ANSYS Mechanical APDL Advanced Analysis

Guide

ANSYS, Inc.
Southpointe
275 Technology Drive
Canonsburg, PA 15317
ansysinfo@ansys.com
http://www.ansys.com
(T) 724-746-3304
(F) 724-514-9494

Release 14.5
October 2012
ANSYS, Inc. is
certified to ISO
9001:2008.

Copyright and Trademark Information

2012 SAS IP, Inc. All rights reserved. Unauthorized use, distribution or duplication is prohibited.
ANSYS, ANSYS Workbench, Ansoft, AUTODYN, EKM, Engineering Knowledge Manager, CFX, FLUENT, HFSS and any
and all ANSYS, Inc. brand, product, service and feature names, logos and slogans are registered trademarks or
trademarks of ANSYS, Inc. or its subsidiaries in the United States or other countries. ICEM CFD is a trademark used
by ANSYS, Inc. under license. CFX is a trademark of Sony Corporation in Japan. All other brand, product, service
and feature names or trademarks are the property of their respective owners.

Disclaimer Notice
THIS ANSYS SOFTWARE PRODUCT AND PROGRAM DOCUMENTATION INCLUDE TRADE SECRETS AND ARE CONFIDENTIAL AND PROPRIETARY PRODUCTS OF ANSYS, INC., ITS SUBSIDIARIES, OR LICENSORS. The software products
and documentation are furnished by ANSYS, Inc., its subsidiaries, or affiliates under a software license agreement
that contains provisions concerning non-disclosure, copying, length and nature of use, compliance with exporting
laws, warranties, disclaimers, limitations of liability, and remedies, and other provisions. The software products
and documentation may be used, disclosed, transferred, or copied only in accordance with the terms and conditions
of that software license agreement.
ANSYS, Inc. is certified to ISO 9001:2008.

U.S. Government Rights

For U.S. Government users, except as specifically granted by the ANSYS, Inc. software license agreement, the use,
duplication, or disclosure by the United States Government is subject to restrictions stated in the ANSYS, Inc.
software license agreement and FAR 12.212 (for non-DOD licenses).

Third-Party Software
See the legal information in the product help files for the complete Legal Notice for ANSYS proprietary software
and third-party software. If you are unable to access the Legal Notice, please contact ANSYS, Inc.
Published in the U.S.A.

Table of Contents
1. Probabilistic Design ................................................................................................................................ 1
1.1. Understanding Probabilistic Design .................................................................................................. 1
1.1.1. Traditional (Deterministic) vs. Probabilistic Design Analysis Methods ......................................... 2
1.1.2. Reliability and Quality Issues .................................................................................................... 3
1.2. Probabilistic Design Terminology ...................................................................................................... 3
1.3. Using Probabilistic Design ................................................................................................................. 8
1.3.1. Create the Analysis File ............................................................................................................. 9
1.3.1.1. Example Problem Description ........................................................................................ 10
1.3.1.2. Build the Model Parametrically ....................................................................................... 10
1.3.1.3. Obtain the Solution ....................................................................................................... 11
1.3.1.4. Retrieve Results and Assign as Output Parameters .......................................................... 11
1.3.1.5. Prepare the Analysis File ................................................................................................ 12
1.3.2. Establish Parameters for Probabilistic Design Analysis ............................................................. 12
1.3.3. Enter the PDS and Specify the Analysis File ............................................................................. 13
1.3.4. Declare Random Input Variables ............................................................................................. 14
1.3.5. Visualize Random Input Variables ........................................................................................... 20
1.3.6. Specify Correlations Between Random Variables ..................................................................... 20
1.3.7. Specify Random Output Parameters ....................................................................................... 23
1.3.8. Select a Probabilistic Design Method ...................................................................................... 24
1.3.8.1. Probabilistic Method Determination Wizard ................................................................... 24
1.3.9. Execute Probabilistic Analysis Simulation Loops ...................................................................... 25
1.3.9.1. Probabilistic Design Looping ......................................................................................... 26
1.3.9.2. Serial Analysis Runs ....................................................................................................... 27
1.3.9.3. PDS Parallel Analysis Runs .............................................................................................. 27
1.3.9.3.1. Machine Configurations ........................................................................................ 29
1.3.9.3.1.1. Choosing Slave Machines ............................................................................. 29
1.3.9.3.1.2. Using the Remote Shell Option ..................................................................... 29
1.3.9.3.1.3. Using the Connection Port Option ................................................................ 31
1.3.9.3.1.4. Configuring the Master Machine ................................................................... 33
1.3.9.3.1.5. Host setup using port option ........................................................................ 35
1.3.9.3.1.6. Host and Product selection for a particular analysis ....................................... 35
1.3.9.3.2. Files Needed for Parallel Run ................................................................................. 36
1.3.9.3.3. Controlling Server Processes ................................................................................. 37
1.3.9.3.4. Initiate Parallel Run ............................................................................................... 38
1.3.10. Fit and Use Response Surfaces .............................................................................................. 38
1.3.10.1. About Response Surface Sets ....................................................................................... 39
1.3.10.2. Fitting a Response Surface ........................................................................................... 39
1.3.10.3. Plotting a Response Surface ......................................................................................... 40
1.3.10.4. Printing a Response Surface ......................................................................................... 40
1.3.10.5. Generating Monte Carlo Simulation Samples on the Response Surfaces ........................ 41
1.3.11. Review Results Data ............................................................................................................. 41
1.3.11.1. Viewing Statistics ......................................................................................................... 42
1.3.11.2. Viewing Trends ............................................................................................................ 43
1.3.11.3. Creating Reports .......................................................................................................... 44
1.4. Guidelines for Selecting Probabilistic Design Variables ..................................................................... 44
1.4.1. Choosing and Defining Random Input Variables ..................................................................... 44
1.4.1.1. Random Input Variables for Monte Carlo Simulations ..................................................... 45
1.4.1.2. Random Input Variables for Response Surface Analyses .................................................. 45
1.4.1.3. Choosing a Distribution for a Random Variable ............................................................... 45
1.4.1.3.1. Measured Data ..................................................................................................... 45
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

iii

Advanced Analysis Guide

1.4.1.3.2. Mean Values, Standard Deviation, Exceedence Values ............................................. 46
1.4.1.3.3. No Data ................................................................................................................ 46
1.4.1.4. Distribution Functions ................................................................................................... 49
1.4.2. Choosing Random Output Parameters .................................................................................... 50
1.5. Probabilistic Design Techniques ...................................................................................................... 51
1.5.1. Monte Carlo Simulations ........................................................................................................ 51
1.5.1.1. Direct Sampling ............................................................................................................. 51
1.5.1.2. Latin Hypercube Sampling ............................................................................................. 52
1.5.1.3. User-Defined Sampling .................................................................................................. 53
1.5.2. Response Surface Analysis Methods ....................................................................................... 55
1.5.2.1. Central Composite Design Sampling .............................................................................. 56
1.5.2.2. Box-Behnken Matrix Sampling ....................................................................................... 58
1.5.2.3. User-Defined Sampling .................................................................................................. 58
1.6. Postprocessing Probabilistic Analysis Results ................................................................................... 59
1.6.1. Statistical Postprocessing ....................................................................................................... 59
1.6.1.1. Sample History .............................................................................................................. 59
1.6.1.2. Histogram ..................................................................................................................... 60
1.6.1.3. Cumulative Distribution Function .................................................................................. 60
1.6.1.4. Print Probabilities .......................................................................................................... 62
1.6.1.5. Print Inverse Probabilities ............................................................................................... 62
1.6.2. Trend Postprocessing ............................................................................................................. 62
1.6.2.1. Sensitivities ................................................................................................................... 62
1.6.2.2. Scatter Plots .................................................................................................................. 65
1.6.2.3. Correlation Matrix .......................................................................................................... 67
1.6.3. Generating an HTML Report ................................................................................................... 67
1.7. Multiple Probabilistic Design Executions ......................................................................................... 67
1.7.1. Saving the Probabilistic Design Database ................................................................................ 68
1.7.2. Restarting a Probabilistic Design Analysis ............................................................................... 68
1.7.3. Clearing the Probabilistic Design Database ............................................................................. 69
1.8. Example Probabilistic Design Analysis ............................................................................................. 69
1.8.1. Problem Description .............................................................................................................. 69
1.8.2. Problem Specifications ........................................................................................................... 69
1.8.2.1. Problem Sketch ............................................................................................................. 70
1.8.3. Using a Batch File for the Analysis ........................................................................................... 70
1.8.4. Using the GUI for the PDS Analysis .......................................................................................... 72
2. Variational Technology ......................................................................................................................... 75
2.1. Harmonic Sweep Using VT Accelerator ............................................................................................ 76
2.1.1. Structural Elements Supporting Frequency-Dependent Properties .......................................... 76
2.1.2. Harmonic Sweep for High-Frequency Electromagnetic Problems ............................................ 77
2.1.2.1. Transmission Line Example Problem ............................................................................... 78
2.1.2.2. Waveguide Example Problem ......................................................................................... 79
2.1.3. Harmonic Sweep for Structural Analysis with Frequency-Dependent Material Properties ......... 81
2.1.3.1. Beam Example ............................................................................................................... 81
3. Adaptive Meshing ................................................................................................................................. 85
3.1. Prerequisites for Adaptive Meshing ................................................................................................. 85
3.2. Employing Adaptive Meshing ......................................................................................................... 85
3.3. Modifying the Adaptive Meshing Process ........................................................................................ 86
3.3.1. Selective Adaptivity ............................................................................................................... 86
3.3.2. Customizing the ADAPT Macro with User Subroutines ............................................................ 87
3.3.2.1. Creating a Custom Meshing Subroutine (ADAPTMSH.MAC) ............................................ 87
3.3.2.2. Creating a Custom Subroutine for Boundary Conditions (ADAPTBC.MAC) ....................... 88
3.3.2.3. Creating a Custom Solution Subroutine (ADAPTSOL.MAC) .............................................. 88

iv

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Advanced Analysis Guide

3.3.2.4. Some Further Comments on Custom Subroutines .......................................................... 88
3.3.3. Customizing the ADAPT Macro (UADAPT.MAC) ....................................................................... 89
3.4. Adaptive Meshing Hints and Comments .......................................................................................... 89
3.5. Where to Find Examples .................................................................................................................. 90
4. Rezoning ............................................................................................................................................... 93
4.1. Benefits and Limitations of Rezoning ............................................................................................... 93
4.1.1. Rezoning Limitations .............................................................................................................. 95
4.2. Rezoning Requirements .................................................................................................................. 96
4.3. Understanding the Rezoning Process .............................................................................................. 98
4.3.1. Overview of the Rezoning Process Flow .................................................................................. 98
4.3.2. Key Commands Used in Rezoning ........................................................................................ 101
4.4. Step 1: Determine the Substep to Initiate Rezoning ........................................................................ 102
4.5. Step 2. Initiate Rezoning ................................................................................................................ 103
4.6. Step 3: Select a Region to Remesh ................................................................................................. 103
4.7. Step 4: Perform the Remeshing Operation ..................................................................................... 104
4.7.1. Choosing a Remeshing Method ............................................................................................ 104
4.7.1.1. Remeshing Using a Program-Generated New Mesh (2-D) .............................................. 105
4.7.1.1.1. Creating an Area to Remesh ................................................................................ 105
4.7.1.1.2. Using Nodes From the Old Mesh ......................................................................... 105
4.7.1.1.3. Hints for Remeshing Multiple Regions ................................................................ 106
4.7.1.1.4. Generating a New Mesh ..................................................................................... 106
4.7.1.2. Remeshing Using a Generic New Mesh (2-D and 3-D) ................................................... 106
4.7.1.2.1. Using the REMESH Command with a Generic New Mesh ...................................... 107
4.7.1.2.2. Requirements for the Generic New Mesh ............................................................. 107
4.7.1.2.3. Using the REGE and KEEP Remeshing Options ..................................................... 108
4.7.1.3. Remeshing Using Manual Mesh Splitting (2-D) ............................................................. 110
4.7.1.3.1. Understanding Mesh Splitting ............................................................................. 110
4.7.1.3.2. Geometry Details for Mesh Splitting ................................................................... 110
4.7.1.3.3. Using the REMESH Command for Mesh Splitting .................................................. 112
4.7.1.3.4. Mesh Transition Options for Mesh Splitting .......................................................... 113
4.7.2. Mesh Control ....................................................................................................................... 114
4.7.3. Remeshing Multiple Regions at the Same Substep ................................................................ 116
4.8. Step 5: Verify Applied Contact Boundaries, Loads, and Boundary Conditions ................................... 116
4.8.1. Contact Boundaries .............................................................................................................. 116
4.8.2. Pressure and Contiguous Displacements ............................................................................... 117
4.8.3. Forces and Isolated Applied Displacements ........................................................................... 117
4.8.4. Nodal Temperatures ............................................................................................................. 117
4.8.5. Other Boundary Conditions and Loads ................................................................................. 118
4.9. Step 6: Automatically Map Variables and Balance Residuals ............................................................ 118
4.9.1. Mapping Solution Variables .................................................................................................. 118
4.9.2. Balancing Residual Forces ..................................................................................................... 118
4.9.3. Interpreting Mapped Results ................................................................................................ 120
4.9.4. Handling Convergence Difficulties ........................................................................................ 120
4.10. Step 7: Perform a Multiframe Restart ............................................................................................ 120
4.11. Repeating the Rezoning Process if Necessary ............................................................................... 121
4.11.1. File Structures for Repeated Rezonings ............................................................................... 121
4.12. Postprocessing Rezoning Results ................................................................................................. 121
4.12.1. The Database Postprocessor ............................................................................................... 122
4.12.1.1. Listing the Rezoning Results File Summary ................................................................. 122
4.12.1.2. Animating the Rezoning Results ................................................................................. 122
4.12.1.3. Using the Results Viewer for Rezoning ........................................................................ 123
4.12.2. The Time-History Postprocessor .......................................................................................... 123
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Advanced Analysis Guide

4.13. Rezoning Restrictions .................................................................................................................. 123
4.14. Rezoning Examples ..................................................................................................................... 124
4.14.1. Example: Rezoning Using a Program-Generated New Mesh ................................................. 124
4.14.1.1. Initial Input for the Analysis ........................................................................................ 125
4.14.1.2. Rezoning Input for the Analysis .................................................................................. 127
4.14.2. Example: Rezoning Using a Generic New Mesh .................................................................... 127
4.14.2.1. Initial Input for the Analysis ........................................................................................ 128
4.14.2.2. Exporting the Distorted Mesh as a CDB File ................................................................ 130
4.14.2.3. Importing the File into ANSYS ICEM CFD and Generating a New Mesh ......................... 131
4.14.2.4. Rezoning Using the New CDB Mesh ........................................................................... 132
5. 2-D to 3-D Analysis .............................................................................................................................. 135
5.1. Benefits of 2-D to 3-D Analysis ....................................................................................................... 135
5.2. Requirements for a 2-D to 3-D Analysis ......................................................................................... 135
5.3. Overview of the 2-D to 3-D Analysis Process .................................................................................. 137
5.3.1. Overview of the 2-D to 3-D Analysis Process Flow .................................................................. 137
5.3.2. Key Commands Used in 2-D to 3-D Analysis .......................................................................... 138
5.4. Performing a 2-D to 3-D Analysis ................................................................................................... 139
5.4.1. Step 1: Determine the Substep to Initiate .............................................................................. 139
5.4.2. Step 2: Initiate the 2-D to 3-D Analysis ................................................................................... 140
5.4.3. Step 3: Extrude the 2-D Mesh to the New 3-D Mesh .............................................................. 140
5.4.4. Step 4: Map Solution Variables from 2-D to 3-D Mesh ............................................................. 141
5.4.5. Step 5: Perform an Initial-State-Based 3-D Analysis ................................................................ 141
5.5. 2-D to 3-D Analysis Restrictions ..................................................................................................... 141
6. Cyclic Symmetry Analysis .................................................................................................................... 143
6.1. Understanding Cyclic Symmetry Analysis ...................................................................................... 143
6.1.1. How the Program Automates a Cyclic Symmetry Analysis ...................................................... 144
6.1.2. Commands Used in a Cyclic Symmetry Analysis ..................................................................... 144
6.2. Cyclic Modeling ............................................................................................................................ 144
6.2.1. The Basic Sector ................................................................................................................... 145
6.2.2. Edge Component Pairs ......................................................................................................... 146
6.2.2.1. CYCOPT Auto Detection Tolerance Adjustments for Difficult Cases ............................... 146
6.2.2.2. Identical vs. Dissimilar Edge Node Patterns ................................................................... 149
6.2.2.3. Unmatched Nodes on Edge-Component Pairs .............................................................. 150
6.2.2.4. Identifying Matching Node Pairs .................................................................................. 150
6.2.3. Model Verification (Preprocessing) ........................................................................................ 150
6.3. Solving a Cyclic Symmetry Analysis ............................................................................................... 150
6.3.1. Understanding the Solution Architecture .............................................................................. 151
6.3.1.1. The Duplicate Sector .................................................................................................... 151
6.3.1.2. Coupling and Constraint Equations (CEs) ...................................................................... 151
6.3.1.3. Non-Cyclically Symmetric Loading ............................................................................... 152
6.3.1.3.1. Specifying Non-Cyclic Loading ............................................................................ 153
6.3.1.3.2. Commands Affected by Non-Cyclic Loading ........................................................ 155
6.3.1.3.3. Plotting and Listing Non-Cyclic Boundary Conditions ........................................... 155
6.3.1.3.4. Graphically Picking Non-Cyclic Boundary Conditions ........................................... 155
6.3.2. Solving a Static Cyclic Symmetry Analysis .............................................................................. 155
6.3.3. Solving a Modal Cyclic Symmetry Analysis ............................................................................ 157
6.3.3.1. Understanding Harmonic Index and Nodal Diameter .................................................... 157
6.3.3.2. Solving a Stress-Free Modal Analysis ............................................................................ 158
6.3.3.3. Solving a Prestressed Modal Analysis ............................................................................ 159
6.3.3.4. Solving a Large-Deflection Prestressed Modal Analysis ................................................. 160
6.3.3.4.1. Solving a Large-Deflection Prestressed Modal Analysis with VT Accelerator .......... 162
6.3.4. Solving a Linear Buckling Cyclic Symmetry Analysis ............................................................... 162

vi

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Advanced Analysis Guide

6.3.5. Solving a Harmonic Cyclic Symmetry Analysis ....................................................................... 163
6.3.6. Solving a Magnetic Cyclic Symmetry Analysis ........................................................................ 163
6.3.7. Database Considerations After Obtaining the Solution .......................................................... 165
6.3.8. Model Verification (Solution) ................................................................................................. 165
6.3.8.1. Solving a Prestressed Harmonic Cyclic Symmetry Analysis ............................................ 165
6.4. Postprocessing a Cyclic Symmetry Analysis .................................................................................... 166
6.4.1. Real and Imaginary Solution Components ............................................................................ 167
6.4.2. Expanding the Cyclic Symmetry Solution .............................................................................. 167
6.4.2.1. Using the /CYCEXPAND Command ............................................................................... 167
6.4.2.1.1. Applying a Traveling Wave Animation to the Cyclic Model .................................... 169
6.4.2.2. Using the EXPAND Command ...................................................................................... 169
6.4.3. Phase Sweep of Repeated Eigenvector Shapes ...................................................................... 169
6.5. Limitations in a Cyclic Symmetry Analysis ...................................................................................... 171
6.5.1. Solver Limitations ................................................................................................................. 171
6.5.2. /CYCEXPAND Limitations ..................................................................................................... 172
6.6. Example Modal Cyclic Symmetry Analysis ...................................................................................... 172
6.6.1. Problem Description ............................................................................................................. 172
6.6.2. Problem Specifications ......................................................................................................... 173
6.6.3. Input File for the Analysis ...................................................................................................... 173
6.6.4. Analysis Steps ...................................................................................................................... 175
6.7. Example Buckling Cyclic Symmetry Analysis .................................................................................. 177
6.7.1. Problem Description ............................................................................................................. 177
6.7.2. Problem Specifications ......................................................................................................... 177
6.7.3. Input File for the Analysis ...................................................................................................... 178
6.7.4. Analysis Steps ...................................................................................................................... 181
6.7.5. Solve For Critical Strut Temperature at Load Factor = 1.0 ........................................................ 182
6.8. Example Harmonic Cyclic Symmetry Analysis ................................................................................. 184
6.8.1. Problem Description ............................................................................................................. 185
6.8.2. Problem Specifications ......................................................................................................... 185
6.8.3. Input File for the Analysis ...................................................................................................... 185
6.8.4. Analysis Steps ...................................................................................................................... 187
6.9. Example Magnetic Cyclic Symmetry Analysis ................................................................................. 190
6.9.1. Problem Description ............................................................................................................. 190
6.9.2. Problem Specifications ......................................................................................................... 191
6.9.3. Input file for the Analysis ...................................................................................................... 192
7. Rotating Structure Analysis ................................................................................................................ 197
7.1. Understanding Rotating Structure Dynamics ................................................................................. 197
7.2. Using a Stationary Reference Frame ............................................................................................... 198
7.2.1. Campbell Diagram ............................................................................................................... 199
7.2.2. Harmonic Analysis for Unbalance or General Rotating Asynchronous Forces .......................... 201
7.2.3. Orbits ................................................................................................................................... 202
7.3. Using a Rotating Reference Frame ................................................................................................. 202
7.4. Choosing the Appropriate Reference Frame Option ....................................................................... 204
7.5. Example Campbell Diagram Analysis ............................................................................................. 205
7.5.1. Problem Description ............................................................................................................. 205
7.5.2. Problem Specifications ......................................................................................................... 205
7.5.3. Input for the Analysis ............................................................................................................ 205
7.5.4. Analysis Steps ...................................................................................................................... 206
7.6. Example Coriolis Analysis .............................................................................................................. 207
7.6.1. Problem Description ............................................................................................................. 208
7.6.2. Problem Specifications ......................................................................................................... 208
7.6.3. Input for the Analysis ............................................................................................................ 208
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

vii

Advanced Analysis Guide

7.6.4. Analysis Steps ...................................................................................................................... 209
7.7. Example Unbalance Harmonic Analysis ......................................................................................... 210
7.7.1. Problem Description ............................................................................................................. 210
7.7.2. Problem Specifications ......................................................................................................... 210
7.7.3. Input for the Analysis ............................................................................................................ 211
7.7.4. Analysis Steps ...................................................................................................................... 213
8. Submodeling ....................................................................................................................................... 219
8.1. Understanding Submodeling ........................................................................................................ 219
8.1.1. Nonlinear Submodeling ....................................................................................................... 220
8.2. Using Submodeling ...................................................................................................................... 220
8.2.1. Create and Analyze the Coarse Model ................................................................................... 221
8.2.2. Create the Submodel ............................................................................................................ 222
8.2.3. Perform Cut-Boundary Interpolation ..................................................................................... 223
8.2.4. Analyze the Submodel ......................................................................................................... 224
8.2.5. Verify the Distance Between the Cut Boundaries and the Stress Concentration ...................... 226
8.3. Example Submodeling Analysis Input ............................................................................................ 227
8.3.1. Submodeling Analysis Input: No Load-History Dependency ................................................... 227
8.3.2. Submodeling Analysis Input: Load-History Dependency ........................................................ 228
8.4. Shell-to-Solid Submodels .............................................................................................................. 233
8.5. Where to Find Examples ................................................................................................................ 235
9. Substructuring .................................................................................................................................... 237
9.1. Benefits of Substructuring ............................................................................................................. 237
9.2. Using Substructuring .................................................................................................................... 237
9.2.1. Step 1: Generation Pass (Creating the Superelement) ............................................................ 238
9.2.1.1. Building the Model ...................................................................................................... 239
9.2.1.2. Applying Loads and Creating the Superelement Matrices ............................................. 240
9.2.2. Step 2: Use Pass (Using the Superelement) ............................................................................ 244
9.2.2.1. Clear the Database and Specify a New Jobname ........................................................... 244
9.2.2.2. Build the Model ........................................................................................................... 245
9.2.2.3. Apply Loads and Obtain the Solution ........................................................................... 247
9.2.3. Step 3: Expansion Pass (Expanding Results Within the Superelement) .................................... 249
9.3. Sample Analysis Input ................................................................................................................... 252
9.4. Top-Down Substructuring ............................................................................................................. 253
9.5. Automatically Generating Superelements ...................................................................................... 255
9.6. Nested Superelements .................................................................................................................. 256
9.7. Prestressed Substructures ............................................................................................................. 256
9.7.1. Static Analysis Prestress ........................................................................................................ 257
9.7.2. Substructuring Analysis Prestress .......................................................................................... 257
9.8. Where to Find Examples ................................................................................................................ 257
10. Component Mode Synthesis ............................................................................................................. 259
10.1. Understanding Component Mode Synthesis ................................................................................ 259
10.1.1. CMS Methods Supported .................................................................................................... 259
10.1.2. Solvers Used in Component Mode Synthesis ....................................................................... 260
10.2. Using Component Mode Synthesis .............................................................................................. 261
10.2.1. The CMS Generation Pass: Creating the Superelement ......................................................... 261
10.2.2. The CMS Use and Expansion Passes ..................................................................................... 263
10.2.3. Superelement Expansion in Transformed Locations ............................................................. 263
10.2.4. Plotting or Printing Mode Shapes ....................................................................................... 264
10.3. Example Component Mode Synthesis Analysis ............................................................................ 264
10.3.1. Problem Description ........................................................................................................... 264
10.3.2. Problem Specifications ....................................................................................................... 264
10.3.3. Input for the Analysis: Fixed-Interface Method ..................................................................... 267

viii

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Advanced Analysis Guide

10.3.4. Analysis Steps: Fixed-Interface Method ................................................................................ 271
10.3.5. Input for the Analysis: Free-Interface Method ...................................................................... 274
10.3.6. Analysis Steps: Free-Interface Method ................................................................................. 275
10.3.7. Input for the Analysis: Residual-Flexible Free-Interface Method ............................................ 276
10.3.8. Analysis Steps: Residual-Flexible Free-Interface Method ....................................................... 278
10.3.9. Example: Superelement Expansion in a Transformed Location ............................................. 279
10.3.9.1. Analysis Steps: Superelement Expansion in a Transformed Location ............................ 281
11. Rigid-Body Dynamics and the ANSYS-ADAMS Interface .................................................................. 283
11.1. Understanding the ANSYS-ADAMS Interface ................................................................................ 283
11.2. Building the Model ...................................................................................................................... 284
11.3. Modeling Interface Points ........................................................................................................... 285
11.4. Exporting to ADAMS ................................................................................................................... 286
11.4.1. Exporting to ADAMS via Batch Mode .................................................................................. 288
11.4.2. Verifying the Results ........................................................................................................... 288
11.5. Running the ADAMS Simulation .................................................................................................. 289
11.6. Transferring Loads from ADAMS .................................................................................................. 289
11.6.1. Transferring Loads on a Rigid Body ..................................................................................... 289
11.6.1.1. Exporting Loads in ADAMS ........................................................................................ 290
11.6.1.2. Importing Loads ........................................................................................................ 291
11.6.1.3. Importing Loads via Commands ................................................................................. 292
11.6.1.4. Reviewing the Results ................................................................................................ 292
11.6.2. Transferring the Loads of a Flexible Body ............................................................................. 292
11.7. Methodology Behind the ANSYS-ADAMS Interface ...................................................................... 293
11.7.1. The Modal Neutral File ........................................................................................................ 293
11.7.2. Adding Weak Springs ......................................................................................................... 294
11.8. Example Rigid-Body Dynamic Analysis ........................................................................................ 294
11.8.1. Problem Description ........................................................................................................... 294
11.8.2. Problem Specifications ....................................................................................................... 295
11.8.3. Command Input ................................................................................................................. 296
12. Element Birth and Death ................................................................................................................... 299
12.1. Elements Supporting Birth and Death ......................................................................................... 299
12.2. Understanding Element Birth and Death ..................................................................................... 299
12.3. Element Birth and Death Usage Hints .......................................................................................... 300
12.3.1. Changing Material Properties ............................................................................................. 301
12.4. Using Birth and Death ................................................................................................................. 301
12.4.1. Build the Model .................................................................................................................. 301
12.4.2. Apply Loads and Obtain the Solution .................................................................................. 301
12.4.2.1. Define the First Load Step .......................................................................................... 301
12.4.2.1.1. Sample Input for First Load Step ........................................................................ 302
12.4.2.2. Define Subsequent Load Steps ................................................................................... 302
12.4.2.2.1. Sample Input for Subsequent Load Steps ........................................................... 302
12.4.3. Review the Results .............................................................................................................. 302
12.4.4. Use Analysis Results to Control Birth and Death ................................................................... 303
12.4.4.1. Sample Input for Deactivating Elements ..................................................................... 303
12.5. Where to Find Examples .............................................................................................................. 303
13. User-Programmable Features and Nonstandard Uses ..................................................................... 305
13.1. User-Programmable Features (UPFs) ............................................................................................ 305
13.1.1. Understanding UPFs ........................................................................................................... 305
13.1.2.Types of UPFs Available ....................................................................................................... 306
13.2. Nonstandard Uses of the ANSYS Program .................................................................................... 307
13.2.1. What Are Nonstandard Uses? .............................................................................................. 308
13.2.2. Hints for Nonstandard Use of ANSYS ................................................................................... 308
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

ix

Advanced Analysis Guide

14. State-Space Matrices Export ............................................................................................................. 309
14.1. State-Space Matrices Based on Modal Analysis ............................................................................. 309
14.1.1. Examples of SPMWRITE Command Usage .......................................................................... 309
14.1.2. Example of Reduced Model Generation in ANSYS and Usage in Simplorer ........................... 310
14.1.2.1. Problem Description .................................................................................................. 310
14.1.2.2. Problem Specifications ............................................................................................... 311
14.1.2.3. Input File for the Analysis ........................................................................................... 311
15. Soil-Pile-Structure Analysis ............................................................................................................... 315
15.1. Soil-Pile-Structure Interaction Analysis ......................................................................................... 315
15.1.1. Automatic Pile Subdivision ................................................................................................. 316
15.1.2. Convergence Criteria .......................................................................................................... 316
15.1.3. Soil Representation ............................................................................................................ 317
15.1.4. Mudslides .......................................................................................................................... 322
15.1.5. Soil-Pile Interaction Results ................................................................................................. 323
15.1.5.1. Displacements and Reactions ..................................................................................... 323
15.1.5.2. Forces and Stresses .................................................................................................... 323
15.1.5.3. UNITY Check Data ...................................................................................................... 323
15.2. Soil Data Definition and Examples ............................................................................................... 325
15.2.1. Soil Profile Data Definition .................................................................................................. 325
15.2.1.1. Mudline Position Definition ........................................................................................ 327
15.2.1.2. Common Factors for P-Y, T-Z Curves ............................................................................ 327
15.2.1.3. Horizontal Soil Properties (P-Y) ................................................................................... 328
15.2.1.3.1. P-Y curves defined explicitly .............................................................................. 329
15.2.1.3.2. P-Y curves generated from given soil properties ................................................ 329
15.2.1.4. Vertical Soil Properties (T-Z) ....................................................................................... 331
15.2.1.4.1. T-Z curves defined explicitly .............................................................................. 331
15.2.1.4.2. T-Z curves generated from given soil properties ................................................. 332
15.2.1.5. End Bearing Properties (ENDB) ................................................................................... 333
15.2.1.5.1. ENDB curve defined explicitly ............................................................................ 333
15.2.1.5.2. ENDB curves generated from given soil properties ............................................. 334
15.2.1.6. Mudslide Definition ................................................................................................... 335
15.2.2. Soil Data File Examples ....................................................................................................... 335
15.2.2.1. Example 1: Constant Linear Soil .................................................................................. 336
15.2.2.2. Example 2: Non-Linear Soil ......................................................................................... 336
15.2.2.3. Example 3: Soil Properties Defined in 5 Layers ............................................................. 337
15.2.2.4. Example 4: Soil Properties Defined in 5 Layers with Mudslide ...................................... 337
15.3. Performing a Soil-Pile Interaction Analysis ................................................................................... 338
15.3.1. Overview of Mechanical APDL Commands Used for Soil-Pile Interaction Analysis ................. 338
15.3.2. Mechanical APDL Component System Example .................................................................. 338
15.3.3. Static Structural Component System Example ..................................................................... 339
15.4. Soil-Pile-Structure Results ............................................................................................................ 341
15.5. References .................................................................................................................................. 342
16. Coupling to External Aeroelastic Analysis of Wind Turbines ............................................................ 343
16.1. Sequential Coupled Wind Turbine Solution in Mechanical APDL ................................................... 343
16.1.1. Procedure for a Sequentially Coupled Wind Turbine Analysis ............................................... 343
16.1.2. Output from the OUTAERO Command ................................................................................ 344
16.1.3. Example Substructured Analysis to Write Out Aeroelastic Analysis Input Data ...................... 345
17. Applying Ocean Loading from a Hydrodynamic Analysis ................................................................ 349
17.1. How Hydrodynamic Analysis Data Is Used .................................................................................... 349
17.2. Hydrodynamic Load Transfer with Forward Speed ........................................................................ 350
17.3. Hydrodynamic Data File Format .................................................................................................. 350
17.3.1. Comment (Optional) .......................................................................................................... 350

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Advanced Analysis Guide

17.3.2. General Model Data ............................................................................................................ 350
17.3.3. Hydrodynamic Surface Geometry ....................................................................................... 351
17.3.4. Wave Periods ...................................................................................................................... 352
17.3.5. Wave Directions .................................................................................................................. 352
17.3.6. Panel Pressures ................................................................................................................... 352
17.3.7. Morison Element Hydrodynamic Definition ......................................................................... 353
17.3.8. Morison Element Wave Kinematics Definition ...................................................................... 354
17.3.9. RAO Definition ................................................................................................................... 355
17.3.10. Mass Properties ................................................................................................................ 355
17.4. Example Analysis Using Results from a Hydrodynamic Diffraction Analysis ................................... 356
Index ........................................................................................................................................................ 361

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

xi

xii

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

List of Figures
1.1. A Beam Under a Snow Load .................................................................................................................... 4
1.2. Probabilistic Design Data Flow ................................................................................................................ 8
1.3. A Beam Under a Snow Load .................................................................................................................. 10
1.4. Histograms for the Snow Height H1 and H2 ........................................................................................... 19
1.5. A Scatter Plot of Snow Height H1 vs. H2 ................................................................................................. 21
1.6. The PDS Method Determination Wizard ................................................................................................. 25
1.7. Graph of X1 and X2 Showing Two Samples with Close Values ................................................................. 52
1.8. Graph of X1 and X2 Showing Good Sample Distribution ........................................................................ 53
1.9. Locations of Sampling Points for Problem with Three Input Variables for CCD ......................................... 57
1.10. Location of Sampling Points for Problem with Three Input Variables for BBM ........................................ 58
1.11. Cumulative Distribution Function of X ................................................................................................. 61
1.12. Sensitivities ........................................................................................................................................ 63
1.13. Range of Scatter .................................................................................................................................. 63
1.14. Effects of Reducing and Shifting Range of Scatter ................................................................................ 66
1.15. The Simple Indeterminate Three-Bar Truss for the Example Problem ..................................................... 70
2.1. Element Plot for Waveguide Example .................................................................................................... 79
2.2. Graph of Phase Angle ............................................................................................................................ 80
2.3. Graph of Magnitude ............................................................................................................................. 81
3.1. Selective Adaptivity .............................................................................................................................. 87
4.1. Rezoning Using a Program-Generated New Mesh .................................................................................. 99
4.2. Rezoning Using a Generic New Mesh Generated by Another Application ............................................... 99
4.3. Rezoning Using Manual Splitting of an Existing Mesh .......................................................................... 100
4.4. Boundary Geometry of a Generic (CDB) New Mesh .............................................................................. 108
4.5. Remeshing Options when Using a Generic (CDB) New Mesh ................................................................ 108
4.6. Splitting of Quadrilateral and Degenerate Linear Elements (PLANE182) ................................................ 111
4.7. Splitting of Quadrilateral, Degenerate and Triangular Quadratic Elements (PLANE183) .......................... 112
4.8. Transition Element Generation Methods ............................................................................................. 113
4.9. /PREP7 Mesh-Control Commands Available in Rezoning ...................................................................... 115
5.1. 2-D to 3-D Process Flow ....................................................................................................................... 138
5.2. 2-D Plane Strain to 3-D Solid Extrusion ................................................................................................. 140
5.3. Axisymmetric to 3-D Solid Extrusion .................................................................................................... 140
6.1. Hydro Rotor -- Model of a Cyclically Symmetric Structure ..................................................................... 143
6.2. A Basic Sector in a Cyclically Symmetric Structure ................................................................................ 145
6.3. Basic Sector Definition ........................................................................................................................ 145
6.4. Full Cyclic Model ................................................................................................................................. 147
6.5. Cyclic Sector ....................................................................................................................................... 147
6.6. Successful Auto Detection with Default FACETOL = 15 Deg ................................................................ 148
6.7. Auto Detection Failure Due to Large Face Tolerance ............................................................................. 149
6.8. Connecting Low and High Edges of Basic and Duplicate Sectors .......................................................... 152
6.9. Process Flow for a Static Cyclic Symmetry Analysis (Cyclic Loading) ...................................................... 156
6.10. Process Flow for a Static Cyclic Symmetry Analysis (Non-Cyclic Loading) ............................................. 156
6.11. Examples of Nodal Diameters (i) ........................................................................................................ 157
6.12. Process Flow for a Stress-Free Modal Cyclic Symmetry Analysis ........................................................... 159
6.13. Process Flow for a Prestressed Modal Cyclic Symmetry Analysis .......................................................... 160
6.14. Process Flow for a Large-Deflection Prestressed Modal Cyclic Symmetry Analysis ............................... 161
6.15. Process Flow for a Linear Buckling Cyclic Symmetry Analysis .............................................................. 162
6.16. Process Flow for a Harmonic Cyclic Symmetry Analysis (Non-Cyclic Loading) ...................................... 163
6.17. Process Flow for a Prestressed Harmonic Cyclic Symmetry Analysis .................................................... 166
6.18. Traveling Wave Animation Example ................................................................................................... 169
6.19. Example Modal Cyclic Symmetry Analysis Results .............................................................................. 177
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

xiii

Advanced Analysis Guide

6.20. Example Buckling Cyclic Symmetry Analysis Results ........................................................................... 182
6.21. Buckling Cyclic Symmetry Results: Load Factor Iterations .................................................................... 182
6.22. Buckling Cyclic Symmetry Results: Load Factor Results Graph ............................................................ 184
6.23. Element Plot Showing Pressure Load on Sector 3 ............................................................................... 188
6.24. Contour Plot of Displacement Sum at Frequency of 866 HZ ................................................................ 189
6.25. Displacement Plot as a Function of Excitation Frequency ................................................................... 190
6.26. Two-Phase Electric Machine Full Model ........................................................................................... 191
6.27. Two-Phase Electric Machine Half Model .......................................................................................... 191
6.28. Vector Plot of Cyclic Flux Density (B) - Half Model ............................................................................... 196
6.29. Contour Line Plot of Equipotentials ................................................................................................... 196
8.1. Submodeling of a Pulley ..................................................................................................................... 219
8.2. Coarse Model ...................................................................................................................................... 221
8.3. Submodel Superimposed Over Coarse Model ...................................................................................... 222
8.4. Cut Boundaries on the Submodel ........................................................................................................ 223
8.5. Loads on the Submodel ...................................................................................................................... 226
8.6. Data Flow Diagram for Submodeling (Without Temperature Interpolation) .......................................... 226
8.7. Contour Plots to Compare Results ....................................................................................................... 227
8.8. Path Plots to Compare Results ............................................................................................................. 227
8.9. Coarse-Mesh Model, Submodel, and Fine-Mesh Model ......................................................................... 230
8.10. Equivalent Plastic Strain Through Various Cut-Boundary Conditions and Load Steps ........................... 231
8.11. Equivalent Plastic Strain Distributions in a Submodeling Analysis with Load-History Dependency ....... 232
8.12. 3-D Solid Submodel Superimposed on Coarse Shell Model ................................................................. 234
8.13. Node Rotations ................................................................................................................................. 235
9.1. Applicable Solvers in a Typical Substructuring Analysis ........................................................................ 238
9.2. Example of a Substructuring Application ............................................................................................. 238
9.3. Node Locations ................................................................................................................................... 245
10.1. Applicable CMS Solvers and Files ....................................................................................................... 260
10.2. Process Flow for Creating a CMS Superelement Matrix ....................................................................... 262
10.3. Example CMS Analysis Results: Fixed-Interface Method ...................................................................... 274
11.1. Connecting a Structure to an Interface Point ...................................................................................... 286
11.2. Export to ADAMS Dialog Box ............................................................................................................. 287
11.3. ADAMS Export FEA Loads Dialog Box ................................................................................................ 290
11.4. Import from ADAMS Dialog Box ........................................................................................................ 291
11.5. Linkage Assembly ............................................................................................................................. 295
11.6. Link3 Component ............................................................................................................................. 295
14.1. Simulation Schematic ........................................................................................................................ 313
14.2. Evolution of Spring Force .................................................................................................................. 313
15.1. Pile/Spring Foundation Model ........................................................................................................... 317
15.2. Soil/load Deflection Characteristics ................................................................................................... 318
15.3. P-Y Curve for Sand ............................................................................................................................ 320
15.4. P-Y Curve for Clay: Static Loading ....................................................................................................... 321
15.5. P-Y Curve for Clay: Cyclic Loading ...................................................................................................... 321
15.6. T-Z Curve for Clay and Sand ............................................................................................................... 322
15.7. ENDB Curve for Clay and Sand ........................................................................................................... 322
17.1. Hydrodynamic Diffraction Analysis Pressure Results ........................................................................... 357
17.2. Analysis Results Using Pressure Data from a Hydrodynamic Diffraction Analysis ................................. 359

xiv

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

List of Tables
6.1. Valid Non-Cyclically Symmetric Loads .................................................................................................. 153
6.2. Buckling Cyclic Symmetry: Load Factor Iteration Results ....................................................................... 184
9.1. Loads Applicable in a Substructure Analysis ......................................................................................... 243

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

xv

xvi

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 1: Probabilistic Design

Probabilistic design is an analysis technique for assessing the effect of uncertain input parameters and
assumptions on your model.
A probabilistic analysis allows you to determine the extent to which uncertainties in the model affect
the results of a finite element analysis. An uncertainty (or random quantity) is a parameter whose value
is impossible to determine at a given point in time (if it is time-dependent) or at a given location (if it
is location-dependent). An example is ambient temperature; you cannot know precisely what the temperature will be one week from now in a given city.
In a probabilistic analysis, statistical distribution functions (such as the Gaussian or normal distribution,
the uniform distribution, etc.) describe uncertain parameters. (See the Modeling and Meshing Guide for
a description of parameters.)
The Probabilistic Design System (PDS) allows you to perform a probabilistic design analysis. The following
topics are available:
1.1. Understanding Probabilistic Design
1.2. Probabilistic Design Terminology
1.3. Using Probabilistic Design
1.4. Guidelines for Selecting Probabilistic Design Variables
1.5. Probabilistic Design Techniques
1.6. Postprocessing Probabilistic Analysis Results
1.7. Multiple Probabilistic Design Executions
1.8. Example Probabilistic Design Analysis

1.1. Understanding Probabilistic Design

Computer models are expressed and described with specific numerical and deterministic values; material properties are entered using certain values, the geometry of the component is assigned a certain
length or width, etc. An analysis based on a given set of specific numbers and values is called a deterministic analysis. Naturally, the results of a deterministic analysis are only as good as the assumptions
and input values used for the analysis. The validity of those results depend on how correct the values
were for the component under real life conditions.
In reality, every aspect of an analysis model is subjected to scatter (in other words, is uncertain in some
way). Material property values are different if one specimen is compared to the next. This kind of scatter
is inherent for materials and varies among different material types and material properties. For example,
the scatter of the Young's modulus for many materials can often be described as a Gaussian distribution
with standard deviation of 3 - 5%. Likewise, the geometric properties of components can only be reproduced within certain manufacturing tolerances. The same variation holds true for the loads that are
applied to a finite element model. However, in this case the uncertainty is often due to a lack of engineering knowledge. For example, at elevated temperatures the heat transfer coefficients are very important
in a thermal analysis, yet it is almost impossible to measure the heat transfer coefficients. This means
that almost all input parameters used in a finite element analysis are inexact, each associated with some
degree of uncertainty.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design
It is neither physically possible nor financially feasible to eliminate the scatter of input parameters
completely. The reduction of scatter is typically associated with higher costs either through better and
more precise manufacturing methods and processes or increased efforts in quality control; hence, accepting the existence of scatter and dealing with it rather than trying to eliminate it makes products
more affordable and production of those products more cost-effective.
To deal with uncertainties and scatter, you can use the Probabilistic Design System (PDS) to answer the
following questions:
If the input variables of a finite element model are subjected to scatter, how large is the scatter of the
output parameters? How robust are the output parameters? Here, output parameters can be any parameter
that the program can calculate. Examples are the temperature, stress, strain, or deflection at a node, the
maximum temperature, stress, strain, or deflection of the model, etc.
If the output is subjected to scatter due to the variation of the input variables, then what is the probability
that a design criterion given for the output parameters is no longer met? How large is the probability that
an unexpected and unwanted event takes place (what is the failure probability)?
Which input variables contribute the most to the scatter of an output parameter and to the failure probability? What are the sensitivities of the output parameter with respect to the input variables?
Probabilistic design can be used to determine the effect of one or more variables on the outcome of
the analysis. In addition to the probabilistic design techniques available, the program offers a set of
strategic tools that can be used to enhance the efficiency of the probabilistic design process. For example,
you can graph the effects of one input variable versus an output parameter, and you can easily add
more samples and additional analysis loops to refine your analysis.

1.1.1. Traditional (Deterministic) vs. Probabilistic Design Analysis Methods

In traditional deterministic analyses, uncertainties are either ignored or accounted for by applying
conservative assumptions.
Uncertainties are typically ignored if the analyst knows for certain that the input parameter has no effect
on the behavior of the component under investigation. In this case, only the mean values or some
nominal values are used in the analysis. However, in some situations the influence of uncertainties exists
but is still neglected; for example, the Young's modulus mentioned above or the thermal expansion
coefficient, for which the scatter is usually ignored. Let's assume you are performing a thermal analysis
and you want to evaluate the thermal stresses (thermal stresses are directly proportional to the Young's
modulus as well as to the thermal expansion coefficient of the material). The equation is:
therm = E T
If the Young's modulus alone has a Gaussian distribution with a 5% standard deviation, then there is
almost a 16% chance that the stresses are more than 5% higher than what you would think they are
in a deterministic case. This figure increases if you also take into account that, typically, the thermal
expansion coefficient also follows a Gaussian distribution.
Random Input Variables taken Probability that the thermal
into account
stresses are more than 5%
higher than expected

Probability that the

thermal stresses are more
than 10% higher than expected

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design Terminology

Young's modulus (Gaussian distribution with 5% standard deviation)

~16%

~2.3%

Young's modulus and thermal

expansion coefficient (each with
Gaussian distribution with 5%
standard deviation)

~22%

~8%

When a conservative assumption is used, this actually tells you that uncertainty or randomness is involved.
Conservative assumptions are usually expressed in terms of safety factors. Sometimes regulatory bodies
demand safety factors in certain procedural codes. If you are not faced with such restrictions or demands,
then using conservative assumptions and safety factors can lead to inefficient and costly over-design.
You can avoid over-design by using probabilistic methods while still ensuring the safety of the component.
Probabilistic methods even enable you to quantify the safety of the component by providing a probability that the component will survive operating conditions. Quantifying a goal is the necessary first step
toward achieving it. Probabilistic methods can tell you how to achieve your goal.

1.1.2. Reliability and Quality Issues

Use probabilistic design when issues of reliability and quality are paramount.
Reliability is usually a concern because product or component failures have significant financial consequences (costs of repair, replacement, warranty, or penalties); worse, a failure can result in injury or
loss of life. Although perfection is neither physically possible nor financially feasible, probabilistic design
helps you to design safe and reliable products while avoiding costly over-design and conserve manufacturing resources (machining accuracy, efforts in quality control, and so on).
Quality is the perception by a customer that the product performs as expected or better. In a quality
product, the customer rarely receives unexpected and unpleasant events where the product or one of
its components fails to perform as expected. By nature, those rare "failure" events are driven by uncertainties in the design. Here, probabilistic design methods help you to assess how often "failure" events
may happen. In turn, you can improve the design for those cases where the "failure" event rate is above
your customers' tolerance limit.

1.2. Probabilistic Design Terminology

To understand the terminology involved in probabilistic design, consider the following problem.
Find the scatter of the maximum deflection of a beam under a random load of snow. The snow should
have a linear distribution along the length of the beam with a height of H1 at one end and a height of
H2 at the other. The beam material is described by various parameters including the Young's modulus,
for which a mean value and standard deviation have been derived.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design
Figure 1.1: A Beam Under a Snow Load

H2

H1

E
PDS Term
random input variables (RVs)

Description
Quantities that influence the result of an analysis.
In probabilistic design, RVs are often called "drivers" because they
drive the result of an analysis. You must specify the type of statistical
distribution the RVs follow and the parameter values of their distribution functions.
For the beam example, the heights H1 and H2 and the Young's modulus
E are clearly the random input variables. Naturally, the heights H1 and
H2 cannot be negative and more often there will be little snow and
only a few times there will be a lot of snow. Therefore, it might be
appropriate to model the height of the snow as an exponential or a
lognormal distribution, both of which have the bulk of the data at
lower values.

correlation

Two (or more) RVs which are statistically dependent upon each other.
In the beam example, it is unlikely that one side of the roof (beam)
supports a great deal of snow while almost no snow exists on the
other. It is not necessary that H1 and H2 are exactly identical, but with
a lot of snow then H1 and H2 both likely have larger values and with
little snowfall then both would be lower. Therefore, H1 and H2 are
correlated, although the correlation must not be mistaken for a direct
mathematical dependency. In the beam example, no numerical dependency exists but rather a certain trend between the two heights;
with this particular H1 and H2 it is unlikely that their values will be
drastically different at any given point in time.

Note
Mathematical dependencies have some numerical dependence on each other. For example, a true correlation exists
if, when one parameter value doubles, another parameter
value also doubles.
random
output
parameters
(RPs)

The results of a finite element analysis.

The RPs are typically a function of the RVs; that is, changing the values
of the random input variables should change the value of the random

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design Terminology

PDS Term

Description
output parameters. In our beam example, the maximum beam deflection is a random output parameter.

probabilistic
design variables

The random input variables (RVs) and random output parameters (RPs)
are collectively known as probabilistic design variables.
In the Probabilistic Design System (PDS), you must identify which
parameters in the model are RVs and which are RPs.

sample

A unique set of parameter values that represents a particular model

configuration.
A sample is characterized by random input variable values. If you
measure the Young's modulus of a given beam, and the snow height
on a given day at each end of that beam, then the three values for E,
H1, and H2 would constitute one sample.
Think of a sample as one virtual prototype. Every component manufactured represents one sample, because you can measure its particular
properties (material, geometry, etc.) and obtain specific values for
each.
In statistics, however, sample also has a wider and more general use.
For example, any single measured value of any physical property is
considered to be one sample. Because a probabilistic analysis is based
on a statistical evaluation of the random output parameters (RPs), the
values of the RPs are also called samples.

simulation

The collection of all samples that are required or that you request for
a given probabilistic analysis.
The simulation contains the information used to determine how the
component would behave under real-life conditions (with all the existing uncertainties and scatter); therefore, all samples represent the
simulation of the behavior.

analysis file

An input file containing a complete analysis sequence (preprocessing,

solution, and postprocessing).
The file must contain a parametrically defined model using parameters
to represent all inputs and outputs to be used as RVs and RPs.

loop

A single pass through the analysis file.

or

In each loop, the PDS uses the values of the RVs from one sample and
executes the user-specified analysis. The PDS collects the values for
the RPs following each loop.

simulation
loop
loop file

The probabilistic design loop file (Jobname.LOOP), created automatically by via the analysis file.
The PDS uses the loop file to perform analysis loops.

probabilistic
model

The combination of definitions and specifications for the deterministic

model (in the form of the analysis file). The model has these components:
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design
PDS Term

Description
Random input variables (RVs)
Correlations
Random output parameters (RPs)
The selected settings for probabilistic method and its parameters.
If you change any part of the probabilistic model, then you will generate different results for the probabilistic analysis (that is, different
results values and/or a different number of results). For example,
modifying the analysis file may affect the results file. If you add or
take away an RV or change its distribution function, you solve a different probabilistic problem (which again leads to different results). If
you add an RP, you will still solve the same probabilistic problem, but
more results are generated.

probabilistic
design database (PDS
database)

The database containing the current probabilistic design environment,

which includes:
Random input variables (RVs)
Correlations between RVs
Random output parameters (RPs)
Settings for probabilistic methods
Which probabilistic analyses have been performed and in which files
the results are stored
Which output parameters of which probabilistic analyses have been
used for a response surface fit, the regression model that has been used
for the fitting procedure, and the results of that fitting procedure.
The database can be saved (to Jobname.PDS) or resumed at any
time. The results of a probabilistic analysis are not stored in the database but in separate files. The samples generated for a fitted response
surface are in neither the database nor in files, because they can be
regenerated very quickly. (Files consume disk space, and reading the
files requires as much time as regenerating the sample data.)

mean value

A measure of location often used to describe the general location of

the bulk of the scattering data of a random output parameter or of a
statistical distribution function.
Mathematically, the mean value is the arithmetic average of the data.
The mean value also represents the center of gravity of the data points.
Another name for the mean value is the expected value.

median
value

The statistical point where 50% of the data is below the median value
and the 50% is above.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design Terminology

PDS Term

Description
For symmetrical distribution functions (Gaussian, uniform, etc.) the
median value and the mean value are identical, while for nonsymmetrical distributions they are different.

standard
deviation

A measure of variability (dispersion or spread) about the arithmetic

mean value, often used to describe the width of the scatter of a random output parameter or of a statistical distribution function.
The larger the standard deviation, the wider the scatter and the more
likely it is that there are data values further apart from the mean value.

solution set

The collection of results derived from the simulation loops performed

for a given probabilistic analysis.
The solution set includes the values of all random input variables and
all random output parameters for all simulation loops of a probabilistic
analysis. A unique label identifies each solution set.

response
surface set

The collection of response surfaces derived from a fitting procedure

(regression analysis) and the results derived from using the response
surfaces for a probabilistic analysis.
A response surface set is identified by a unique response surface label.

remote host

A computer in your local area network used to execute a probabilistic

analysis in parallel mode.
A remote host can have more than one CPU. In parallel processing,
you can use multiple CPUs on remote hosts.

The following figure shows the flow of information during a probabilistic design analysis. The analysis
file must exist as a separate entity, and that the probabilistic design database is not part of the model
database.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design
Figure 1.2: Probabilistic Design Data Flow
File.DB
ANSYS

Database
File
SAVE

RESUME

ANSYS

/EXIT
(Save
Everything)
Analysis File
(parametrically
defined model)

PDEXE

Model
Database

/CLEAR
Probabilsitic
Design
Database

PDSAVE

PDEXE

File.LOOP
Loop File

PDRESU

File.PDS
Prob. Design
Database
File

1.3. Using Probabilistic Design

You can approach a probabilistic design as a batch run or interactively through the Graphical User Interface (GUI). The approach you take depends upon your experience and preference for interacting with
the program.
You may want to perform the entire probabilistic design analysis by creating a command input file and
submitting it as a batch job. The command method may be more efficient for complex analyses (for
example, nonlinear) requiring extensive run time.
The interactive features of probabilistic design offer flexibility and immediate feedback for review of
loop results. When performing probabilistic design via the GUI, it is important to first establish the
analysis file for your model.
The usual process for probabilistic design consists of the following general steps. The steps may vary
slightly, depending on whether you are performing probabilistic design interactively (through the GUI)
or in batch mode. The items in parentheses indicate which processor is necessary to perform the given
task.
1. Create an analysis file for use during looping. The file should represent a complete analysis sequence
and must do the following:
Build the model parametrically (PREP7).
Obtain the solution(s) (SOLUTION).

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

Retrieve and assign to parameters the quantities that will be used as random input variables and
random output parameters (POST1/POST26).
2. Establish parameters in the database which correspond to those used in the analysis file. This step is
typical, but not required (Begin or PDS); however, if you skip this step, then the parameter names are
not available for selection in interactive mode.
3. Enter PDS and specify the analysis file (PDS).
4. Declare random input variables (PDS).
5. Visualize random input variables (PDS). Optional.
6. Specify any correlations between the RVs (PDS).
7. Specify random output parameters (PDS).
8. Select the probabilistic design tool or method (PDS).
9. Execute the loops required for the probabilistic design analysis (PDS).
10. Fit the response surfaces (if you did not use a Monte Carlo Simulation method) (PDS).
11. Review the results of the probabilistic analysis (PDS).
Because analyzing complex problems can be time-consuming, you have the option of running a probabilistic analysis on a single processor or distributing the analyses across multiple processors. By using
the PDS parallel-run capabilities, you can run many analysis loops simultaneously and reduce the
overall run time for a probabilistic analysis.

1.3.1. Create the Analysis File

The analysis file is crucial to probabilistic design. The probabilistic design system (PDS) uses the analysis
file to form the loop file, which in turn is used to perform analysis loops. Any type of analysis (structural,
thermal, magnetic, etc.; linear or nonlinear) may be incorporated into the analysis file.
The model must be defined in terms of parameters (both RVs and RPs). Only numerical scalar parameters
are used by the PDS. See Use ANSYS Parameters in the Modeling and Meshing Guide for a discussion of
parameters. See the ANSYS Parametric Design Language Guide for a discussion of the ANSYS Parametric
Design Language (APDL).
It is your responsibility to create and verify the analysis file. It must represent a clean analysis that will
run from start to finish. Most nonessential commands (such as those that perform graphic displays,
listings, status requests, etc.) should be stripped off or commented out of the file. Maintain only those
display commands that you want to see during an interactive session (such as EPLOT), or direct desired
displays to a graphics file (/SHOW). Because the analysis file will be used iteratively during probabilistic
design looping, any commands not essential to the analysis will decrease efficiency.
You can create an analysis file by inputting commands line by line via a system editor, or you can create
the analysis interactively in the program and use the command log as the basis for the analysis file.
Creating the file with a system editor is the same as creating a batch input file for the analysis. (If you
are performing the entire probabilistic design in batch mode, the analysis file is usually the first portion
of the complete batch input stream.) This method allows you full control of parametric definitions
through exact command inputs. It also eliminates the need to clean out unnecessary commands later.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design
You prefer to perform the initial analysis interactively, and then use the resulting command log as the
basis for the analysis file. In this case, you must edit the log file to make it suitable for probabilistic
design looping. For more information about using the log files, see Using the ANSYS Session and
Command Logs in the Operations Guide.

1.3.1.1. Example Problem Description

The simple beam problem introduced earlier illustrates a probabilistic design analysis.
Figure 1.3: A Beam Under a Snow Load

H2

H1

E
Young's modulus is 20E4.

1.3.1.2. Build the Model Parametrically

PREP7 is used to build the model in terms of the RV parameters. For our beam example, the RV parameters are H1 (snow height at left end), H2 (snow height at right end), and the Young's modulus E.
...
! Initialize parameters:
H1=100
! Initialize snow height H1 @ left end (in mm)
H2=100
! Initialize snow height H2 @ right end(in mm)
YOUNG=200.0e3
! Initialize the Young's modulus (in N/mm**2)
ROOFWDT=1000.0
! Initialize roof width left and right of beam (in mm)
BWDT=10.0
! Initialize beam width (in mm)
BHGT=40.0
! Initialize beam height (in mm)
BLEN=3000.0
! Initialize beam length (in mm)
SNOWDENS = 200e-9
! Density of snow (200 kg/m**3)
GRAVACC = 9.81
! Earth gravity (in N/kg)
LOAD1 = H1*GRAVACC*ROOFWDT*SNOWDENS ! Pressure load due to snow @ left end
LOAD2 = H2*GRAVACC*ROOFWDT*SNOWDENS ! Pressure load due to snow @ right end
DELLOAD = LOAD2-LOAD1
!
! Material definitions:
MP,EX,1,YOUNG
! Young's modulus
MP,PRXY,1,0.3
! Poisson's ratio
!
! Create the geometry (a line)
K,1,0,0,0
! keypoint at left end
K,2,BLEN,0,0
! keypoint at right end
L,1,2,100
! line between keypoints
!
! Mesh definitions
ET,1,BEAM188
! 3-D beam element
SECTYPE,1,BEAM,RECT
! Define a rectangular cross-section
SECDATA,BWDT,BHGT
! Describe the cross-section using RV
! parameters
LATT,1,1,1
LMESH,1
! Mesh the line
FINISH
! Leave PREP7
...

As mentioned earlier, you can vary virtually any aspect of the design: dimensions, shape, material
property, support placement, applied loads, etc. The only requirement is that the design be defined in

10

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

terms of parameters. The RV parameters (H1, H2, and E in this example) may be initialized anywhere,
but are typically defined in PREP7.

Caution
If you build your model interactively (through the GUI), you will encounter many situations
where data can be input through graphical picking (such as when defining geometric entities).
Because some picking operations do not allow parametric input (and PDS requires parametric
input), you should avoid picking operations. Instead, use menu options that allow direct input
of parameters.

1.3.1.3. Obtain the Solution

The SOLUTION processor is used to define the analysis type and analysis options, apply loads, specify
load step options, and initiate the finite element solution. All data required for the analysis should be
specified: appropriate convergence criteria for nonlinear analyses, frequency range for analysis, and so
on. Loads and boundary conditions may also be RVs as illustrated for the beam example here.
The SOLUTION input for our beam example could look like this:
...
/SOLU
ANTYPE,STATIC
D,1,UX,0,,,,UY
D,2,UY,0,,,,
!D,2,UX,0,,,,UY
elem=0
*get,numele,ELEM,,COUNT
*DO,i,1,numele
elem=elnext(elem)
node1=NELEM(elem,1)
node2=NELEM(elem,2)
x1 = NX(node1)
x2 = NX(node2)
ratio1 = x1/BLEN
ratio2 = x2/BLEN
p1 = LOAD1 + ratio1*DELLOAD
p2 = LOAD1 + ratio2*DELLOAD
SFBEAM,elem,1,PRES,p1,p2
*ENDDO
SOLVE
FINISH
...

!
!
!
!

Static analysis (default)

UX=UY=0 at left end of the beam
UY=0 at right end of the beam
UX=UY=0 at right end of the beam

!
!
!
!
!

get
get
get
get
get

!
!
!
!

evaluate pressure at left node

evaluate pressure at left node
Transverse pressure varying linearly
as load per unit length in negative Y

number of next selected element

the node number at left end
the node number at right end
the x-location of left node
the x-location of rigth node

! Leave SOLUTION

This step is not limited to just one analysis. You can, for instance, obtain a thermal solution and then
obtain a stress solution (for thermal stress calculations).
If your solution uses the multiframe restart feature, all changes to the parameter set that are made after
the first load step will be lost in a multiframe restart. To ensure that the correct parameters are used
in a multiframe restart, you must explicitly save (PARSAV) and resume (PARESU) the parameters for
use in the restart. See the Basic Analysis Guide for more information on multiframe restarts.

1.3.1.4. Retrieve Results and Assign as Output Parameters

This is where you retrieve results data and assign them to random output parameters to be used for
the probabilistic portion of the analysis. Use the *GET command (Utility Menu> Parameters> Get
Scalar Data), which assigns program-calculated values to parameters, to retrieve the data. POST1 is
typically used for this step, especially if the data are to be stored, summed, or otherwise manipulated.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

11

Probabilistic Design
In our beam example, the maximum deflection and the maximum stress of the beam are random output
parameters (RPs). The parameters for these data may be defined as follows:
...
/POST1
SET,FIRST
NSORT,U,Y
*GET,DMAX,SORT,,MIN
!
! Derived data for line
ETABLE,VOLU,VOLU
ETABLE,SMAX_I,NMISC,1
ETABLE,SMAX_J,NMISC,3
!
ESORT,ETAB,SMAX_I,,1
*GET,SMAXI,SORT,,MAX
ESORT,ETAB,SMAX_J,,1
*GET,SMAXJ,SORT,,MAX
SMAX=SMAXI>SMAXJ

! Sorts nodes based on UY deflection

! Parameter DMAX = maximum deflection
elements are accessed through ETABLE:
! VOLU = volume of each element
! SMAX_I = max. stress at end I of each
! element
! SMAX_J = max. stress at end J of each
! element
!
!
!
!
!
!
!
!

Sorts elements based on absolute value

of SMAX_I
Parameter SMAXI = max. value of SMAX_I
Sorts elements based on absolute value
of SMAX_J
Parameter SMAXJ = max. value of SMAX_J
Parameter SMAX = greater of SMAXI and
SMAXJ, that is, SMAX is the max. stress

FINISH
...

See the *GET and ETABLE commands for more information.

1.3.1.5. Prepare the Analysis File

If you create your model interactively, you must derive the analysis file from the interactive session.
Use the command log or the session log file to do so. For more information about the log files, see
Using the ANSYS Session and Command Logs in the Operations Guide.

Note
Do not use the /CLEAR command in your analysis file as this will delete the probabilistic
design database during looping. If this happens, the random input variables are no longer
recognized during looping and you will get the same (deterministic) results for all simulation
loops; however, resume the database using the RESUME command as part of your analysis
file. For example, this is helpful if the variations of the random input variables do not require
that meshing is done in every loop (because the mesh is not effected). In this case you can
mesh your model, save the database, and resume the database at the beginning of the
analysis file.

1.3.2. Establish Parameters for Probabilistic Design Analysis

After completing the analysis file, you can begin the probabilistic design analysis. (You may need to
reenter the program if you edited the analysis file at the system level.)
When performing probabilistic design interactively, it is advantageous (but optional) to first establish
the parameters from the analysis file in the database. (It is unnecessary to do so in batch mode.)
To establish the parameters in the database:
Resume the database file (Jobname.DB) associated with the analysis file. This establishes your entire
model database, including the parameters. To resume the database file:
Command(s): RESUME

12

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

GUI: Utility Menu> File> Resume Jobname.db
Utility Menu> File> Resume from
Read the analysis file to perform the complete analysis. This establishes your entire model database,
but might be time-consuming for a large model. To read the analysis file:
Command(s): /INPUT
GUI: Utility Menu> File> Read Input from
Restore only the parameters from a previously saved parameter file; that is, read in a parameter file
that you saved using either the PARSAV command or the Utility Menu> Parameters> Save Parameters menu path. To resume the parameters:
Command(s): PARRES
GUI: Utility Menu> Parameters> Restore Parameters
Recreate the parameter definitions as they exist in the analysis file. Doing this requires that you know
which parameters were defined in the analysis file.
Command(s): *SET
GUI: Utility Menu> Parameters> Scalar Parameters
You may elect to do none of the above, and instead use the PDVAR command to define the parameters
that you declare as probabilistic design variables. See Declare Random Input Variables (p. 14) for information on using PDVAR.

Note
The database does not need to contain model information corresponding to the analysis file
to perform probabilistic design. The model input is automatically read from the analysis file
during probabilistic design looping.

1.3.3. Enter the PDS and Specify the Analysis File

The remaining steps are performed within the PDS processor.
Command(s): /PDS
GUI: Main Menu> Prob Design
In interactive mode, you must specify the analysis file name. This file is used to derive the probabilistic
design loop file Jobname.LOOP. The default for the analysis file name is Jobname.pdan. You can
also specify a name for the analysis file:
Command(s): PDANL
GUI: Main Menu> Prob Design> Analysis File> Assign
For a probabilistic design run in batch mode, the analysis file is usually the first portion of the batch
input stream, from the first line down to the first occurrence of /PDS. In batch mode, the analysis file
name defaults to Jobname.BAT (a temporary file containing input copied from the batch input file).
Therefore, you normally do not need to specify an analysis filename in batch mode. However, if for
some reason you have separated the batch probabilistic design input into two files (one representing

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

13

Probabilistic Design
the analysis and the other containing all probabilistic design operations), then you will need to specify
the analysis file using PDANL after entering the PDS (/PDS).

Note
In the analysis file, the /PREP7 and /PDS commands must occur as the first nonblank characters
on a line. (Do not use the $ delimiter on a line containing either of these commands.) This
requirement is necessary for proper loop file construction.
You cannot assign a different analysis file using the PDANL command after a probabilistic
analysis has been performed. This ensures the integrity of the previously generated results
with the specified probabilistic model.
The program cannot restrain you from editing the analysis file or exchanging it with system level commands. If you do so, then it is your responsibility to ensure the integrity of the generated results with
the definitions in the analysis file. If you are not sure that this integrity is maintained or if you know
that it is not, then we recommend that you save the current PDS database via the PDSAVE command
and then clear the probabilistic analysis results from the probabilistic design database using the PDCLR,
POST command. The PDCLR command does not delete the result files that have been generated; it
erases the link to the result files from the database.
In the example of a beam supporting a roof with a snow load you could store the analysis file in a
macro named beam.mac. Here, the analysis is specified with the commands:
...
/PDS
PDANL,beam,mac
...

1.3.4. Declare Random Input Variables

The next step is to declare random input variables, that is, specify which parameters are RVs. To declare
random input variables:
Command(s): PDVAR
GUI: Main Menu> Prob Design> Prob Definitns> Random Input
If the parameter name that you specify with the PDVAR command is not an existing parameter, the
parameter is automatically defined in the database with an initial value of zero.
For random input variables you must specify the type of statistical distribution function used to describe
its randomness as well as the parameters of the distribution function. For the distribution type, you can
select one of the following:
Gaussian (Normal) (GAUS):

14

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

Ax@f

You provide values for the mean value and the standard deviation of the random variable x.
Truncated Gaussian (TGAU):

nim

You provide the mean value and the standard deviation of the non-truncated Gaussian distribution
and the truncation limits xmin and xmax.
Lognormal option 1 (LOG1):


You provide values for the mean value and the standard deviation of the random variable x. The
PDS calculates the logarithmic mean and the logarithmic deviation :

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

15

Probabilistic Design

Lognormal option 2 (LOG2):

Ax@f

x
You provide values for the logarithmic mean value and the logarithmic deviation . The parameters
and are the mean value and standard deviation of ln(x):

Triangular (TRIA):

vlm

nim

You provide the minimum value xmin, the most likely value limit xmlv and the maximum value xmax.
Uniform (UNIF):

16

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

Ax@f
nimx

xmx
x

You provide the lower and the upper limit xmin and xmax of the random variable x.
Exponential (EXPO):

You provide the decay parameter and the shift (or lower limit) xmin of the random variable x.
Beta (BETA):

tDr




You provide the shape parameters r and t and the lower and the upper limit xmin and xmax of the
random variable x.
Gamma (GAMM):

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

17

Probabilistic Design
Ax@f

k
x

You provide the decay parameter and the power parameter k.

Weibull (Type III smallest) (WEIB):

rhDm
nim

You provide the Weibull characteristic value xchr , the Weibull exponent m and the minimum value
xmin. Special cases: For xmin = 0 the distribution coincides with a two-parameter Weibull distribution.
The Rayleigh distribution is a special case of the Weibull distribution with = xchr - xmin and m = 2.
You may change the specification of a previously declared random input variable by redefining it. You
may also delete a probabilistic design variable (PDVAR,Name,DEL). The delete option does not delete
the parameter from the database; it simply deactivates the parameter as a probabilistic design variable.

Note
Changing the probabilistic model by changing a random input variable is not allowed after
a probabilistic analysis has been performed. This ensures the integrity of the previously
generated results with the specified probabilistic model. If you need to change one or more
random input variables (for example, because you learned that some specifications were incorrect after running an analysis), then we recommend that you save the current PDS database
(using the PDSAVE command) and then clear the probabilistic analysis results from the
probabilistic design database (using the PDCLR,POST command). The PDCLR command does
not delete the result files that have been generated, it simply removes the link to the results
file from the database.

18

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

In the example of a beam supporting a roof with a snow load, you could measure the snow height on
both ends of the beam 30 different times. Suppose the histograms from these measurements look like
the figures given below.
Figure 1.4: Histograms for the Snow Height H1 and H2

6.0E-01

Relative Frequency

5.0E-01
4.0E-01
3.0E-01
2.0E-01
1.0E-01

02
E+

66

02
3.

E+
85

02
2.

E+
2.

03

02
E+

22
1.

4.

07

E+

01

0.0E+00

Snow Height H1

8.0E-01

Relative Frequency

7.0E-01
6.0E-01
5.0E-01
4.0E-01
3.0E-01
2.0E-01
1.0E-01

9.

85
E+
01
2.
95
E+
02
4.
92
E+
02
6.
89
E+
02
8.
86
E+
02
1.
08
E+
03

0.0E+00

Snow Height H2

From these histograms you can conclude that an exponential distribution is suitable to describe the
scatter of the snow height data for H1 and H2. Suppose from the measured data we can evaluate that
the average snow height of H1 is 100 mm and the average snow height of H2 is 200 mm. The parameter
can be directly derived by 1.0 divided by the mean value which leads to 1 = 1/100 = 0.01 for H1,
and 1 = 1/200 = 0.005 for H2. From measurements of the Young's modulus you see that the Young's
modulus follows a Gaussian distribution with a standard deviation of 5%. Given a mean value of 200,000
N/mm2 for the Young's modulus this gives a standard deviation of 10,000 N/mm2. These definitions can
be specified using the following commands:
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

19

Probabilistic Design
...
PDVAR,H1,EXPO,0.01
PDVAR,H2,EXPO,0.005
PDVAR,YOUNG,GAUS,200000,10000
...

1.3.5. Visualize Random Input Variables

After you define your random input variables, use visual tools to verify them. You can plot individual
RVs, and you can obtain specific information about them through an inquiry.
Command(s): PDPLOT, PDINQR
GUI: Main Menu> Prob Design> Prob Definitns> Plot
Main Menu> Prob Design> Prob Definitns> Inquire
The PDPLOT command plots the probability density function as well as the cumulative distribution
function of a defined random input variable. This allows you to visually check that your definitions are
correct.
Use the PDINQR command to inquire about specific information for a defined random input variable
by retrieving statistical properties or probing the two function curves that are plotted with the PDPLOT
command. For example you can inquire about the mean value of a random input variable or evaluate
at which value the cumulative distribution function reaches a certain probability. The result of this inquiry
is stored in a scalar parameter.

1.3.6. Specify Correlations Between Random Variables

In a probabilistic design analysis, random input variables can have specific relationships to each other,
called correlations. If two (or more) random input variables are statistically dependent on each other,
then there is a correlation between those variables.
To define these correlations:
Command(s): PDCORR
GUI: Main Menu> Prob Design> Prob Definitns> Correlation
You specify the two random input variables for which you want to specify a correlation, and the correlation coefficient (between -1 and 1). To remove a correlation, enter DEL in the correlation coefficient
field (PDCORR,Name1,Name2,DEL)
In the example of a beam supporting a roof with a snow load, the data for the snow height indicates
that there is a correlation between the snow height at one end versus the snow height at the other
end. This is due to the fact that it is unlikely that one end of the beam has very little snow (or no snow
at all) at the same time that the other end carries a huge amount of snow. In the average the two snow
heights tend to be similar. This correlation is obvious if you plot the measured data values for H2 versus
the data value for H1. This scatter plot looks like this:

20

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

Figure 1.5: A Scatter Plot of Snow Height H1 vs. H2
600

Snow height H2

500
400
300
200
100
0

100

200

300

400

Snow height H1

Performing a statistical evaluation of the data, we can conclude that the linear correlation coefficient
between the values of H1 and H2 is about 0.8. You can define this correlation using the commands:
...
PDVAR,H1,EXPO,0.01
PDVAR,H2,EXPO,0.005
PDCORR,H1,H2,0.8
...

You may have a more complex correlation where you have a spatial dependency. If so, you can use the
PDCFLD command to calculate a correlation field and store it into an array.
Random fields are random effects with a spatial distribution; the value of a random field not only varies
from simulation to simulation at any given location, but also from location to location. The correlation
field describes the correlation coefficient between two different spatial locations. Random fields can be
either based on element properties (typically material) or nodal properties (typically surface shape
defined by nodal coordinates). Hence, random fields are either associated with the selected nodes or
the selected elements. If a random field is associated with elements, then the correlation coefficients
of the random field are calculated based on the distance of the element centroids.
Note that for correlation fields, the domain distance D({xi} , {xj}) is not the spatial distance |{xi} - {xj}|,
but the length of a path between {xi} and {xj} that always remains inside the finite element domain.
However, exceptions are possible in extreme meshing cases. For elements that share at least one node,
the PDCFLD evaluates the distance by directly connecting the element centroids with a straight line.
If these neighboring elements form a sharp inward corner then it is possible that the domain distance
path lies partly outside the finite element domain, as illustrated below.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

21

Probabilistic Design
After the correlation coefficients have been calculated and stored in a parameter (PDCFLD,ParR), then
you can use the PDCORR command to define the correlations between the elements of the random
field.

Note
When specifying one variable (A) with correlations to two or more other variables (B, C,
etc.), be certain that you consider the relationship implied between the other variables
B and C, etc. If you specify high correlations between A and B and A and C, without
specifying the relationship between B and C, you might receive an error. Specifying a
relatively high correlation between A and B, with only a moderate correlation between
A and C might work because the logical correlation between B and C could still be low
or nonexistent.

Example
The structure illustrated below is modeled with 12 elements. We will evaluate the domain distances of
the element centroids.
Y
4.
1

3.
5

2.

1.
9

10

11

12

0.
0.

1.

2.

3.

4. X

First, build the finite element structure:

...
/PREP7
et,1,shell181,,,2
! create the nodes
N,1,0,4
N,2,1,4
N,3,2,4
N,4,3,4
N,5,4,4
N,6,0,3
N,7,1,3
N,8,2,3
N,9,3,3
N,10,4,3
N,11,1,2
N,12,2,2
N,13,3,2
N,14,4,2
N,15,0,1
N,16,1,1
N,17,2,1
N,18,3,1
N,19,4,1

22

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

N,20,0,0
N,21,1,0
N,22,2,0
N,23,3,0
N,24,4,0
! create the elements
E,1,2,7,6
E,2,3,8,7
E,3,4,9,8
E,4,5,10,9
E,7,8,12,11
E,9,10,14,13
E,11,12,17,16
E,13,14,19,18
E,15,16,21,20
E,16,17,22,21
E,17,18,23,22
E,18,19,24,23
...

Next, calculate the domain distances and store the results in the array elemdist:
...
/PDS
PDCFLD,elemdist,ELEM,DIST
...

Finally, get all the element domain distances and print them:
...
*GET,numsel,ELEM,0,COUNT
! Get the number of selected elements
!
! Outer loop through all selected elements from first to last
index=0
elem1=0
! Pipe output to file
/OUT,elements,dat
*DO,i,1,numsel
elem1=ELNEXT(elem1)
! get number of next selected element
*IF,elem1,EQ,0,CYCLE
! Leave do loop if no more elements
!
! Inner loop through selected elements from "elem1+1" to last
elem2=elem1
*DO,j,i+1,numsel
elem2=ELNEXT(elem2)
! get number of next selected element
*IF,elem2,EQ,0,CYCLE
! Leave do loop if no more elements
index=index+1
!
! Print out the element distance
*MSG,INFO,elem1,elem2,elemdist(index)
Distance between element %i and %i is %g
*ENDDO
! go to next element for inner loop
*ENDDO
! go to next element for outer loop
...

The print out will show that for the structure illustrated above the "domain distance" between the element
centroids of elements 1 and 9 is 3.8284 and between the element centroids of elements 1 and 12 it is
4.8284. The paths related to these distances are sketched in the illustration with a solid line and a dashed
line respectively. In this example there are 12 elements, thus the array "elemdist" has a length of 12*(121)/2 = 66.

1.3.7. Specify Random Output Parameters

After declaring your input variables and correlations among them you must define the random output
parameters. The random output parameters (RPs) are results parameters that you are interested in. To
define random output parameters:
Command(s): PDVAR,Name,RESP
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

23

Probabilistic Design
GUI: Main Menu> Prob Design> Prob Definitns> Random Output

1.3.8. Select a Probabilistic Design Method

In the Probabilistic Design System, several probabilistic design methods are available.
You can select one of two primary methods, the Monte Carlo Simulation (default) or the Response
Surface Method.
Options under the Monte Carlo Simulation method include the Latin Hypercube Sampling method
(default) and the Direct Monte Carlo Sampling method.
Options under the Response Surface Method include the Central Composite Design and the Box-Behnken
Matrix Design method.
Both the Monte Carlo Simulation and the Response Surface Methods allow a user-defined option. See
the PDDMCS, PDLHS, and the PDDOEL, commands or Probabilistic Design Techniques (p. 51) for further
details about these methods.
To specify a method to be used for probabilistic design looping:
Command(s): PDMETH
GUI: Main Menu> Prob Design> Prob Method> Monte Carlo Sims
Main Menu> Prob Design> Prob Method> Response Surface

Note
To use Response Surface Methods, the random output parameters must be smooth and
continuous functions of the involved random input variables. Do not use Response
Surface Methods if this condition is not satisfied.

1.3.8.1. Probabilistic Method Determination Wizard

You can use the Probabilistic Method Determination Wizard to find the fastest method for solving your
probabilistic analysis. You should have completed one analysis prior to starting the wizard.
Use Main Menu> Prob Design> Prob Method> Methods Wizard to access the wizard. Answer the
questions on the screens as they are presented. Before you start the wizard, you should know:
How long did it take to run the analysis file (hours, minutes, seconds)? Or, you should give an estimation of the time it will take if you haven't run the analysis yet.
How many CPUs are available to run the probabilistic analysis (if running parallel processing)?
Is your output parameter a smooth/continuous function of the input parameters?
What results are you interested in evaluating? Options are mean values, standard deviation, sensitivities, or acceptable part failures.
Below is one of the wizard screens, as an example.

24

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

Figure 1.6: The PDS Method Determination Wizard

Based on the information you provide, the wizard will tell you the fastest method for solving your
probabilistic design problem. The wizard will issue the PDMETH command and either the PDLHS or
the PDDOEL command. You will still need to run the analysis, then fit the results to a response surface,
etc. to evaluate your results.

1.3.9. Execute Probabilistic Analysis Simulation Loops

You can execute your analysis on your computer alone (serial execution), or using other computers in
your network to save running time and speed processing (parallel execution).
If you want to use serial processing only, select the serial option from the Run menu.
If you want to run parallel analyses on multiple CPUs, you must first set the parallel options before
performing the analysis. (See PDS Parallel Analysis Runs (p. 27) for more details).
Command(s): PDEXE
GUI: Main Menu> Prob Design> Run> Exec Serial> Run Serial
Main Menu> Prob Design> Run> Exec Parallel> Run Parallel

Caution
For security reasons, ANSYS, Inc. recommends that you use parallel processing only within
the firewall of your local area network.
If you use serial processing to perform a probabilistic analysis, you will use only the CPU of the computer
you are working on. If you have access to only one product license, or if you have access to only one
computer, then this is the only way in which you can run a probabilistic analysis. While the simulation
loops are running in serial mode, your session is locked (you cannot perform other tasks in the same
session). If you are running your session in interactive mode, the simulation loops are also performed
in interactive mode. If you are running your session in batch mode, the simulation loops are performed
in batch mode.
If you select the PDS parallel-processing option, you can use other CPUs that you have access to for
running the probabilistic analysis. PDS parallel processing can distribute the necessary jobs in a local
area network. With this option, the simulation loops are sent to CPUs that you can specify, where they
are executed in "server mode." This looks the same as a batch run (in other words, there is no interactive
visualization during the execution of a simulation loop). While the simulation loops are running in parallel mode, your session is locked; however, you can instruct the session to start postprocessing the
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

25

Probabilistic Design
probabilistic results as they are calculated so you can review and visualize the results before all simulation
loops are finished.
In parallel processing, you can monitor the running jobs and the completed results.
When using parallel-processing, for n available licenses, n - 1 will be available for PDS solutions, as one
license is used as a PDS administrator. For example, if you have 8 licenses available, you will be able to
run 7 PDS solutions concurrently.

1.3.9.1. Probabilistic Design Looping

Regardless of whether you opt for serial or parallel processing, the program controls the analysis looping.
You cannot modify this process, which does the following:
Always reads the analysis file from the beginning.
Always ignores the RV settings and replaces their value with the derived value that the probabilistic
method has assigned to an RV for a particular loop.
Always reads the PDS database after each loop.
For the execution of the simulation loops you must specify a solution label (Slab on the PDEXE command). The solution label is used for several purposes:
The results are stored in an ASCII readable file under the name "jobname_Slab.pdrs". Here, Slab is
the user-specified solution label.
If you want to fit response surfaces to the results of the random output parameters then you need
to specify the solution label to indicate which set of results you want to use for the fitting procedure.
If you want to postprocess the results generated with the PDEXE command then you must specify
the solution label to indicate which results you want to postprocess.
When you execute the probabilistic analysis (PDEXE), the program creates a probabilistic design loop
file (Jobname.LOOP) from the analysis file. This loop file is used to perform analysis loops. Looping
continues until all parameters have been evaluated.
If a loop is interrupted due to an error in the execution run (for example, a meshing failure, a nonconverged nonlinear solution, etc.), PDS aborts that loop. Further processing depends if you are in
serial or parallel processing mode. If you are using:
Serial interactive processing: you can terminate when you receive the error, or continue processing.
Serial batch processing: processing terminates at the first error.
Parallel processing: processing terminates if the allowed number of failed loops is exceeded (set in PDEXE),
otherwise it continues.
Note that for all failed loops (loops with errors), the results for that loop are discarded, no data from
that loop is used for post processing.
After the PDEXE command is issued, the PDS generates a file containing the input sample values. The
file is called jobname.samp. An example of the content of the file is given below:
TEST1
ITER CYCL

26

LOOP

X1

X2

X3

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

1
1
1
..
..

1
1
1
..
..

1
2
3
..
..

1.619379209e+000
2.237676559e-001
7.931615474e+000
...
...

2.364528435e-001
5.788049712e-001
8.278689033e-001
...
...

1.470789050e+000
1.821263115e+000
2.170793522e+000
...
...

The first line contains the solution label (the parameter Slab is set via the PDEXE command); the
second line contains the headers of the data columns - the iteration number, cycle number, loop
number, and the random variable names. The iteration number and cycle number tell the PDS to which
group (with specific PDS method and settings) the loops belong. Subsequent lines provide specific iteration, cycle, loop, and input sample values for the defined input variables.
The PDS also creates a file where the results are stored. The name of the results file is jobname_Slab.pdrs. Before the job is executed, the file looks like this:
TEST1
ITER CYCL

LOOP ERR

X1

X2

X3

RESULT

In the first line, the PDS enters the solution label. In the second line are the headers for the data columns:
the first four columns are the iteration, cycle number, loop number, and an error flag. The fifth and
subsequent columns are for the random input variable and random output parameter values. If you
run a subsequent analysis (same type of analysis with the same solution label), the cycle is incremented,
the loop count is reset to 1, and the result file is appended.
For example, the content of the result file could look like this:
TEST1
ITER CYCL
1
1
1
1
1
1
..
..
..
..

LOOP ERR
1
0
2
0
3
0
.. ..
.. ..

X1
1.619379209e+000
2.237676559e-001
7.931615474e+000
...
...

X2
2.364528435e-001
5.788049712e-001
8.278689033e-001
...
...

X3
1.470789050e+000
1.821263115e+000
2.170793522e+000
...
...

RESULT
4.162928057e+000
4.744249212e+000
1.149997825e+001
...
...

Note
Loops ending with an error are deemed "not trustworthy", i.e. if the loop lead to an error,
then the calculated results are probably wrong. Those loops will have the error flag in the
fourth column set to "1" instead of "0". Those loops will be excluded from the probabilistic
postprocessing altogether, i.e. the loops will not be used for the response surface fitting and
also the statistical analysis in connection with a Monte Carlo simulation will skip those loops.

1.3.9.2. Serial Analysis Runs

Serial execution of the probabilistic analysis is useful for smaller models, or if you have access to only
one machine.
Command(s): PDEXE
GUI: Main Menu> Prob Design> Run> Exec Serial> Run Serial

1.3.9.3. PDS Parallel Analysis Runs

To save time when analyzing large models with PDS, use the parallel processing option available under
the PDEXE command, if you have multiple CPUs available.
To successfully execute a probabilistic analysis in parallel mode you need to follow three steps:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

27

Probabilistic Design
1. Configure the remote machines that you want to use for parallel processing. See Machine Configurations (p. 29) for details. You must have account privileges on all remote hosts.
2. Configure the local machine that you want to use for managing the parallel processing. See Configuring the Master Machine (p. 33) for details.
3. Start the parallel execution.
Each of these steps is explained in detail in the following sections.
An understanding of the following terms will be useful in the discussion of the parallel analysis.
Simulation: A simulation is a set of input variables used with the analysis file to produce the output
variables
Parent Process: The parent process is the executable which manages the creation of input parameters,
communication with child processes, and postprocessing of output parameters (i.e. this is where the
session runs, which you started to perform a probabilistic design analysis). There is only one parent
process for a parallel analysis.
Child Process: A child process is a server running on a particular slave machine which processes a
simulation. There can be any number of children for a given parent process.
Master Machine: This is the machine on which you are running the parent process. If this machine
has more than one CPU you may want to also use it as a slave machine.
Slave Machine: This is the name of the machine on which you are running the child process. If the
slave machine has more than one CPUyou may want to run more than one child process.
ANSYS Nanny: This is a program in the parent process that starts a child process on a slave machine;
manages the transfer of files and data between the parent process and child processes and eventually
terminates the child processes on the various slave machines and removes temporary files.
ANSYS Thin Server: This is an application, which must be running on the slave machine, which is
responsible for tranferring files between the master machine and slave machine as well as starting a
session for each child process. This application may be started automatically (remote shell option) or
you can start it manually (connection port option).
ANSYS Server: This is the application using TCP/IP which receives commands and return information
to the parent process (client-server).
Connection Port: This is the port on which the ANSYS Thin Server accepts connections from a Master
Machine.
Communication Port: This is a port on which the ANSYS Thin Server will allow communications from
a Master Machine.

Caution
For security reasons, ANSYS, Inc. recommends that you use parallel processing within your
local network protected by a firewall. There is minimal security when using the ANSYS Thin
Server. The client that connects to the server has all of the permissions allowed by the person
or account starting the ANSYS Thin Server.

28

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

1.3.9.3.1. Machine Configurations

The slave machines host one or more child processes. To be able to host a child process the slave machine
must have the ANSYS Thin Server running. The ANSYS Thin Server may be started automatically using
the remote shell option or manually using the connection port option.

1.3.9.3.1.1. Choosing Slave Machines

You can designate any machine that has a licensed copy of the program installed as a remote host
(slave machine) but it is a good practice to follow these guidelines when selecting remote hosts:
Use idling computer power. During regular work hours, computers usually have designated "owners"
using the machines for interactive work. Using these computers for parallel processing can slow down
the machine performance and possibly disturb your colleagues in their work. However, computers
are usually idle at night and can then be used without disturbing others.
Use machines that are dedicated as computer servers. There might be some computers in your work
environment that are not used for interactive work at all. It is fairly safe to use these computers as
remote hosts for parallel-processing at all times.
Check with your System Administrator for additional guidelines.

1.3.9.3.1.2. Using the Remote Shell Option

With this option the ANSYS Thin Server is started and configured to communicate with the Master
machine automatically. To use the remote shell option requires the following on each slave machine:
A remote shell daemon must be running
An account that the master machines remote shell command can communicate with
The master machine and user name must be granted access for the account
The installation path for the executable must be in the lookup path
A valid product license available for the child process
Remote Shell Daemon
Your system administrator should install and configure your machine to assure that this daemon or
service is running. The PC does not ship with a remote shell daemon facility so you will need to acquire
one (you may use the connection port ooption instead).
Account
Your system administrator should create an account specifically for you or an account utilized only for
PDS parallel.
Account Access
Depending on how and which remote shell daemon is installed it will need to know which master
machines and users can access this account. This is typically done using the .rhosts file located in
the account's HOME directory, refer to the documentation for your particular remote shell daemon. So
for example:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

29

Probabilistic Design
On all slave machines, edit/create the .rhosts file in your home directory to include:
MasterMachine1 UserId
MasterMachine2 UserId
MasterMachine3 UserId
MasterMachine4 UserId

each on a separate line. This will allow the user with the account user identification UserId to access
the slave machine from the master machines listed in the file using the remote shell service.
ANSYS in PATH
To be able to run the parallel process the executable and ANSYS Thin Server script must be in your
path. The most effective way to determine this is to do the following.
Determining if the executable is in your path:
On the master machine, use the command rsh SlaveMachine which ansys145 (replace rsh by
remsh if master machine is an HP machine). This should return the string
/ansys145/bin
Here "" is your installation directory If you do not get this then you need to modify your PATH variable
to contain the installation path. The way this is done depends on the operating system being used,
refer to the documentation for your particular O/S environment. Here are some examples for typical
uses:
UNIX
To change the PATH requires you to modify a certain system file depending which shell you are running
under. To find out which shell you are using, issue the UNIX command
echo $shell or echo $SHELL
If the prompt is /bin/csh" or similar, then you are running under c-shell. If the prompt is/bin/ksh
or similar, then you are running under k-shell. If the prompt is /bin/tcsh or similar, then you are
running under tc-shell, which can be treated the same as c-shell.
If you are running under c-shell (or tc-shell) you need include the following line at the end of your
.cshrc file in your home directory:
set path=( .../ansys145/bin $path)
If you are running under k-shell you need include the following line at the end of your .kshrc file in
your home directory:
export PATH={ .../ansys145/bin $PATH }
If you don't have a .cshrc file or .kshrc file in your home directory, then you need to create one
and include the respective commands mentioned above.
PC
To change the PATH variable, go to Control Panel and select System on the Advanced tab. Select
Environment Variables and add the executable to the PATH under System variables.

30

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

Determining if the ANSYS Thin Server script is in your path:
On the master machine issue the command rsh SlaveMachine which ansysts145. This should
return the string
/ansys145/bin
Here "" is your installation directory. If you do not get this then you need to modify your PATH variable
as described above.
Licensing
If the licensing is not managed from a license server, but installed locally then the access to the license
must be made available. Include the following line at the end of your .cshrc file or .kshrc file in
your home directory depending which shell you are running under:
setenv ANSYSLMD_LICENSE_FILE. ../shared_files/licensing/FileName.lic
Here "" is your installation directory. The name FileName.lic is for the file containing the license
key.

1.3.9.3.1.3. Using the Connection Port Option

With this option the communication between the ANSYS Nanny and the ANSYS Thin Server is manually
configured and started. The ANSYS Nanny running on a master machine can only make a connection
to an ANSYS Thin Server using a specific connection port. The ANSYS Nanny reads the connection port
from the hosts145.ans file. When the ANSYS Thin Server is started, with the same specific connection
port, it reads the file AnsysClients from the same directory it is started within to determine the
communication ports and master machines with which it will communicate.
The purpose of these two different port numbers is a two-level authentication mechanism. You can
only use an ANSYS Thin Server on a certain slave machine, if you know which connection port it is using
and those communication ports from which it will accept communications. If the connection port is
not correct, then you will not be able to even make a connection to the ANSYS Thin Server. If the
communication port number is incorrect, then the ANSYS Thin Server will refuse the connection.
To use the connection port option requires the following on each slave machine:
An account that you have access to
The installation path for the executable must be in the lookup path
Configuring the AnsysClients file
A valid product license available for the child process
Start the ANSYS Thin Server
Stop the ANSYS Thin Server
Account
Your system administrator should create an account specifically for you.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

31

Probabilistic Design
ANSYS in PATH
See Using the Remote Shell Option (p. 29) for information on how to place the executable in your path.
AnsysClients
The AnsysClients file is read by the ANSYS Thin Server on startup and must be in the directory in
which you are going to run the ANSYS Thin Server. This file must contain a list of each master machine
IP address and the communication port that the ANSYS Thin Server will accept communications from.
The master machine's IP address must be used, but may be specified with a wildcard to allow connection
from a network domain. The communication port is to be a value between 49512 and 65535. The
communication port can also be specified as a range, which allows for that many connections. It is best
to always specify a range for the communication port numbers. For example:
10.3.* 59100-59130
10.3.20.1 59200-59230
10.2.5.55 59300-59330
10.1.1.104 59400-59430
192.1.10.34 59500-59530

Make sure that the port number ranges are not overlapping and are unique for each master machine
or network domain. You should make the range at least as wide as the number of possible users connecting from a master machine.
Start the ANSYS Thin Server
Starting on a PC
On a PC open a command prompt window and go to the directory containing the AnsysClients
file.
Issue the command: ansysts145 connection port
This will start the ANSYS Thin Server using the specific connection port. The value of connection port
is the port on which the master machines will connect to the ANSYS Thin Server. The value of "connection_port" should be between 49512 and 65535. For example:
ansysts145 62000
The ANSYS Thin Server starts without any message and continues running until the command prompt
window is closed or using Ctrl-C to stop the process. The command prompt window may be minimized.
Starting on a UNIX Machine
On a UNIX machine go to the directory containing the AnsysClients file.
Issue the command: ansysts145 connection_port &
The command starts the ANSYS Thin Server using the specific connection port. The value of connection
port is the port on which the master machines will connect to the ANSYS Thin Server. The value of
"connection_port" should be between 49512 and 65535. For example:
ansysts145 62000&

32

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

The ANSYS Thin Server will start without any message and will continue running in the background
until the machine is restarted or you kill the process. You may close or minimize the window in which
the ANSYS Thin Server was started.
Stop the ANSYS Thin Server
The ANSYS Thin Server should be stopped once the parallel process is complete.
Stopping on a PC
You may stop the process by closing the command prompt window or using Ctrl-C.
Stopping on a UNIX Machine
On a UNIX machine use ps -u UserId | grep tclsh in the command line to determine under
which process ID the ANSYS Thin Server is running. Here, UserId is your user account name on the
machine. With this "ps" command should get two processes running. Typically, there will be two processes
listed, one process for anstclsh and another for tclsh. Use kill -9 process-id to kill both
processes. Under certain circumstances, killing the anstclsh process also kills the tclsh process.
(Issue the ps command again to verify.) If this does not happen, then kill the tclsh process separately.

1.3.9.3.1.4. Configuring the Master Machine

After you have configured the slave machines you need to configure the master machine for the type
of ANSYS Thin Server startup you chose to use on the slave machines. The communication from the
master machine to the slave machines is done by the ANSYS Nanny which is a program running in the
parent process. It takes care of running the necessary simulations on different slave machines. It will
establish the connection to the slave machines, copy the necessary files over to the slave machines,
start and monitor the running simulations and clean up the working directories after the entire sequence
of simulations is finished. The ANSYS Nanny will be started automatically as you start executing PDS
simulations in parallel, distributed mode.
For parallel processing you need to specify the remote hosts you want to use. This information is placed
in a file named hosts145.ans. You can create this file using a text editor or you can use the
ANS_ADMIN utility (see the online help available with the ANS_ADMIN utility for more information).
This file contains host information for all remote machines on which you may want to run. This file is
global and does not contain job-specific information. Typically, your IT administrator would provide
this file, including all the information about the slave machines a typical user can use in your network.
If you have multiple users running parallel or distributed jobs at your site, you should have one
hosts145.ans file for all users. But you can copy this file to your local working directory and make
adjustments. ANSYS searches for this file first in the local directory, followed by the home directory,
and finally the apdl directory.
You have two options when setting up the Master Machine to use the ANSYS Thin Server on the slave
machines.
Configuration when the Thin Server uses the Remote Shell Option
Let's assume that the slave machine called "MySlaveMachine" has been prepared to work under the
remote shell option as outlined above. The slave machine "MySlaveMachine" is an SGI UNIX machine
and it has 4 CPUs. On the slave machine we also want to use the directory "/tmp/sdr/pds_runs" as the
local directory for the remote simulations to run in. In this case your hosts145.ans file must include
the lines:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

33

Probabilistic Design
#
# HOST
OS
PORT CPU
MySlaveMachine SGI64 0

TIME LocPORT I/O Directory

4
15
0
/tmp/sdr/pds_runs

Configuration when the Thin Server uses the Port Option

Assume that you now want to use the same slave machine "MySlaveMachine" using the ANSYS Thin
Server. The ANSYS Thin Server has been started on a slave machine called MySlaveMachine" using the
command ansysts145 62000 as shown above. Assume also that the file AnsysClients looks
exactly like shown in the section above, i.e. your master machine "MyMasterMachine" can communicate
to the ANSYS Thin Server using the communication port numbers 59400-59430. In this case, your
hosts145.ans file must include the lines:
#
# HOST
OS
PORT
MySlaveMachine SGI64

CPU TIME
62000
4

LocPORT
I/O Directory
15
59400-59430
/tmp/sdr/pds_runs

This will ensure that the ANSYS Thin Server on the slave machine will be contacted using the same
connection port it has been started with, i.e. 62000 in this case. Also the communication will use the
same communication port numbers the ANSYS Thin Server accepts from the machine "MyMasterMachine"
where you try to connect from.
An example hosts145.ans file looks like this:
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#

This file is used to specify those hosts that the ANSYS Nanny may
run children on.
Each host entry is to be on its own line.
several fields which are space delimited.

The host entry consists of

Field 1 - host IP address or name

Field 2 - host machine type
Field 3 - execution key (used for Probabilistic Design only):
0-Use a remote shell to start the child process;
this requires a remote shell server to be
running on the host machine.
>1024-Use a running ANSYS thin server on the host
which is listening on this port number.
Field 4 - The default maximum number of jobs to run on this host
Field 5 - The time in minutes to check again if the host is available.
If this is zero then the host will not be checked again.
Field 6 - The local port number to start the communication with the
ANSYS Thin Server on. This is tied to authentication on the
ANSYS Thin Server.
Field 7 - The directory to create the children subdirectories in
Field 8 - The cluster type. Only valid entry is MPI.
Field 9 - The speed factor (relative speed to other machines listed).
Only valid entry is 1.
# Field 10 - Number of OpenMP threads. Only valid entry is 1.
# Example:
#
# UNIX box that has five processors
# zeus sgi64
0
5 30
2000
/scratch/wjc
# Microsoft box using the ANSYS Thin Server
# wjcpc XP
2010 1 0
2000
C:\TEMP
alpha1
alpha
0
1
15
2000
/scratch/epc
MPI 1 1
athena
sgi64
0
1
15
2000
/scratch/epc
MPI 1 1
rs43p
rs6000
0
1
15
2000
/home/pdstest MPI 1 1
rs260
rs64
0
1
15
2000
/home/pdstest MPI 1 1
snoopy
hppa8000 0
1
15
2000
/home/pdstest MPI 1 1
alpha24 alpha
0
1
15
2000
/home/pdstest MPI 1 1
hp770
hppa8000 0
1
15
2000
/home/pdstest MPI 1 1
us60
usparc
0
1
15
2000
/home/pdstest MPI 1 1
ss60
sun64
0
1
15
2000
/home/pdstest MPI 1 1

34

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

1.3.9.3.1.5. Host setup using port option

The picture below illustrates an example of the set-up of a network with 2 master machines and 4 slave
machines using the connection port option. Here, the first master machines uses slave machines 1, 2
and 3, while the second master machine is using only slave machines 2, 3 and 4. In this illustration "NN"
represents the revision number of ANSYS.
IIIIT yn
WWIPSEHHIPS Pretsw
WWIISEHHIIS Iretsw
steilgsse






ynn gis
oisses xe

Iretsw

PPPPT yn
WWPPSEHHPPS Pretsw
WWIISEHHIIS Iretsw
steilgsse

sFstsoh



WWIISE HHIIS QQQQT Qrevre

WWIISE HHIIS PPPPT Prevre
WWIISE HHIIS IIIIT Irevre




QQQQT yn
WWPPSEHHPPS Pretsw
WWIISEHHIIS Iretsw
steilgsse



ynn gis
oisses xe


Pretsw

RRRRT yn
WWPPSEHHPPS Pretsw
WWIISEHHIIS Iretsw
steilgsse

sFstsoh





WWPPSE HHPPS RRRRT Rrevre

WWPPSE HHPPS QQQQT Qrevre
WWPPSE HHPPS PPPPT Prevre

1.3.9.3.1.6. Host and Product selection for a particular analysis

After specifying the hosts145.ans file, provide the specific information that may change from analysis to analysis.
If you are working in interactive mode then select:
Main Menu> Prob Design> Run> Exec Parallel> Host Select
In this menu you can:
Select the slave machines you want to use for a particular analysis. This is necessary, for example, if
you know that some machines in your network are in use by another user at the time you want to
start the analysis.
Select the licenses you want to use for the particular analysis. If there are multiple levels of licenses
of ANSYS available in your network, then you should first select the lower license that will be able
to run your analysis.
Based on the information provided in the menu, a Jobname.hosts file is created or updated as you
press OK in the menu.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

35

Probabilistic Design
If you are working in batch mode, you must create this file using a text editor. However, for sake of
simplicity it is recommended to let ANSYS create this file using interactive mode and then to proceed
with batch mode operation. This file must reside in the directory where you are running ANSYS PDS.
This file must include the following information:
Remote hosts to be used for this particular parallel run.
Number of processes that can run in parallel on each host. If a remote host has more than one CPU,
you can use all of the CPUs on the remote host. For performance reasons, we recommend leaving
one CPU for system tasks and using only N-1 CPUs for parallel processing (if N is the number of CPUs
on the remote host).
Directories in which you want the child processes to be executed. It is recommend that you use
temporary directories like "/scratch" or "/tmp" on UNIX or "C:\TEMP" on PCs. These remote directories
are cleaned up automatically after the parallel processes finish running. Make sure that there is enough
disk space in the directory for the files created by the analysis.
An example jobname.hosts file looks like this:
# This file is used to specify those hosts that the ANSYS Parent may
# run children processes on.
#
# Each host entry is to be on its own line. The host entry consists of
# several fields which are space delimited.
#
# Field 1 - host IP address or name
# Field 2 - username to use for a remote shell to the host
# Field 3 - execution key:
# 0 - Use a remote shell to start the child process
# >1024 - Use a running ANSYS thin server on the host
# which is communicating on this port number
# Field 4 - the product code sequence to attempt to start ANSYS jobs
# on the host with. This is a colon delimited list.
# Field 5 - the number of jobs to run on this host
# Field 6 - The time in minutes to check again if the host is available.
# If this is zero then the host will not be checked again.
# Field 7 - If field 3 is nonzero then this is the local port number
# or range to start the communication with the ANSYS Thin
# Server on. This is tied to the "authentication" on the
# ANSYS Thin Server.
# If field 3 is zero then this should be set to zero.
# Field 8 - directory to create the children subdirectories in
#
# Example:
#
# UNIX box that has five processors and will first attempt to
# run with ANSYS Mechanical Batch Child and then ANSYS Mechanical
# zeus wjc 0 MEBACH:ANSYS 5 30 2000 /scratch/wjc
# XP box running the thin server on port 2010
# wjcpc wjc 2010 ANSYS 1 0 2000 C:\TEMP
alpha1 epc 0 MEBA:MEBACH 1 15 2000 /scratch/epc
athena epc 0 MEBA:MEBACH 1 15 2000 /scratch/epc
rs43p pdstest 0 MEBA:MEBACH 1 15 2000 /home/pdstest
rs260 pdstest 0 MEBA:MEBACH 1 15 2000 /home/pdstest
snoopy pdstest 0 MEBA:MEBACH 1 15 2000 /home/pdstest
hp160 pdstest 0 MEBA:MEBACH 1 15 2000 /home/pdstest
alpha24 pdstest 0 MEBA:MEBACH 2 15 2000 /home/pdstest
hp770 pdstest 0 MEBA:MEBACH 1 15 2000 /home/pdstest
us60 pdstest 0 MEBA:MEBACH 1 15 2000 /home/pdstest
ss60 pdstest 0 MEBA:MEBACH 1 15 2000 /home/pdstest

1.3.9.3.2. Files Needed for Parallel Run

You also need to specify all other files that will be used in the distributed processing run. These files
are listed in a separate file called Jobname.ifiles. At a minimum, this file must include the analysis
36

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

file. Additionally, you may need other files, such as data files or ANSYS macros. You must include all
files that the analysis files calls in order for the PDS run to execute successfully. These files, which are
typically found only on your local machine, will then be copied to all remote hosts during the PDS run.
The Jobname.ifiles must reside in your working directory. If you are working in batch mode, you
need to create this file manually. The first file listed must be the analysis file. Other files listed must include
the full path.
Command(s): PDANL
GUI: Main Menu> Prob Design> Analysis File> Assign
An example jobname.ifiles file looks like this:
# Files to copy to the ANSYS server for job Jobname
# created on Fri Oct 13 10:33:44 EDT 2000.
pdsrun1.inp
/home/staff/epc/ddts/pds/tags

1.3.9.3.3. Controlling Server Processes

If you are working interactively, you need to complete one additional step. When you select Run Parallel,
you see a dialog box listing all remote hosts specified in the Jobname.hosts file, as well as their
current status. You can use this dialog box to specify a solution label, set the number of allowed failures
(loops leading to an ANSYS error during execution) as well as monitor the processes and remote machines, and stop or start processes on those machines as necessary.
The number of allowed failures is used to automatically stop the parallel execution process if the
number of loops leading to an error exceeds the allowed maximum number. In a typical case you don't
expect your analysis file to cause an ANSYS error. If in this case there are several loops leading to an
error, then this is most likely due to a bug in the APDL programming in the analysis file. Here, the ANSYS
PDS automatically terminates the parallel execution if the number of allowed failures (loops with errors)
exceeds the allowed maximum.
In certain cases however, you will know that executing the analysis file always leads to an error that
can be ignored. An example is a non-linear burst analysis, which terminates with an error as the maximum displacement grows out of proportion. I.e. ultimately all loops will terminate with an error and
the result of a burst analysis is the value of the load factor at which this instability happens. In this case
you can force the ANSYS PDS to ignore the error and set the error flag in the result file to "0". To do
this set the number of allowed failures in the exactly equal to the total number of loops. Since you are
bypassing the error checking mechanism it is strongly recommended to thoroughly review of the
probabilistic results in the file jobname_Slab.pdrs. It is not recommended to do this as a standard
procedure, but only in cases where you are sure that the error generated in the analysis loop can be
safely ignored.
The status codes you could see include:
DOWN - The remote host is not available (reason unknown).
DN-TS - The Thin Server is down.
DN-DRF - ANSYS was unable to create the necessary directories on that host.
DN-ANS - ANSYS would not start on that host.
DN-ATH - Authentication failure (the batch child product was unable to authenticate with the parent
product).

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

37

Probabilistic Design
UP - The Thin Server is up.
RUN - The ANSYS simulation is running on that host.
If you want to check diagnostics for a status code in parallel processing mode, select Diagnostics in
the Server Process Control dialog. The diagnostics system shows you the status of the slave machine(s),
and details to help you address any issues that arise. Suggested actions are presented.

Note
In batch mode, ANSYS will always attempt to start remote processing on all machines listed
in the Jobname.hosts file.

1.3.9.3.4. Initiate Parallel Run

You can now activate the parallel execution:
Command(s): PDEXE
GUI: Main Menu> Prob Design> Run> Exec Parallel> Run Parallel
Several things happen when you initiate a parallel processing run.
1. The networked machines are initialized (ANSYS checks for machine permissions, licenses, and directory
permissions), and any necessary directories are created.
2. The relevant files are copied to the networked machines.
3. ANSYS is launched in server mode on each networked machine.
Simulations are sent to each machine as that machine finishes a previous simulation; faster machines
will naturally process more simulations. If a slave machine (for whatever reason) does not complete a
simulation, that simulation is automatically sent to another machine to be processed.
When the analyses are finished (either completed or stopped manually), then the PDS removes any files
or directories it created and stops any processes it started.

1.3.10. Fit and Use Response Surfaces

After you have executed a probabilistic analysis, you can use the results stored in the result files to fit
response surfaces.
If the probabilistic analysis was based on the Response Surface Method, this step is mandatory. The
random output parameter values generated using the Response Surface Method are meant to be fitted
with a response surface; therefore, the Response Surface Method determines the values of the random
input variables such that fitting of a response surface is most efficient (that is, so that it uses the fewest
sampling points).
If the probabilistic analysis was based on the Monte Carlo Simulation method, this step is optional and
you can go directly to the results postprocessing. If you use Monte Carlo Simulation results to derive
response surfaces, then the sampling points are not as efficiently located for the fitting process, so you
should accommodate by using more sample points.
You should use at least 20% more Monte Carlo Simulation points than what would be required for
a Response Surface Method for the same problem. For a list of the number of sampling points required
for a Response Surface Method please see Probabilistic Design Techniques (p. 51).
38

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

If you cannot determine how many sampling points a Response Surface Method needs (for example,
because there are too many random input parameters), then you should have at least two times
more sampling points than the number of coefficients of the response surfaces.

1.3.10.1. About Response Surface Sets

The results generated by the response surface fitting procedure as well as the results generated by
applying the approximation equation (Monte Carlo Simulation) are combined in a response surface set.
Each response surface set includes the following information:
A unique name that you provide. This name is used to identify the response surface set during
probabilistic postprocessing.
The name of the solution set you used for the fitting procedure (containing the data points).
The number and the names of the random output parameters for which a response surface has been
evaluated. If you have many random output parameters you might not be interested in fitting a response surface for every one, but only for those that are most important.
For each random output parameter that was fitted with a response surface, the response surface set
includes information about the regression model that was used to fit the response surface (linear,
quadratic, or quadratic with cross-terms), as well as the terms and coefficients that were derived as
result of the regression analysis.
The Monte Carlo Simulation samples created using the response surface equations.
There is a one-to-one relationship between solution sets and response surface sets. For each solution
set containing sample points you can have only one response surface set containing the response surfaces
fitting these sample points. The reverse is also true, that each response surface set can only contain the
response surfaces that are based on the sample points of one solution set.

1.3.10.2. Fitting a Response Surface

To fit a response surface you must specify a name for the response surface set where you want the
results to be stored, the solution set you want to use, one random output parameter, the regression
model, an output transformation technique (if any), and whether to filter terms.
The regression model describes which regression terms are used for the approximation function of the
response surface. In the ANSYS PDS, the following regression models are implemented:
Linear approximation function
Quadratic approximation function without cross-terms
Quadratic approximation function including cross-terms
While you can use all terms included in the regression model, the ANSYS PDS also offers an option that
automatically filters out insignificant terms. This technique is called the forward-stepwise regression
analysis. For example, where the Young's modulus E and the thermal expansion coefficient are random
input variables, a full quadratic regression model reads:

mreht

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

39

Probabilistic Design
A full regression model uses the available sampling points to determine values for all regression coefficients c0 to c3. Of course the values for c0 to c2 will be zero or very close to zero; taking more coefficients
into account than really necessary reduces the degrees of freedom of the algebraic equation to be
solved to evaluate the coefficients. This in turn reduces the accuracy of the coefficients that are important
for the regression fit. The forward-stepwise regression analysis takes this into account and automatically
eliminates terms that are not needed.
The ANSYS PDS offers a variety of transformation functions that can be used to make the random response parameter to be more appropriately described by a quadratic function after the transformation
has been applied. These transformation functions can be found in Transformation of Random Output
Parameter Values for Regression Fitting in the Mechanical APDL Theory Reference.
i
B

is the
Here, yi is the value of a random output parameter obtained in the i-th sampling loop and
corresponding transformed value. The physical nature of the problem should indicate which transformation to use; for example, lifetime parameters (such as the number of cycles until low cycle fatigue occurs)
are usually transformed with a logarithmic transformation. If you do not have this kind of information,
then you should start with the Box-Cox transformation. The PDS automatically searches for an optimum
value for the Box-Cox parameter within the interval (-2,2). As guidelines:
If is close to -1.0 then the data is best transformed by a reciprocal transformation, which is a power
transformation with an exponent of -1.0.
If is close to zero then the data is best transformed by a logarithmic transformation.
If is close to 0.5 then use the square root transformation.
If is close to 1.0, then no transformation should be used.
If is not close to any of these specific values then the Box-Cox transformation is appropriate.
To fit a response surface:
Command(s): RSFIT
GUI: Main Menu> Prob Design> Response Surf> Fit Resp Surf

1.3.10.3. Plotting a Response Surface

Whether a response surface is a good representation of the sampling point that it is supposed to fit
can be best illustrated by plotting the response surface. The ANSYS PDS plots the sampling points as
symbols and the response surface as a contour plot so you can visually compare them. However, you
can only plot one random output parameter as a function of two random input variables at a time.
To plot a response surface:
Command(s): RSPLOT
GUI: Main Menu> Prob Design> Response Surf> Plt Resp Surf

1.3.10.4. Printing a Response Surface

After using a regression analysis to fit a response surface, ANSYS automatically prints all necessary results
in the output window:
The transformation function that has been used or has been determined automatically (in case of
Box-Cox transformation)

40

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

Regression terms
Regression coefficients
Goodness-of-fit measures
The goodness-of-fit measures provide a means to verify the quality of the response surface and
whether it is a good representation of the underlying data (in other words, the sample points).
You can request a print out of this data at any time.
Command(s): RSPRNT
GUI: Main Menu> Prob Design> Response Surf> Prn Resp Surf

1.3.10.5. Generating Monte Carlo Simulation Samples on the Response Surfaces

After you have generated a response surface set that includes one or more response surfaces for one
or more random output parameters then you also need to perform Monte Carlo Simulations using these
response surfaces to generate probabilistic results. This is where the PDS generates sampling values for
the random input variables in the same way it did for the simulation looping performed using your
analysis file. But instead of using the random input variable values in the analysis file and running
through the analysis file, it uses the approximation function derived for the response surfaces to calculate
approximated response values. The process of calculating an explicitly known equation is much faster
than running through the analysis file and performing a finite element analysis, so you can run a large
number of simulation loops in a relatively short time. Usually, several thousand simulation loops are
performed if you utilize the response surfaces.
After you have generated the Monte Carlo Simulation loops on the response surfaces, you can begin
probabilistic postprocessing and review the probabilistic results the same way as you would for Monte
Carlo Simulations. However, there is one difference for postprocessing between Monte Carlo results
and Monte Carlo results derived from response surface approximations. For Monte Carlo simulation
results, the accuracy of the results is determined by the number of simulation loops that are performed.
The PDS can visualize the accuracy of Monte Carlo results by means of confidence limits or confidence
bounds. For Monte Carlo results derived from response surface approximations, the confidence bounds
are suppressed. This is necessary because the accuracy is not determined by the number of simulation
loops (as mentioned above, you typically perform a large number of these) but by the goodness-of-fit
or the response surface model itself. With increasing numbers of simulation loops the confidence bounds
tend to merge with the result curve you are plotting (the width of the confidence band shrinks to zero).
This could lead you to conclude that the results are very, very accurate. However, the underlying response
surface approximation could have been completely inadequate (for example, using a linear approximation
function for a highly nonlinear problem).
Command(s): RSSIMS
GUI: Main Menu> Prob Design> Response Surf> RS Simulation

1.3.11. Review Results Data

After probabilistic design looping is complete, you can review the results sets in a variety of ways using
the commands described in this section. These commands can be applied to the results from any
probabilistic design method or tool.
Statistics
Sample History
Histogram
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

41

Probabilistic Design
Cumulative Distribution Function
Probabilities
Inverse Probabilities
Trends
Scatter Plot
Sensitivities
Correlation Matrix
Report
Print HTML Report

1.3.11.1. Viewing Statistics

To postprocess one particular design variable, select this option. The statistics options are described
below.

Plot Sample History

Use the PDSHIS command to request a sample history plot.
Command(s): PDSHIS
GUI: Main Menu> Prob Design> Prob Results> Statistics> Sampl History
You must specify the results set you want to use, the design variable you want to review, the plot type
to use, and the confidence level.

Plot Histogram
Use the PDHIST command to request a histogram plot of a design variable.
Command(s): PDHIST
GUI: Main Menu> Prob Design> Prob Results> Statistics> Histogram
You must specify the results set you want to use, the design variable you want to review, the number
of classes/points to use, and the type of histogram.

CumulativeDF
Use the PDCDF command to request a histogram plot of a design variable.
Command(s): PDCDF
GUI: Main Menu> Prob Design> Prob Results> Statistics> CumulativeDF
You must specify the results set you want to use, the design variable you want to review, and the
confidence level.
The confidence level is a probability expressing the confidence that the values for the cumulative distribution function are in fact between the confidence bounds. The larger the confidence level, the wider
the confidence bounds. Plotting of the confidence bounds only makes sense for the postprocessing of
Monte Carlo simulation results. Here, the confidence bounds represent the accuracy of the results and
with increasing sample size the width of the confidence bounds gets smaller for the same confidence
level. For response surface methods the number of simulations done on the response surface is usually

42

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Probabilistic Design

very large. Therefore, the accuracy of the results is determined by the goodness of the response surface
fit and not by the confidence level.

Probabilities
Use the PDPROB command to request the value of a design variable at a specific point on the cumulative distribution curve.
Command(s): PDPROB
GUI: Main Menu> Prob Design> Prob Results> Statistics> Probabilities
You must specify the results set you want to use, the design variable you want to review, the relation
(greater than, less than), the limit value, and the confidence level.

Inverse Probabilities
Use the PDPINV command to request the value of a design variable at a specific point on the cumulative
distribution curve.
Command(s): PDPINV
GUI: Main Menu> Prob Design> Prob Results> Statistics> Inverse Prob
You must specify the results set you want to use, the design variable you want to review, the relation
(greater than, less than), the limit value, and the confidence level.

1.3.11.2. Viewing Trends

To postprocess one particular design variable as it relates to another, select this option. The trend options
are described below.

Scatter Plot
Use the PDSCAT command to request a scatter plot showing the correlation between two design
variables.
Command(s): PDSCAT
GUI: Main Menu> Prob Design> Prob Results> Trends> Scatter Plot
You must select the results set that you want to use, the design variables that you want to review, the
type of trendline curve to use (and if plotted, the polynomial order), and the maximum number of point
to include in the scatter plot.

Sensitivities
Use the PDSENS command to request the sensitivities of an output parameter to the input variables.
Command(s): PDSENS
GUI: Main Menu> Prob Design> Prob Results> Trends> Sensitivities
You must specify the results set and output parameter you want to use, the type of chart to plot, the
type of correlation coefficient, and the sensitivity level.

Correlation Matrix
Use the PDCMAT command to calculate the correlation coefficient matrix.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

43

Probabilistic Design
Command(s): PDCMAT
GUI: Main Menu> Prob Design> Prob Results> Statistics> Probabilities
You must specify the results set you want to use, which type of design variables you are looking at, the
specific design variable names, the type of correlation, the significance level, and whether you want to
see the probabilities with the correlation coefficients.

1.3.11.3. Creating Reports

To visualize and summarize all results of a probabilistic analysis, select this option. Details on specifying
a report are described below.

Report Options
Use the PDROPT command to request an HTML report.
Command(s): PDROPT
GUI: Main Menu> Prob Design> Prob Results> Report> Report Options
You must specify which statistics and trends to show in the report and in what form you want to see
them. See the PDROPT command for details, and see the other probabilistic results options for further
details.

Generate Report
Use the PDWRITE command to request the sensitivities of an output parameter to the input variables.
Command(s): PDWRITE
GUI: Main Menu> Prob Design> Prob Results> Report> Generate Report
You must enter a name for the report file, your first and last name, and whether links should be generated
between your report and the analysis file, each analysis loop, and the response surface output parameter
details (if the response surface method was used).

1.4. Guidelines for Selecting Probabilistic Design Variables

This section presents useful guidelines for defining your probabilistic design variables.

1.4.1. Choosing and Defining Random Input Variables

Here are a few tips you can use to determine which random input variable in your finite element
model follows which distribution function and what the parameters of the distribution function are.
First, you should know to
Specify a reasonable range of values for each random input variable.

44

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Guidelines for Selecting Probabilistic Design Variables

Set reasonable limits on the variability for each RV.

Note
The values and hints given below are simply guidelines; none are absolute. Always
verify this information with an expert in your organization and adapt it as needed to
suit your analysis.

1.4.1.1. Random Input Variables for Monte Carlo Simulations

The number of simulation loops that are required for a Monte Carlo Simulation does not depend on
the number of random input variables. The required number of simulation loops only depends on the
amount of the scatter of the output parameters and the type of results you expect from the analysis.
In a Monte Carlo Simulation, it is a good practice to include all of the random input variables you can
think of even if you are not sure about their influence on the random output parameters. Exclude only
those random input variables where you are very certain that they have no influence. The probabilistic
design system will then automatically tell you which random input variables have turned out to be
significant and which one are not. The number of simulations that are necessary in a Monte Carlo analysis to provide that kind of information is usually about 50 to 200. However, the more simulation loops
you perform, the more accurate the results will be.

1.4.1.2. Random Input Variables for Response Surface Analyses

The number of simulation loops that are required for a Response Surface analysis depends on the
number of random input variables. Therefore, you want to select the most important input variable(s),
the ones you know have a significant impact on the random output parameters. If you are unsure which
random input variables are important, it is usually a good idea to include all of the random variables
you can think of and then perform a Monte Carlo Simulation. After you learn which random input
variables are important and therefore should be included in your Response Surface Analysis, you can
eliminate those that are unnecessary.

1.4.1.3. Choosing a Distribution for a Random Variable

The type and source of the data you have determines which distribution functions can be used or are
best suited to your needs.

1.4.1.3.1. Measured Data

If you have measured data then you first have to know how reliable that data is. Data scatter is not just
an inherent physical effect, but also includes inaccuracy in the measurement itself. You must consider
that the person taking the measurement might have applied a "tuning" to the data. For example, if the
data measured represents a load, the person measuring the load may have rounded the measurement
values; this means that the data you receive are not truly the measured values. Depending on the
amount of this "tuning," this could provide a deterministic bias in the data that you need to address
separately. If possible, you should discuss any bias that might have been built into the data with the
person who provided that data to you.
If you are confident about the quality of the data, then how to proceed depends on how much data
you have. In a single production field, the amount of data is typically sparse. If you have only few data
then it is reasonable to use it only to evaluate a rough figure for the mean value and the standard deviation. In these cases, you could model the random input variable as a Gaussian distribution if the
physical effect you model has no lower and upper limit, or use the data and estimate the minimum
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

45

Probabilistic Design
and maximum limit for a uniform distribution. In a mass production field, you probably have a lot of
data, in which case you could use a commercial statistical package that will allow you to actually fit a
statistical distribution function that best describes the scatter of the data.

1.4.1.3.2. Mean Values, Standard Deviation, Exceedence Values

The mean value and the standard deviation are most commonly used to describe the scatter of data.
Frequently, information about a physical quantity is given in the form that its value is; for example,
"1005.5". Often, but not always, this form means that the value "100" is the mean value and "5.5" is
the standard deviation. To specify data in this form implies a Gaussian distribution, but you must verify
this (a mean value and standard deviation can be provided for any collection of data regardless of the
true distribution type). If you have more information (for example, you know that the data must be
lognormal distributed), then the PDS allows you to use the mean value and standard deviation for a
definition of a lognormal distribution.
Sometimes the scatter of data is also specified by a mean value and an exceedence confidence limit.
The yield strength of a material is sometimes given in this way; for example, a 99% exceedence limit
based on a 95% confidence level is provided. This means that derived from the measured data we can
be sure by 95% that in 99% of all cases the property values will exceed the specified limit and only in
1% of all cases they will drop below the specified limit. The supplier of this information is using mean
value, the standard deviation, and the number of samples of the measured data to derive this kind of
information. If the scatter of the data is provided in this way, the best way to pursue this further is to
ask for more details from the data supplier. Because the given exceedence limit is based on the measured
data and its statistical assessment, the supplier might be able to provide you with the details that were
used.
If the data supplier does not give you any further information, then you could consider assuming that
the number of measured samples was large. If the given exceedence limit is denoted with x1 - /2 and
the given mean value is denoted with x then the standard deviation can be derived from the equation:

PG

where the values for the coefficient C are:

Exceedence Probability

99.5%

2.5758

99.0%

2.3263

97.5%

1.9600

95.0%

1.6449

90.0%

1.2816

1.4.1.3.3. No Data
In situations where no information is available, there is never just one right answer. Below are hints
about which physical quantities are usually described in terms of which distribution functions. This
might help you with the particular physical quantity you have in mind. Also below is a list of which
distribution functions are usually used for which kind of phenomena. Keep in mind that you might need
to choose from multiple options.

46

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Guidelines for Selecting Probabilistic Design Variables

Geometric Tolerances
If you are designing a prototype, you could assume that the actual dimensions of the manufactured
parts would be somewhere within the manufacturing tolerances. In this case it is reasonable to use
a uniform distribution, where the tolerance bounds provide the lower and upper limits of the distribution function.
Sometimes the manufacturing process generates a skewed distribution; for example, one half of the
tolerance band is more likely to be hit than the other half. This is often the case if missing half of the
tolerance band means that rework is necessary, while falling outside the tolerance band on the other
side would lead to the part being scrapped. In this case a Beta distribution is more appropriate.
Often a Gaussian distribution is used. The fact that the normal distribution has no bounds (it spans
minus infinity to infinity), is theoretically a severe violation of the fact that geometrical extensions
are described by finite positive numbers only. However, in practice this is irrelevant if the standard
deviation is very small compared to the value of the geometric extension, as is typically true for
geometric tolerances.

Material Data
Very often the scatter of material data is described by a Gaussian distribution.
In some cases the material strength of a part is governed by the "weakest-link-theory". The "weakestlink-theory" assumes that the entire part would fail whenever its weakest spot would fail. for material
properties where the "weakest-link" assumptions are valid, then the Weibull distribution might be
applicable.
For some cases, it is acceptable to use the scatter information from a similar material type. Let's assume
that you know that a material type very similar to the one you are using has a certain material
property with a Gaussian distribution and a standard deviation of 5% around the measured mean
value; then let's assume that for the material type you are using, you only know its mean value. In
this case, you could consider using a Gaussian distribution with a standard deviation of 5% around
the given mean value.
For temperature-dependent materials it is prudent to describe the randomness by separating the
temperature dependency from the scatter effect. In this case you need the mean values of your material property as a function of temperature in the same way that you need this information to perform
a deterministic analysis. If M(T) denotes an arbitrary temperature dependent material property then
the following approaches are commonly used:
Multiplication equation:
M(T)rand = Crand

(T)

Additive equation:
M(T)rand =

(T) + Mrand

Linear equation:
M(T)rand = Crand

(T) + Mrand

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

47

Probabilistic Design
Here, (T) denotes the mean value of the material property as a function of temperature. In
the "multiplication equation" the mean value function is scaled with a coefficient Crand and this
coefficient is a random variable describing the scatter of the material property. In the "additive
equation" a random variable Mrand is added on top of the mean value function (T). The
"linear equation" combines both approaches and here both Crand and Mrand are random variables.
However, you should take into account that in general for the "linear equation" approach Crand
and Mrand are, correlated.
Deciding which of these approaches is most suitable to describing the scatter of the temperature dependent material property requires that you have some raw data about this material property. Only by
reviewing the raw data and plotting it versus temperature you can tell which approach is the better
one.

Load Data
For loads, you usually only have a nominal or average value. You could ask the person who provided
the nominal value the following questions: If we have 1000 components that are operated under real
life conditions, what would the lowest load value be that only one of these 1000 components is
subjected to and all others have a higher load? What would the most likely load value be, i.e. the
value that most of these 1000 components have (or are very close to)? What would the highest load
value be that only one of the 1000 components is subjected to and all others have a lower load? To
be safe you should ask these questions not only of the person who provided the nominal value, but
also to one or more experts who are familiar with how your products are operated under real-life
conditions. From all the answers you get, you can then consolidate what the minimum, the most
likely, and the maximum value probably is. As verification you can compare this picture with the
nominal value that you would use for a deterministic analysis. If the nominal value does not have a
conservative bias to it then it should be close to the most likely value. If the nominal value includes
a conservative assumption (is biased), then its value is probably close to the maximum value. Finally,
you can use a triangular distribution using the minimum, most likely, and maximum values obtained.
If the load parameter is generated by a computer program then the more accurate procedure is to
consider a probabilistic analysis using this computer program as the solver mechanism. Use a probabilistic design technique on that computer program to assess what the scatter of the output parameters are, and apply that data as input to a subsequent analysis. In other words, first run a probabilistic analysis to generate an output range, and then use that output range as input for a subsequent
probabilistic analysis.
Here, you have to distinguish if the program that generates the loads is ANSYS itself or your
own in-house program. If you have used ANSYS to generate the loads (for example, FLOTRAN
analysis calculating fluid loads on a structure or a thermal analysis calculating the thermal loads
of a structure) then we highly recommend that you include these load calculation steps in the
analysis file (and therefore in the probabilistic analysis). In this case you also need to model the
input parameters of these load calculation steps as random input variables. If you have used
your own in-house program to generate the loads, you can still integrate the load calculation
program in the analysis file (see the /SYS command for details), but you must have an interface
between that program and ANSYS that allows the programs to communicate with each other
and thus automatically transfer data.
You also have to distinguish if the load values are random fields or single random variables. If the load
is different from node to node (element to element) then it is most appropriate to include the program
calculating the load in the analysis file. If the load is described by one or very few constant values then

48

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Guidelines for Selecting Probabilistic Design Variables

you can also consider performing a probabilistic analysis with the program calculating these load values.
Again you need to provide an interface to transfer input data to this program and get output data (the
loads) back to ANSYS. If there is more than just one single load value generated by the program then
you should also check for potential correlations.

1.4.1.4. Distribution Functions

Beta Distribution
The Beta distribution is very useful for random variables that are bounded at both sides. If linear operations are performed on random variables that are all subjected to a uniform distribution then the results
can usually be described by a Beta distribution. An example is if you are dealing with tolerances and
assemblies, where the components are assembled and the individual tolerances of the components
follow a uniform distribution. In this case the overall tolerances of the assembly are a function of adding
or subtracting the geometrical extension of the individual components (a linear operation). Hence, the
overall tolerances of the assembly can be described by a Beta distribution. Also, as previously mentioned,
the Beta distribution can be useful for describing the scatter of individual geometrical extensions of
components as well. The uniform distribution is a special case of the Beta distribution.

Exponential Distribution
The exponential distribution is useful in cases where there is a physical reason that the probability
density function is strictly decreasing as the random input variable value increases. The distribution is
mostly used to describe time-related effects; for example, it describes the time between independent
events occurring at a constant rate. It is therefore very popular in the area of systems reliability and
lifetime-related systems reliability, and it can be used for the life distribution of non-redundant systems.
Typically, it is used if the lifetime is not subjected to wear-out and the failure rate is constant with time.
Wear-out is usually a dominant life-limiting factor for mechanical components, which would preclude
the use of the exponential distribution for mechanical parts. However in cases where preventive maintenance exchanges parts before wear-out can occur, then the exponential distribution is still useful to
describe the distribution of the time until exchanging the part is necessary.

Gamma Distribution
The Gamma distribution is again a more time-related distribution function. For example it describes
the distribution of the time required for exactly k events to occur under the assumption that the events
take place at a constant rate. It is also used to describe the time to failure for a system with standby
components.

Gaussian (Normal) Distribution

The Gaussian or normal distribution is a very fundamental and commonly used distribution for statistical matters. It is typically used to describe the scatter of the measurement data of many physical phenomena. Strictly speaking, every random variable follows a normal distribution if it is generated by a
linear combination of a very large number of other random effects, regardless which distribution these
random effects originally follow. The Gaussian distribution is also valid if the random variable is a linear
combination of two or more other effects if those effects also follow a Gaussian distribution.

Lognormal Distribution
The lognormal distribution is a basic and commonly used distribution. It is typically used to describe
the scatter of the measurement data of physical phenomena, where the logarithm of the data would
follow a normal distribution. The lognormal distribution is very suitable for phenomena that arise from
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

49

Probabilistic Design
the multiplication of a large number of error effects. It is also correct to use the lognormal distribution
for a random variable that is the result of multiplying two or more random effects (if the effects that
get multiplied are also lognormally distributed). It is often used for lifetime distributions; for example,
the scatter of the strain amplitude of a cyclic loading that a material can endure until low-cycle-fatigue
occurs is very often described by a lognormal distribution.

Uniform Distribution
The uniform distribution is a very fundamental distribution for cases where no other information apart
from a lower and an upper limit exists. It is very useful to describe geometric tolerances. It can also be
used in cases where there is no evidence that any value of the random variable is more likely than any
other within a certain interval. In this sense it can be used for cases where "lack of engineering knowledge" plays a role.

Triangular Distribution
The triangular distribution is most helpful to model a random variable when actual data is not available.
It is very often used to cast the results of expert-opinion into a mathematical form, and is often used
to describe the scatter of load parameters. However, regardless of the physical nature of the random
variable you want to model, you can always ask some experts questions like "What is the one-in-athousand minimum and maximum case for this random variable? and other similar questions. You
should also include an estimate for the random variable value derived from a computer program, as
described earlier. This is also described in more detail above for load parameters in Choosing a Distribution for a Random Variable (p. 45).

Truncated Gaussian Distribution

The truncated Gaussian distribution typically appears where the physical phenomenon follows a Gaussian distribution, but the extreme ends are cut off or are eliminated from the sample population by
quality control measures. As such, it is useful to describe the material properties or geometric tolerances.

Weibull Distribution
In engineering, the Weibull distribution is most often used for strength or strength-related lifetime
parameters, and it is the standard distribution for material strength and lifetime parameters for very
brittle materials (for these very brittle material the "weakest-link-theory" is applicable). For more details
see Choosing a Distribution for a Random Variable (p. 45).

1.4.2. Choosing Random Output Parameters

Output parameters are usually parameters such as length, thickness, diameter, or model coordinates.
The ANSYS PDS does not restrict you with regard the number of random output parameters, provided
that the total number of probabilistic design variables (that is random input variables and random
output parameters together) does not exceed 5000.
ANSYS, Inc. recommends that you include all output parameters that you can think of and that might
be useful to you. The additional computing time required to handle more random output parameters
is marginal when compared to the time required to solve the problem. It is better to define random
output parameters that you might not consider important before you start the analysis. If you forgot
to include a random output parameter that later turns out to be important, you must redo the entire
analysis.

50

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design Techniques

1.5. Probabilistic Design Techniques

Understanding the algorithm used by a computer program is always helpful; this is particularly true in
the case of probabilistic design. This section presents details on the method types and the sampling
options associated with each. See the Mechanical APDL Theory Reference for more information.

1.5.1. Monte Carlo Simulations

The Monte Carlo Simulation method is the most common and traditional method for a probabilistic
analysis. This method lets you simulate how virtual components behave the way they are built. One
simulation loop represents one manufactured component that is subjected to a particular set of loads
and boundary conditions.
For Monte Carlo simulations, you can employ either the Direct Sampling method or the Latin Hypercube
Sampling method.
When you manufacture a component, you can measure its geometry and all of its material properties
(although typically, the latter is not done because this can destroy the component). In the same sense,
if you started operating the component then you could measure the loads it is subjected to. Again, to
actually measure the loads is very often impractical. But the bottom line is that once you have a component in your hand and start using it then all the input parameters have very specific values that you
could actually measure. With the next component you manufacture you can do the same; if you compared
the parameters of that part with the previous part, you would find that they vary slightly. This comparison of one component to the next illustrates the scatter of the input parameters. The Monte Carlo
Simulation techniques mimic this process. With this method you virtually manufacture and operate
components or parts one after the other.
The advantages of the Monte Carlo Simulation method are:
The method is always applicable regardless of the physical effect modeled in a finite element analysis.
It not based on assumptions related to the random output parameters that if satisfied would speed
things up and if violated would invalidate the results of the probabilistic analysis. Assuming the deterministic model is correct and a very large number of simulation loops are performed, then Monte
Carlo techniques always provide correct probabilistic results. Of course, it is not feasible to run an
infinite number of simulation loops; therefore, the only assumption here is that the limited number
of simulation loops is statistically representative and sufficient for the probabilistic results that are
evaluated. This assumption can be verified using the confidence limits, which the PDS also provides.
Because of the reason mentioned above, Monte Carlo Simulations are the only probabilistic methods
suitable for benchmarking and validation purposes.
The individual simulation loops are inherently independent; the individual simulation loops do not
depend on the results of any other simulation loops. This makes Monte Carlo Simulation techniques
ideal candidates for parallel processing.
The Direct Sampling Monte Carlo technique has one drawback: it is not very efficient in terms of required
number of simulation loops.

1.5.1.1. Direct Sampling

Direct Monte Carlo Sampling is the most common and traditional form of a Monte Carlo analysis. It is
popular because it mimics natural processes that everybody can observe or imagine and is therefore
easy to understand. For this method, you simulate how your components behave based on the way
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

51

Probabilistic Design
they are built. One simulation loop represents one component that is subjected to a particular set of
loads and boundary conditions.
The Direct Monte Carlo Sampling technique is not the most efficient technique, but it is still widely
used and accepted, especially for benchmarking and validating probabilistic results. However, benchmarking and validating requires many simulation loops, which is not always feasible. This sampling
method is also inefficient due to the fact that the sampling process has no "memory."
For example, if we have two random input variables X1 and X2 both having a uniform distribution
ranging from 0.0 to 1.0, and we generate 15 samples, we could get a cluster of two (or even more)
sampling points that occur close to each other if we graphed the two variables (see figure below). While
in the space of all random input variables, it can happen that one sample has input values close to
another sample, this does not provide new information and insight into the behavior of a component
in a computer simulation if the same (or almost the same) samples are repeated.
Figure 1.7: Graph of X1 and X2 Showing Two Samples with Close Values

To use Direct Monte Carlo Sampling, do the following

Command(s): PDMETH,MCS,DIR PDDMCS
GUI: Main Menu> Prob Design> Prob Method> Monte Carlo Sims
In this sampling method, you set the number of simulations, whether to stop simulations loops when
certain criteria are met (accuracy for mean values and standard deviations), and the seed value for
randomizing input variable sample data.

1.5.1.2. Latin Hypercube Sampling

The Latin Hypercube Sampling (LHS) technique is a more advanced and efficient form for Monte Carlo
Simulation methods. The only difference between LHS and the Direct Monte Carlo Sampling technique
is that LHS has a sample "memory," meaning it avoids repeating samples that have been evaluated
before (it avoids clustering samples). It also forces the tails of a distribution to participate in the sampling
process. Generally, the Latin Hypercube Sampling technique requires 20% to 40% fewer simulations
loops than the Direct Monte Carlo Simulation technique to deliver the same results with the same accuracy. However, that number is largely problem dependent.

52

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design Techniques

Figure 1.8: Graph of X1 and X2 Showing Good Sample Distribution

To use the Latin Hypercube Sampling technique:

Command(s): PDMETH,MCS,LHS PDLHS
GUI: Main Menu> Prob Design> Prob Method> Monte Carlo Sims
In this sampling method, you set the number of simulations and repetitions, the location in the interval
for the sample, whether the simulations stop when certain criteria are met (accuracy of mean values
and standard deviations), and random number seed for variability in the sample input variable data.

1.5.1.3. User-Defined Sampling

For this method, you provide the file containing the samples.
Command(s): PDMETH,MCS,USER PDUSER
GUI: Main Menu> Prob Design> Prob Method> Monte Carlo Sims
By using this option you have complete control over the sampling data. You are required to give the
file name and path.
If user-specified sampling methods are requested with the PDMETH,MCS,USER command or the
PDMETH,RSM,USER command, then you need to specify which file contains the sample data. The sample
data is a matrix, where the number of columns is equal to the number of defined random variables and
the number of lines is equal to the number of simulation loops requested. This data must be contained
in an ASCII file and the content must obey the following notations and format requirements:
Column separators allowed: blank spaces, commas, semicolons, and tabs.
Multiple blank spaces and multiple tabs placed directly one after the other are allowed and are considered as one single column separator.
Multiple commas or semicolons placed directly one after the other are not allowed; for example, two
commas with no data between them (just blanks) are read as an empty column, which leads to an
error message.
The first line of the file must contain a solution label. No additional data is allowed on the first line,
and if found, will lead to an error message. An error message is also issued if the solution label is
missing.
The solution label is just a placeholder. For consistency, you should use the same solution label you
specify in the PDEXE command, but if they are different, you will always use the solution label specified
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

53

Probabilistic Design
in the PDEXE command for postprocessing. The PDS system does not check if the solution label in
the user-specified file and the one given in the PDEXE command match.
The second line of the file must contain the headers of the data columns. The first three column
headers must be ITER CYCL and LOOP respectively; then subsequent columns should contain
,
,
,
the names of the random variables. You must use one of the allowed separators as described above
between the column headers. No additional data is allowed on this line, and if found, will prompt an
error message. An error message is also issued if any of the required column headers are missing.
The random variable names in your file must match the names of the defined random variables. The
variable names that you specify must consist of all uppercase characters (regardless of the case used
in the defined variable names).
Columns four to n can be in arbitrary order. The ANSYS PDS tool determines the order for the random
variable data based on the order of the random variable names in the second line.
The third and subsequent lines must contain the order number for the iteration, cycle, and simulation
loop, then the random variable values for that loop. The iteration, cycle, and simulation loop numbers
must be in the first, second, and third columns respectively, followed by the random variable values.
The iteration and cycle numbers are used by the ANSYS PDS (internally) and for a user-defined
sampling method you will typically use a value of "1" for all simulation loops. The loop number is an
ascending number from 1 to the total number of loops requested. Additional data is not allowed,
and if found, will lead to an error message. An error message is also issued if any of the data columns
are missing.
You must be sure that the order of the random variable values in each line is identical to the order
of the random variable names in the second line.
The user-specified sampling file must contain a minimum of one data line for the random variable
values.
When the PDUSER command is issued, the PDS checks that the specified file exists, then verifies it for
completeness and consistency. Consistency is checked according to the possible minimum and maximum
boundaries of the distribution type of the individual random variables. An error message is issued if a
random variable value is found in the file that is below the minimum boundary or above the maximum
boundary of the distribution of that random variable. This also means that any value will be accepted
for a random variable if its distribution has no minimum or maximum boundary; for example, this is
the case for the Gaussian (normal) distribution. Apart from this check, it is your responsibility to provide
values for the random variables that are consistent with their random distribution.

Note
It is your responsibility to ensure that parameters defined as random input variables are actually input parameters for the analysis defined with the PDANL command. Likewise, you
must ensure that parameters defined as random output parameter are in fact results generated
in the analysis file.

Example
An excerpt of the content of a user-specified sampling file is given below. This example is based on
three random variables named X1, X2, and X3. A total of 100 simulation loops are requested.
USERSAMP
ITER CYCL

54

LOOP

X1

X2

X3

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design Techniques

1
1
1
..
..
1
1
1

1
1
1
..
..
1
1
1

1
2
3
..
..
98
99
100

1.619379209e+000
2.237676559e-001
7.931615474e+000
...
...
1.797221666e+000
1.290815540e+001
4.699281922e+000

2.364528435e-001
5.788049712e-001
8.278689033e-001
...
...
3.029471373e-001
9.271606216e-001
6.526505821e-001

1.470789050e+000
1.821263115e+000
2.170793522e+000
...
...
1.877701081e+000
2.091047328e+000
1.901013985e+000

1.5.2. Response Surface Analysis Methods

For response surface analysis, you can choose from three sampling methods: Central Composite Design,
Box-Behnken Matrix, and user-defined.
Response surface methods are based on the fundamental assumption that the influence of the random
input variables on the random output parameters can be approximated by mathematical function.
Hence, Response Surface Methods locate the sample points in the space of random input variables such
that an appropriate approximation function can be found most efficiently; typically, this is a quadratic

ji

is described by

x x
iaj

Ia i

x
i

polynomial. In this case the approximation function

Ia i

where c0 is the coefficient of the constant term, ci, i = 1,...NRV are the coefficients of the linear terms
and cij, i = 1,...NRV and j = i, ...,NRV are the coefficients of the quadratic terms. To evaluate these coefficients a regression analysis is used and the coefficients are usually evaluated such that the sum of
squared differences between the true simulation results and the values of the approximation function
is minimized.
Hence, a response surface analysis consists of two steps:
1. Performing the simulation loops to calculate the values of the random output parameters that correspond to the sample points in the space of random input variables.
2. Performing a regression analysis to derive the terms and the coefficients of the approximation
function.
The fundamental idea of Response Surface Methods is that once the coefficients of a suitable approximation function are found, then we can directly use the approximation function instead of looping
through the finite element model. To perform a finite element analysis might require minutes to hours
of computation time; in contrast, evaluating a quadratic function requires only a fraction of a second.
Hence, if using the approximation function, we can afford to evaluate the approximated response
parameter thousands of times.
A quadratic polynomial is sufficient in many cases of engineering analysis (for example, the evaluation
of the thermal stress mentioned above). For that evaluation, the Young's modulus and the thermal expansion coefficient both have a linear effect on the thermal stresses, which is taken into account in a
quadratic approximation by the mixed quadratic terms. However, there are cases where a quadratic
approximation is not sufficient; for example, if the finite element results are used to calculate the lifetime
of a component. For this evaluation, the lifetime typically shows an exponential behavior with respect
to the input parameters; thus the lifetime results cannot be directly or sufficiently described by a
quadratic polynomial. But often, if you apply a logarithmic transformation to the lifetime results, then
these transformed values can be approximated by a quadratic polynomial. The ANSYS PDS offers a
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

55

Probabilistic Design
variety of transformation functions that you can apply to the response parameters, and the logarithmic
transformation function is one of them.
Assuming the approximation function is suitable for your problem, the advantages of the Response
Surface Method are:
It often requires fewer simulation loops than the Monte Carlo Simulation method.
It can evaluate very low probability levels. This is something the Monte Carlo Simulation method
cannot do unless you perform a great number of simulation loops.
The goodness-of-fit parameters tell you how good the approximation function is (in other words,
how accurate the approximation function is that describes your "true" response parameter values).
The goodness-of-fit parameters can warn you if the approximation function is insufficient.
The individual simulation loops are inherently independent (the individual simulation loops do not
depend on the results of any other simulation loops). This makes Response Surface Method an ideal
candidate for parallel processing.
The disadvantages of the Response Surface Method are:
The number of required simulation loops depends on the number of random input variables. If you
have a very large number of random input variables (hundreds or even thousands), then a probabilistic analysis using Response Surface Methods would be impractical.
This method is not usually suitable for cases where a random output parameter is a non-smooth
function of the random input variables. For example, a non-smooth behavior is given if you observe
a sudden jump of the output parameter value even if the values for the random input variables vary
only slightly. This typically occurs if you have instability in your model (such as bulking). The same
might happen if the model includes a sharp nonlinearity such as a linear-elastic-ideal-plastic material
behavior. Or, if you are analyzing a contact problem, where only a slight variation in your random
input variables can change the contact situation from contact to non-contact or vice versa, then you
also might have problems using the Response Surface Method.

Note
To use Response Surface Methods, the random output parameters must be smooth and
continuous functions of the involved random input variables. Do not use Response
Surface Methods if this condition is not satisfied.

1.5.2.1. Central Composite Design Sampling

A central composite design consists of a central point, the N axis point plus 2N-f factorial points located
at the corners of an N-dimensional hypercube. Here, N is the number of random input variables and f
is the fraction of the factorial part of the central composite design. A fraction f = 0 is a called a full
factorial design, f = 1 gives a half-factorial design, and so on. The PDS gradually increases the fraction
f as you increase the number of random input variables. This keeps the number of simulation loops
reasonable. The fraction f is automatically evaluated such that a resolution V design is always maintained.
A resolution V design is a design where none of the second order terms of the approximation function
are confined with each other. This ensures a reasonable accuracy for the evaluation of the coefficients
of the second order terms.
The locations of the sampling points for a problem with three random input variables is illustrated below.

56

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Probabilistic Design Techniques

Figure 1.9: Locations of Sampling Points for Problem with Three Input Variables for CCD

The number of sample points (simulation loops) required for a central composite design as a function
of the number of random input variables is given in the table below:
Number of random input variables

Number of coefficients in
a quadratic function (with
cross-terms)

Factorial number f

Number sample
points (simulation
loops)

N/A

N/A

10

15

15

25

21

27

28

45

36

79

45

81

55

147

10

66

149

11

78

151

12

91

281

13

105

283

14

120

285

15

136

287

16

153

289

17

171

291

18

190

549

19

210

10

551

20

231

11

553

To use the Response Surface Method with a Central Composite Design, do the following:
Command(s): PDMETH,RSM,CCD PDDOEL,Name,CCD,...
GUI: Main Menu> Prob Design> Prob Method> Response Surface
PDDOEL allows you to specify design of experiment options.
See the Mechanical APDL Theory Reference for more details.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

57

Probabilistic Design

1.5.2.2. Box-Behnken Matrix Sampling

A Box-Behnken Design consists of a central point plus the midpoints of each edge of an N-dimensional
hypercube.
The location of the sampling points for a problem with three random input variables is illustrated below.
Figure 1.10: Location of Sampling Points for Problem with Three Input Variables for BBM

The number of sample points (simulation loops) required for a Box-Behnken design as a function of the
number of random input variables is given in the table below:
Number of random input variables

Number of coefficients in
a quadratic function (with
cross-terms)

Number sample
points (simulation
loops)
N/A

N/A

10

12

15

25

21

41

28

49

36

57

45

65

55

121

10

66

161

11

78

177

12

91

193

To use Response Surface Analysis with the Box-Behnken design option, do the following:
Command(s): PDMETH,RSM,BBM PDDOEL,Name,BBM,...
GUI: Main Menu> Prob Design> Prob Method> Response Surface
PDDOEL allows you to specify design of experiment options.
See the Mechanical APDL Theory Reference for more details.

1.5.2.3. User-Defined Sampling

For this method, you provide the file containing the samples.

58

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Postprocessing Probabilistic Analysis Results

Command(s): PDMETH,RSM,USER PDUSER
GUI: Main Menu> Prob Design> Prob Method> Response Surface
By using this option, you have complete control over the sampling data. You are required to give the
file name and path.

1.6. Postprocessing Probabilistic Analysis Results

There are two groups of postprocessing functions in the probabilistic design system: statistical and
trend.
A statistical analysis is an evaluation function performed on a single probabilistic design variable; for
example, a histogram plot of a random output parameter. The ANSYS PDS allows statistical evaluation
of either random output parameters or random input variables.
A trend analysis typically involves two or more probabilistic design variables; for example, a scatter plot
of one probabilistic design variable versus another.
Probabilistic postprocessing is generally based on result sets. A result set is either a solution set generated
by a probabilistic analysis run (PDEXE) or a response surface set (RSFIT). With a few exceptions, you
can perform the same type of probabilistic postprocessing operations regardless of whether you postprocess the results of a solution set or the results of a response surface set.

1.6.1. Statistical Postprocessing

Statistical postprocessing allows you several options for reviewing your results.

1.6.1.1. Sample History

The most fundamental form of postprocessing is directly reviewing the simulation loop results as a
function for the number of simulation loops. Here, you can review the simulation values (for either response surface or Monte Carlo methods), or for Monte Carlo Simulations only, the mean, minimum, or
maximum values, or the standard deviations.
It is most helpful to review the mean values and standard deviation history for Monte Carlo Simulation
results if you want to decide if the number of simulation loops was sufficient. If the number of simulation
loops was sufficient, the mean values and standard deviations for all random output parameters should
have converged. Convergence is achieved if the curve shown in the respective plots approaches a
plateau. If the curve shown in the diagram still has a significant and visible trend with increasing number
of simulation loops then you should perform more simulation loops. In addition, the confidence bounds
plotted for the requested history curves can be interpreted as the accuracy of the requested curve. With
more simulation loops, the width of the confidence bounds is reduced.
Typically, postprocessing Monte Carlo results based on response surfaces is based on a very large
number of simulation loops using the response surface equations. Therefore, the accuracy of the results
is not a function of the number of simulation loops, but rather the goodness-of-fit measures of the individual response surfaces. As one example, suppose the goodness-of-fit measures indicate a very poor
quality of a response surface fit, but the mean value and standard deviation history indicate that the
results have converged (because of the large number of loops) and the accuracy is excellent (again
because confidence bounds shrink with increasing number of loops). This could lead you to an incorrect
conclusion. This is why the PDS only allows you to visualize the sample value history directly, but not
the mean value history and so on.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

59

Probabilistic Design
To review the simulation loop results:
Command(s): PDSHIS
GUI: Main Menu> Prob Design> Prob Results> Statistics> Sampl History
You need to select the results set label and the design variable, select a plot type, and set the confidence
level.

1.6.1.2. Histogram
A histogram plot is most commonly used to visualize the scatter of a probabilistic design variable. A
histogram is derived by dividing the range between the minimum value and the maximum value into
intervals of equal size. Then the PDS determines how many samples fall within each interval, that is,
how many "hits" landed in the intervals.
Most likely, you will use histograms to visualize the scatter of your random output parameters. The
ANSYS PDS also allows you to plot histograms of your random input variables so you can double check
that the sampling process generated the samples according to the distribution function you specified.
For random input variables, the PDS not only plots the histogram bars, but also a curve for values derived
from the distribution function you specified. Visualizing histograms of the random input variables is
another way to make sure that enough simulation loops have been performed. If the number of simulation loops is sufficient, the histogram bars will:
Be close to the curve that is derived from the distribution function
Be "smooth" (without large steps)
Not have major gaps
A major gap is given if you have no hits in an interval where neighboring intervals have many hits.
However, if the probability density function is flattening out at the far ends of a distribution (for example,
the exponential distribution flattens out for large values of the random input variable) then there might
logically be gaps. Hits are counted only as positive integer numbers and as these numbers gradually
get smaller, a zero hit can happen in an interval.
To plot histograms:
Command(s): PDHIST
GUI: Main Menu> Prob Design> Prob Results> Statistics> Histogram

1.6.1.3. Cumulative Distribution Function

The cumulative distribution function is a primary review tool if you want to assess the reliability or the
failure probability of your component or product. Reliability is defined as the probability that no failure
occurs. Hence, in a mathematical sense reliability and failure probability are two sides of the same coin
and numerically they complement each other (are additive to 1.0). The cumulative distribution function
value at any given point expresses the probability that the respective parameter value will remain below
that point. The figure below shows the cumulative distribution function of the random property X:

60

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Postprocessing Probabilistic Analysis Results

Figure 1.11: Cumulative Distribution Function of X

Ax@p

7HHI
7HW
7HV
7HU
7HT
7HS
7HR
7HQ
7HP

ix

FFF

7HI
7H

The value of the cumulative distribution function at the location x0 is the probability that the values of
X stay below x0. Whether this probability represents the failure probability or the reliability of your
component depends on how you define failure; for example, if you design a component such that a
certain deflection should not exceed a certain admissible limit then a failure event occurs if the critical
deflection exceeds this limit. Thus for this example, the cumulative distribution function is interpreted
as the reliability curve of the component. On the other hand, if you design a component such that the
eigenfrequencies are beyond a certain admissible limit then a failure event occurs if an eigenfrequency
drops below this limit. Thus for this example, the cumulative distribution function is interpreted as the
failure probability curve of the component.
The cumulative distribution function also lets you visualize what the reliability or failure probability
would be if you chose to change the admissible limits of your design.
Often you are interested in visualizing low probabilities and you want to assess the more extreme ends
of the distribution curve. In this case plotting the cumulative distribution function in one of the following
ways is more appropriate:
As a Gauss plot (also called a "normal plot"). If the probabilistic design variable follows a Gaussian
distribution then the cumulative distribution function is displayed as a straight line in this type of
plot.
As a lognormal plot. If the probabilistic design variable follows a lognormal distribution then the cumulative distribution function is displayed as a straight line in this type of plot
As a Weibull plot. If the probabilistic design variable follows a Weibull distribution then the cumulative
distribution function is displayed as a straight line in this type of plot.
The advantage of these plots is that the probability axis is scaled in a nonlinear fashion such that the
extreme ends of the distribution function are emphasized and more visible.
To plot the cumulative distribution function:
Command(s): PDCDF
GUI: Main Menu> Prob Design> Statistics> Prob Results> CumulativeDF

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

61

Probabilistic Design

1.6.1.4. Print Probabilities

The PDS offers a function where you can determine the cumulative distribution function at any point
along the axis of the probabilistic design variable, including an interpolation function so you can evaluate the probabilities between sampling points. This feature is most helpful if you want to evaluate the
failure probability or reliability of your component for a very specific and given limit value.
To print probabilities:
Command(s): PDPROB
GUI: Main Menu> Prob Design> Prob Results> Statistics> Probabilities

1.6.1.5. Print Inverse Probabilities

The PDS offers a function where you can probe the cumulative distribution function by specifying a
certain probability level; the PDS tells you at which value of the probabilistic design variable this probability will occur. This is helpful if you want to evaluate what limit you should specify to not exceed a
certain failure probability, or to specifically achieve a certain reliability for your component.
To print inverse probabilities:
Command(s): PDPINV
GUI: Main Menu> Prob Design> Prob Results> Statistics> Inverse Prob

1.6.2. Trend Postprocessing

Trend postprocessing allows you several options for reviewing your results.

1.6.2.1. Sensitivities
Probabilistic sensitivities are important in allowing you to improve your design toward a more reliable
and better quality product, or to save money while maintaining the reliability or quality of your product.
You can request a sensitivity plot for any random output parameter in your model.
There is a difference between probabilistic sensitivities and deterministic sensitivities. Deterministic
sensitivities are mostly only local gradient information. For example, to evaluate deterministic sensitivities you can vary each input parameters by 10% (one at a time) while keeping all other input parameters constant, then see how the output parameters react to these variations. As illustrated in the
figure below, an output parameter would be considered very sensitive with respect to a certain input
parameter if you observe a large change of the output parameter value.

62

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Postprocessing Probabilistic Analysis Results

Figure 1.12: Sensitivities
steep gradient = higher sensitivity

Y1

X
Y

Y2

lower gradient = lower sensitivity

X

These purely deterministic considerations have various disadvantages that are taken into consideration
for probabilistic sensitivities, namely:
A deterministic variation of an input parameter that is used to determine the gradient usually does
not take the physical range of variability into account. An input parameter varied by 10% is not
meaningful for the analysis if 10% is too much or too little compared with the actual range of
physical variability and randomness. In a probabilistic approach the physical range of variability is
inherently considered because of the distribution functions for input parameters. Probabilistic sensitivities measure how much the range of scatter of an output parameter is influenced by the scatter
of the random input variables. Hence, two effects have an influence on probabilistic sensitivities: the
slope of the gradient, plus the width of the scatter range of the random input variables. This is illustrated in the figures below. If a random input variable has a certain given range of scatter, then the
scatter of the corresponding random output parameter is larger, and the larger the slope of the
output parameter curve is (first illustration). But remember that an output parameter with a moderate
slope can have a significant scatter if the random input variables have a wider range of scatter (second
illustration).
Figure 1.13: Range of Scatter
range of

scatter X

Y1

range of
scatter X

range of
scatter Y

scatter

range of

Y2

scatter Y

Y2
range
Y2
X

Gradient information is local information only. It does not take into account that the output parameter
may react more or less with respect to variation of input parameters at other locations in the input
parameter space. However, the probabilistic approach not only takes the slope at a particular location
into account, but also all the values the random output parameter can have within the space of the
random input variables.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

63

Probabilistic Design
Deterministic sensitivities are typically evaluated using a finite-differencing scheme (varying one
parameter at a time while keeping all others fixed). This neglects the effect of interactions between
input parameters. An interaction between input parameters exists if the variation of a certain parameter has a greater or lesser effect if at the same time one or more other input parameters change
their values as well. In some cases interactions play an important or even dominant role. This is the
case if an input parameter is not significant on its own but only in connection with at least one other
input parameter. Generally, interactions play an important role in 10% to 15% of typical engineering
analysis cases (this figure is problem dependent). If interactions are important, then a deterministic
sensitivity analysis can give you completely incorrect results. However, in a probabilistic approach,
the results are always based on Monte Carlo simulations, either directly performed using you analysis
model or using response surface equations. Inherently, Monte Carlo simulations always vary all random
input variables at the same time; thus if interactions exist then they will always be correctly reflected
in the probabilistic sensitivities.
To display sensitivities, the PDS first groups the random input variables into two groups: those having
a significant influence on a particular random output parameter and those that are rather insignificant,
based on a statistical significance test. This tests the hypothesis that the sensitivity of a particular random
input variable is identical to zero and then calculates the probability that this hypothesis is true. If the
probability exceeds a certain significance level (determining that the hypothesis is likely to be true),
then the sensitivity of that random input variable is negligible. The PDS will plot only the sensitivities
of the random input variables that are found to be significant. However, insignificant sensitivities are
printed in the output window. You can also review the significance probabilities used by the hypothesis
test to decide which group a particular random input variable belonged to.
The PDS allows you to visualize sensitivities either as a bar chart, a pie chart, or both. Sensitivities are
ranked so the random input variable having the highest sensitivity appears first.
In a bar chart the most important random input variable (with the highest sensitivity) appears in the
leftmost position and the others follow to the right in the order of their importance. A bar chart describes
the sensitivities in an absolute fashion (taking the signs into account); a positive sensitivity indicates
that increasing the value of the random input variable increases the value of the random output parameter for which the sensitivities are plotted. Likewise, a negative sensitivity indicates that increasing
the random input variable value reduces the random output parameter value. In a pie chart, sensitivities
are relative to each other.
In a pie chart the most important random input variable (with the highest sensitivity) will appear first
after the 12 o'clock position, and the others follow in clockwise direction in the order of their importance.
Using a sensitivity plot, you can answer the following important questions.

How can I make the component more reliable or improve its quality?
If the results for the reliability or failure probability of the component do not reach the expected levels,
or if the scatter of an output parameter is too wide and therefore not robust enough for a quality
product, then you should make changes to the important input variables first. Modifying an input
variable that is insignificant would be waste of time.
Of course you are not in control of all random input parameters. A typical example where you have
very limited means of control are material properties. For example, if it turns out that the environmental
temperature (outdoor) is the most important input parameter then there is probably nothing you can
do. However, even if you find out that the reliability or quality of your product is driven by parameters
that you cannot control, this has importance it is likely that you have a fundamental flaw in your
product design! You should watch for influential parameters like these.

64

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Postprocessing Probabilistic Analysis Results

If the input variable you want to tackle is a geometry-related parameter or a geometric tolerance, then
improving the reliability and quality of your product means that it might be necessary to change to a
more accurate manufacturing process or use a more accurate manufacturing machine. If it is a material
property, then there is might be nothing you can do about it. However, if you only had a few measurements for a material property and consequently used only a rough guess about its scatter and the
material property turns out to be an important driver of product reliability and quality, then it makes
sense to collect more raw data. In this way the results of a probabilistic analysis can help you spend
your money where it makes the most sense in areas that affect the reliability and quality of your
products the most.

How can I save money without sacrificing the reliability or the quality of the product?
If the results for the reliability or failure probability of the component are acceptable or if the scatter
of an output parameter is small and therefore robust enough for a quality product then there is usually
the question of how to save money without reducing the reliability or quality. In this case, you should
first make changes to the input variables that turned out to be insignificant, because they do not effect
the reliability or quality of your product. If it is the geometrical properties or tolerances that are insignificant, you can consider applying a less expensive manufacturing process. If a material property turns
out to be insignificant, then this is not typically a good way to save money, because you are usually
not in control of individual material properties. However, the loads or boundary conditions can be a
potential for saving money, but in which sense this can be exploited is highly problem dependent.
Command(s): PDSENS
GUI: Main Menu> Prob Design> Prob Results> Trends> Sensitivities

1.6.2.2. Scatter Plots

While the sensitivities point indicate which probabilistic design parameters you need to modify to have
an impact on the reliability or failure probability, scatter plots give you a better understanding of how
and how far you should modify the input variables. Improving the reliability and quality of a product
typically means that the scatter of the relevant random output parameters must be reduced.
The PDS allows you to request a scatter plot of any probabilistic design variable versus any other one,
so you can visualize the relationship between two design variables (input variables or output parameters).
This allows you to verify that the sample points really show the pattern of correlation that you specified
(if you did so). Typically, random output parameters are correlated as because they are generated by
the same set of random input variables. To support the process of improving the reliability or quality
of your product, a scatter plot showing a random output parameter as a function of the most important
random input variable can be very helpful.
When you display a scatter plot, the PDS plots the sampling points and a trendline. For this trendline,
the PDS uses a polynomial function and lets you chose the order of the polynomial function. If you plot
a random output parameter as a function of a random input variable, then this trendline expresses how
much of the scatter on the random output parameter (Y-axis) is controlled by the random input variable
(X-axis). The deviations of the sample points from the trendline are caused and controlled by all the
other random input variables. If you want to reduce the scatter of the random output parameter to
improve reliability and quality, you have two options:
Reduce the width of the scatter of the most important random input variable(s) (that you have control
over).
Shift the range of the scatter of the most important random input variable(s) (that you have control over).

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

65

Probabilistic Design
The effect of reducing and shifting the scatter of a random input variable is illustrated in the figures
below. "Input range before" denotes the scatter range of the random input variable before the reduction
or shifting, and "input range after" illustrates how the scatter range of the random input variable has
been modified. In both cases, the trendline tells how much the scatter of the output parameter is affected
and in which way if the range of scatter of the random input variable is modified.

Input range before

Random Input Variable

Output range before

Input range after

Output range
after

Random Output Parameter

Input range
after

Output
range
after

Output range
before

Random Output Parameter

Figure 1.14: Effects of Reducing and Shifting Range of Scatter

Input range
before

Random Input Variable

It depends on your particular problem if either reducing or shifting the range of scatter of a random
input variable is preferable. In general, reducing the range of scatter of a random input variable leads
to higher costs. A reduction of the scatter range requires a more accurate process in manufacturing or
operating the product the more accurate, the more expensive it is. This might lead you to conclude
that shifting the scatter range is a better idea, because it preserves the width of the scatter (which
means you can still use the manufacturing or operation process that you have). Below are some considerations if you want to do that:
Shifting the scatter range of a random input variable can only lead to a reduction of the scatter of a
random output parameter if the trendline shows a clear nonlinearity. If the trendline indicates a linear
trend (if it is a straight line), then shifting the range of the input variables anywhere along this straight
line doesn't make any difference. For this, reducing the scatter range of the random input variable
remains your only option.
It is obvious from the second illustration above that shifting the range of scatter of the random input
variable involves an extrapolation beyond the range where you have data. Extrapolation is always
difficult and even dangerous if done without care. The more sampling points the trendline is based
on the better you can extrapolate. Generally, you should not go more than 30-40% outside of the
range of your data. But the advantage of focusing on the important random input variables is that
a slight and careful modification can make a difference.

Note
ANSYS strongly recommends that you redo the entire probabilistic analysis using the new
and modified random input variables if you have reduced of shifted the scatter range of any
input variables using the procedure and recommendations above. To redo the probabilistic
analysis, save the probabilistic model using the PDSAVE command and clear the current
probabilistic analysis results using the PDCLR,POST command. Then you can modify the
random input variable definitions and redo the probabilistic analysis.

66

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Multiple Probabilistic Design Executions

Command(s): PDSCAT
GUI: Main Menu> Prob Design> Prob Results> Trends> Scatter

1.6.2.3. Correlation Matrix

Probabilistic sensitivities are based on a statistical correlation analysis between the individual probabilistic design variables. The PDS lets you review the correlation data that has been used to derive sensitivities and decide if individual sensitivity values are significant or not. This information is collected in
the correlation matrix of the random output parameters versus the random input variables. The PDS
also lets you review the correlations that have been sampled between random input variables, which
is stored in the random input variables correlation matrix. The correlations between random output
parameters are important if you want to use the probabilistic results of your probabilistic analysis as
input for another probabilistic analysis.
To print out a correlation matrix:
Command(s): PDCMAT
GUI: Main Menu> Prob Design> Prob Results> Trends> Correl Matrix

1.6.3. Generating an HTML Report

The probabilistic design system automatically generates an HTML report for all probabilistic analyses
that you performed with your probabilistic model. The report explains the problem you analyzed, which
probabilistic methods were used, and the results. The report uses pictures as well as verbal description
and explanations to document all this. The report is a predefined HTML document that you can modify
and add to using an HTML editor. However, you can also influence the contents of the report to some
extent.
To generate an HTML report for probabilistic design:
Command(s): PDROPT, PDWRITE
GUI: Main Menu> Prob Design> Prob Results> Report> Report Options
Main Menu> Prob Design> Prob Results> Report> Generate Report
Use the PDROPT command to specify the items you want to include in your report, then use the
PDWRITE command to generate it.

1.7. Multiple Probabilistic Design Executions

There are various reasons why you may wish to perform more than one probabilistic design execution.
For example, your initial probabilistic design run may not contain enough samples or loops to find the
desired probability limits. Or, you may start by using one probabilistic design analysis method, then try
another method to compare or validate results. The knowledge you gain from the first few loops may
prompt you to change the method you use or other analysis settings.
If you run a probabilistic analysis using a new probabilistic methods make sure that you also provide
a new solution label in the PDEXE command. If you want to add data to an existing solution set then
make sure that you use the same probabilistic method as the one that has been used in the existing
solution set. In this case please also note, that if the existing solution set has already a response surface
set associated to it, then all response surface sets will be deleted and need to redifined using the RSFIT
command.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

67

Probabilistic Design
If you perform all executions within the same session (or within the same batch input stream), the
procedure is very straightforward. After an execution, redefine the probabilistic design input as desired
and initiate the next execution. To initiate the execution:
Command(s): PDEXE
GUI: Main Menu> Prob Design> Run> Exec Serial> Run Serial
Main Menu> Prob Design> Run> Exec Parallel> Run Parallel
For more information on these execution options, see Execute Probabilistic Analysis Simulation
Loops (p. 25). If you left the program after performing an analysis and would like to continue your
probabilistic design analysis at some later time, you can do a save and restart as described next.

1.7.1. Saving the Probabilistic Design Database

Because the probabilistic database is completely separate from the database, if you save the database
using the command SAVE it does not save the probabilistic design database. An exception is when you
leave the session and select Save Everything in the EXIT from ANSYS menu, in which case the PDS
database is also saved. If you already saved the probabilistic database to a specific file, then Save
Everything causes the current PDS database to overwrite the data in that file; otherwise, the data is
written to jobname.pds in your current working directory.
The probabilistic design data is automatically saved at the end of each probabilistic design analysis
loop. You can save the probabilistic design data at any time by using the PDSAVE command.
To save a probabilistic design analysis:
Command(s): PDSAVE
GUI: Main Menu> Prob Design> Prob Database> Save

Caution
Currently, you must use the Save and Resume commands on the Prob Design menu to save
your work if interrupted or if you simply want to resume the project later.

1.7.2. Restarting a Probabilistic Design Analysis

The probabilistic database is maintained completely separate from the database. This means that if you
resume the database then the probabilistic design database is not also resumed.
To restart a probabilistic design analysis, resume the probabilistic design database file (Jobname.PDS):
Command(s): PDRESU
GUI: Main Menu> Prob Design> Prob Database> Resume
Once the data is read in, you can respecify probabilistic design type, controls, etc., and initiate looping.
(The analysis file corresponding to the resumed database must be available in order to perform probabilistic design.) Note that the previous results are not loaded into the database until a probabilistic
postprocessing command is issued.

68

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Probabilistic Design Analysis

See the /PDS, PDRESU, PDSAVE, and PDEXE command descriptions for more details.

Note
In addition to probabilistic design data, the jobname is saved to the probabilistic design
database file (Jobname.PDS). Therefore, when a probabilistic design data file is resumed
(PDRESU), the jobname saved in that file will overwrite the current jobname (/FILNAME).
You can use the PDRESU command (Main Menu> Prob Design> Prob Database> Resume) in an interactive session to resume probabilistic design data that was created through a batch run, thus allowing
convenient interactive viewing of batch probabilistic design results.

1.7.3. Clearing the Probabilistic Design Database

The probabilistic database is completely separate from the database. This usually means that saving
and resuming the database has no effect on saving and resuming the PDS database. However, if you
issue the /CLEAR command to clear the database, then the PDS database is also cleared. For this reason
you should never use the /CLEAR command as part of your PDS analysis file. Clearing the PDS database
has no effect on the database.
For clearing the probabilistic database you have two options. You can clear the entire PDS database or
just the results and solution portion. The latter maintains the probabilistic model definition (the random
input variable definitions, correlations, and random output parameter definitions). Clearing only the
solution and results part of the PDS database is helpful if you want to modify the probabilistic model
(change RVs, set correlations between RVs or RPs, etc.). After you have performed a probabilistic analysis,
the PDS will not allow you to change any of the probabilistic model details. This ensures consistency
of the probabilistic model with the results.
To clear the probabilistic design database:
Command(s): PDCLR
GUI: Main Menu> Prob Design> Prob Database> Clear & Reset
Because the database is not affected by the PDCLR command, it may also be necessary to clear the
database if the resumed probabilistic design problem is totally independent of the previous one. See
the /CLEAR command for details.

1.8. Example Probabilistic Design Analysis

In the following example, you will perform a probabilistic design analysis of a simple indeterminate
three-bar truss. This analysis uses the Direct Monte Carlo method.

1.8.1. Problem Description

You will analyze the effect of variation in the three-bar truss cross-section that is subjected to a force
with both vertical and horizontal components.

1.8.2. Problem Specifications

The loading for this example is a force of 20000 lbs. located at the vertex of the three truss members.
The force acts at a 45 angle from the vertical direction.
The following material property is used for this analysis:
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

69

Probabilistic Design
Young's modulus (E) = 30E6 psi
The following geometric properties are used for this analysis. These properties are the initial cross-sectional areas for each truss member:
ARE1 = 5.0
ARE2 = 5.0
ARE3 = 5.0

1.8.2.1. Problem Sketch

Figure 1.15: The Simple Indeterminate Three-Bar Truss for the Example Problem

1.8.3. Using a Batch File for the Analysis

You can perform the example probabilistic design analysis of this three-bar truss using the commands
shown below.
The analysis file is created for use during the probabilistic analysis. It is a parametric model of the
problem geometry, materials, and loads. Within the analysis file, input variables are initialized and output
variables are retrieved. If you prefer, you can perform the second pass of the example analysis using
the GUI method rather than commands. See Using the GUI for the PDS Analysis (p. 72) for details.
The following input listing sets up the analysis file for the three-bar truss.
/com
/com, Create an analysis file to be used during looping
/com
*create,pds3bar,pdan
*SET,ARE1,5.00
!INITIALIZE CROSS SECTIONAL AREAS
*SET,ARE2,5.00
*SET,ARE3,5.00
/PREP7
/title, PROBABILISTIC ANALYSIS OF A SIMPLE INDETERMINATE THREE-BAR TRUSS
ET,1,1
EX,1,30E6
R,1,ARE1
R,2,ARE2
R,3,ARE3
N,1
N,2,10
N,3,20
N,4,10,-10,,-45
! ROTATE TIPNODE 45
REAL,1
E,1,4
REAL,2
E,2,4

70

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Probabilistic Design Analysis

REAL,3
E,3,4
D,1,ALL,,,3
F,4,FX,20000
FINISH
/SOLU
SOLVE
FINISH
/POST1
SET,1
ETABLE,VOLU,VOLU
ETABLE,AXST,LS,1
*GET,SIG1,ELEM,1,ETAB,AXST
*GET,SIG2,ELEM,2,ETAB,AXST
*GET,SIG3,ELEM,3,ETAB,AXST
SSUM
*GET,TVOL,SSUM,,ITEM,VOLU
FINI
*end

!
!
!
!
!

STORE MEMBER VOLUMES

STORE MEMBER AXIAL STRESSES
SIG1 IS DEFINED TO BE AXIAL STRESS IN ELEMENT 1
SIG2 IS DEFINED TO BE AXIAL STRESS IN ELEMENT 2
SIG3 IS DEFINED TO BE AXIAL STRESS IN ELEMENT 3

After the analysis file has been created, you can proceed with the probabilistic design analysis. You can
do this though commands or though the GUI. If you prefer using commands, the following input sets
up the probabilistic analysis for the three-bar truss example.
/inp,pds3bar,pdan
/com
/com, Enter PDS and specify the analysis file
/com
/PDS
! enter probabilistic design system
pdanl,pds3bar,pdan
/com
/com, Declare random input variables
/com
PDVAR,ARE1,GAUS,5,0.5
! define area1 with Gaussian distribution
! having mean of 5 and std. dev of 0.5
PDVAR,ARE2,tria,10,11,12
! define area2 with triangular distribution
! having low bound of 10, most likely point of 11
! and high bound of 12
PDVAR,ARE3,unif,5,6
! define area3 with uniform distribution
! with low bound of 5 and high bound of 6
/com
/com, Specify any correlations between the random variables
/com
PDCOR,ARE1,ARE3,0.25
! define a correlation coef of 0.25 between ARE1 and ARE3
/com
/com, Declare random output variables
/com
PDVAR,SIG1,resp
! define SIG1 a response parameter
PDVAR,SIG2,resp
! define SIG2 a response parameter
PDVAR,SIG3,resp
! define SIG3 a response parameter
PDVAR,TVOL,resp
! define TVOL a response parameter
/com
/com, Select the probabilistic design tool or method
/com
PDMETH,MCS,DIR
! specify direct Monte Carlo simulation
PDDMCS,100,NONE,ALL,,,,123457 ! use all 100 samples, initial seed of 123457
/com
/com, Execute the loops required for the probabilistic design analysis
/com
PDEXE,mcs3bar
! run analysis and define solution label 3bar_mcs
/com
/com, Review the results of the probabilistic analysis
/com
PDSENS,MCS3BAR,TVOL,BOTH,RANK,0.025 !Create Sensitivity plot
fini
/exit

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

71

Probabilistic Design

1.8.4. Using the GUI for the PDS Analysis

Because of the parametric definition of some variables in the analysis file, the GUI method is not recommended for analysis file creation and is not presented here. It is acceptable, however, to perform the
probabilistic design analysis of the three-bar truss example via the GUI method. The GUI procedure for
performing the probabilistic design analysis follows. Using a Batch File for the Analysis (p. 70) describes
how to create an analysis file. The GUI procedure for performing the probabilistic design analysis pass
follows.

Step 1: Test Analysis File

To test the analysis file, you clear the database and then read input from the pds3bar.pdan file.
1. Select menu path Utility Menu> File> Clear & Start New. Click OK.
2. In the Verify dialog box, click Yes.
3. Change the jobname. Select menu path Utility Menu> File> Change Jobname. The Change Jobname
dialog box appears.
4. Change the jobname to pds3bar.pdan and click OK.
5. Select menu path Utility Menu> File> Read Input from. In the Files list, click on pds3bar.pdan. Then
click OK. You see a replay of the entire analysis. Click on Close when the Solution is done! message
appears.
In the next several steps of this problem, you explore the effects of variation in the parameters.

Step 2: Enter the Probabilistic Design Module and Identify Analysis File
First, enter the optimizer and identify the analysis file.
1. Select menu path Main Menu> Prob Design> Analysis File> Assign. The Assign Analysis File dialog
box appears.
2. In the Files list, click once on pds3bar.pdan and then clickOK.

Step 3: Identify the Probabilistic Design Variables

1. Select menu path Main Menu>Prob Design>Prob Definitions>Random Input. The Random Input
Variables dialog box appears.
2. Click Add. The Define a Random Variable dialog box appears.
3. In the list of parameter names, click ARE1. Select GAUSS in the distribution type list. Click OK.
4. Type 5 in the MEAN VALUE field and 0.5 in the STANDARD DEVIATION field. Click OK.
5. Click Add. The Define a Random Variable dialog box appears.
6. In the list of parameter names, click ARE2. Select TRIANGULAR in the distribution type list. Click OK.
7. Type 10 in the LOWER BOUNDARY field, 11 in the MOST LIKELY VALUE field, and 12 in the UPPER
BOUNDARY field. Click OK.

72

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Probabilistic Design Analysis

8. Click Add. The Define a Random Variable dialog box appears.
9. In the list of parameter names, click ARE3. Select UNIFORM in the distribution type list. Click OK.
10. Type 5 in the LOW BOUND field, and 6 in the HIGH BOUND field. Click OK.
11. Click on Close to close the Define a Random Variable dialog box.
12. Select menu path Main Menu> Prob Design> Prob Definitions> Correlation. The Add/Edit or Delete
Correlation dialog box appears.
13. Select ARE1 and ARE2 from the list of variable names. Click OK. Type 0.25 in the Correlation Coeff
field.
14. Select menu path Main Menu> Prob Design> Prob Definitions> Random Output. The Random Output
Parameters dialog box appears.
15. Click Add. The Define a Random Output Parameter dialog box appears. In the list of parameter names,
click on SIG1. Click OK.
16. Click Add. The Define a Random Output Parameter dialog box appears. In the list of parameter names,
click on SIG2. Click OK.
17. Click Add. The Define a Random Output Parameter dialog box appears. In the list of parameter names,
click on SIG3. Click OK.
18. Click on Close to close the Random Output Parameters dialog box.

Step 4: Run the Probabilistic Design Analysis

This step involves specifying the probabilistic design method, executing the run, and saving the data.
1. Select menu path Main Menu> Prob Design> Prob Method> Monte Carlo Sims. The Monte Carlo
Simulation dialog box appears.
2. Select Direct Sampling from Sampling Methods.
3. Type 100 in the Number of Simulations field. Select Use 123457 from the Random Seed Option. Click
OK.
4. Select menu path Main Menu> Prob Design> Run> Exec Serial> Run Serial. Type mcs3bar in the
Solution Set Label field. Click OK.
5. The Run Monte Carlo Simulations dialog box appears. Click OK.
6. Select menu path Main Menu> Prob Design> Prob Database> Save. Type mcs3bar in the selection
field. Click OK.
When the data is saved, you are ready to review the results.

Step 5: Review the Probabilistic Results

In this step, you visualize the probabilistic results.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

73

Probabilistic Design
1. Select menu path Main Menu> Prob Design> Prob Results> Trends> Sensitivities. The Sensitivity of
a Response Parameter dialog box appears.
2. Select MCS3BAR in the Select Results Set field. Select SIG1 in the Select Response Parameter field.
Click OK.
3. A Rank-Order Correlation Sensitivity bar and pie chart is shown in the Graphics window.

Step 6: Exit
1. Click on Quit in the Toolbar.
2. Select an option to save, then click OK.

74

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 2: Variational Technology

Variational Technology for Improved Solver Performance
The Variational Technology implementation speeds up the solution itself and has been applied to two
distinct types of mathematical problems: Nonlinear solutions for structural and thermal analyses, and
harmonic analysis. These capabilities are referred to as VT Accelerator. VT Accelerator provides a 2X-5X
speedup for the initial solutions depending on the hardware, model, and type of analysis. VT Accelerator
makes re-solves 3X to 10X faster for parameter changes, allowing for effective simulation driven parametric studies of nonlinear and transient analyses in a cost-effective manner. You can make the following
types of changes to the model before a VT Accelerator re-solve:
Modify, add, or remove loads (constraints may not be changed, although their value may be modified)
Change materials and material properties
Change section data and real constants
Change geometry, although the mesh connectivity must remain the same (i.e., the mesh must be
morphed)
VT Accelerator for Nonlinear Solution Speed-Up VT Accelerator for nonlinear solutions speeds up
the solution of applicable nonlinear analysis types by reducing the total number of iterations. Examples
include:
Nonlinear structural static or transient analyses not involving contact or plasticity
Nonlinear thermal static or transient analyses
VT Accelerator for Harmonic Analysis The harmonic sweep feature of VT Accelerator provides a
high-performance solution for forced-frequency simulations in high-frequency electromagnetic problems
and structural analysis.
For a structural harmonic analysis, VT Accelerator provides a harmonic analysis over a range of userdefined frequencies. The structural material may have frequency dependent elasticity or damping.
For a high-frequency electromagnetic harmonic analysis, VT Accelerator provides a high-frequency
analysis over a range of user-defined frequencies. The module computes S-parameters over the entire
frequency range. In practice, the harmonic sweep feature of VT Accelerator completes one normal ANSYS
run at the mid-frequency of the specified frequency range. It then performs accurate approximations
of the results across the frequency range (in user-specified steps). In addition to controlling the steps
and the frequency range, you can also specify the accuracy of the approximations. Two harmonic sweep
solution methods are available: Variational Technology and Variational Technology Perfect Absorber.
The Variational Technology Perfect Absorber method provides about a 20% faster solution but it is
slightly less accurate.
The following Variational Technology topics are available:
2.1. Harmonic Sweep Using VT Accelerator
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

75

Variational Technology

2.1. Harmonic Sweep Using VT Accelerator

The harmonic sweep feature of VT Accelerator provides a high-performance solution for forced-frequency
simulations in high-frequency electromagnetic problems and structural analysis when the material
properties are frequency dependent and no modal responses can be used.
The harmonic sweep feature of VT Accelerator completes one normal ANSYS run at the mid-frequency
of the specified frequency range.
The harmonic sweep interpolates the stiffness, damping, mass, and loading on the whole range to give
an approximation of the results across the frequency range. The harmonic sweep evaluates these values
for different frequencies (between 2 and 6) across the frequency range, and then it computes a polynomial approximation of the values.
The harmonic sweep performs accurate approximations of the results across the frequency range in
user-specified steps.

Note
If the loading, material properties, or real constants defined with a data table (BF ,TBFIELD
command, or COMBIN14) are frequency-dependent and too few data points exist in the
table, the polynomial interpolation of the values will be poor. In case of abrupt variations
for a given frequency f2, ANSYS advises that you split the frequency range [f1 f3] into two
ranges (e.g., [f1 f2] ; [f2 f3] ) and perform separate analyses on each range.
High-frequency electromagnetic problems use ANSYS high-frequency elements HF119 or HF120 only.
The following Variational Technology topics are available for harmonic sweep:
2.1.1. Structural Elements Supporting Frequency-Dependent Properties
2.1.2. Harmonic Sweep for High-Frequency Electromagnetic Problems
2.1.3. Harmonic Sweep for Structural Analysis with Frequency-Dependent Material Properties

2.1.1. Structural Elements Supporting Frequency-Dependent Properties

Frequency-dependent property structural elements can be used with the following elements:

Element

Frequency-Dependent Material Properties

(defined using TB,ELAS and TB,SDAMP commands)

PLANE182

EX, SDAMP

PLANE183

EX, SDAMP

SOLID185

EX, SDAMP

SOLID186

EX, SDAMP

SOLID187

EX, SDAMP

SOLID272

EX, SDAMP

SOLID273

EX, SDAMP

SOLID285

EX, SDAMP

76

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Harmonic Sweep Using VT Accelerator

2.1.2. Harmonic Sweep for High-Frequency Electromagnetic Problems

The harmonic sweep feature of VT Accelerator provides a high-frequency analysis over a range of userdefined frequencies. The module computes S-parameters over the entire frequency range. (For more
information, see the SPSWP command description.)
In practice, the harmonic sweep feature of VT Accelerator completes one normal ANSYS run at the midfrequency of the specified frequency range. It then performs accurate approximations of the results
across the frequency range (in user-specified steps). In addition to controlling the steps and the frequency
range, you can also specify the accuracy of the approximations.
There are two harmonic sweep solution methods available: Variational Technology and Variational
Technology Perfect Absorber. The Variational Technology Perfect Absorber method provides about a
20% faster solution but it is slightly less accurate. To specify the solution method use the SPSWP
command and set SwpOpt to 0 or 2 or the HROPT command and set Method to VT or VTPA.
S-parameters are stored in a Touchstone format file, called jobname.snp, where n is the number of
ports.
To compute a matrix of S-parameters for waveguide ports
1.

Define the port regions (flags) and boundary conditions. Use the
BFA, BFL, or BF commands for interior waveguide.
SF or SFA commands for exterior waveguide ports.

2.

Define the waveguide port type and excitation using the HFPORT command.

3.

Compute the matrix of S-parameters using the SPSWP command.

4.

Display the S-parameters using the PLSYZ command.

To compute a matrix of S-parameters for transmission line ports

1.

Define the transmission line port (flags) and options.

2.

Define a port excitation using BF, BFL, or BFA and a current density source. Offset this excitation from
the transmission line port plane.

3.

Define a voltage path from the central conductor to the ground on the transmission line port plane
using the PATH command. Save the path using the PASAVE command.

4.

Specify transmission line inputs using the HFPORT command.

5.

Compute the matrix of S-parameters using the SPSWP command.

6.

Display the S-parameters using the PLSYZ command.

Note
A large number of frequency steps will produce a very large database file.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

77

Variational Technology

2.1.2.1. Transmission Line Example Problem

/batch,list
/TITLE, Example for SPSWP macro with transmission line ports
/com, Brick structure: 3 x 1.5 x 7 cm^3 with material step
/prep7
! ---------- geometry parametrs -----------ch=0.015
$ cw=0.03
$ cl=0.0375
!
! ---------- define elements and material properties ------et,11,200,7
! temporary element
et,1,120,1
! brick 1st order
et,2,120,1,,,1
! PML elemnts
mp,murx,1,1
mp,perx,1,1
mp,murx,2,1
mp,perx,2,4
! ---------- define geometry and mesh the model ---------numel=6
rect,-cw/2,cw/2,-ch/2,ch/2
type,11
esize,,cw/numel
amesh,1
type,1
! create brick mesh by extrusion
esize,,8
mat,1
asel,s,loc,z,0.
vext,all,,,0,0,-0.8*cl
mat,2
asel,s,loc,z,0.
vext,all,,,0,0,0.8*cl
type,2
! create PML elements
esize,,4
mat,1
asel,s,loc,z,-0.8*cl
vext,all,,,0,0,-0.4*cl
mat,2
asel,s,loc,z,0.8*cl
vext,all,,,0,0,0.4*cl
asel,s,loc,z,0.
aclear,all
alls
! ---------- assign PEC boundary conditions ---------nsel,s,loc,y,-ch/2
nsel,a,loc,y,ch/2
d,all,ax,0.
! ---------- assign PEC BC on PML exterior ----------nsel,s,loc,z,-cl*1.2
nsel,a,loc,z,cl*1.2
d,all,ax,0.
nsel,all
! ---------- create transmission line port #1 ------------nsel,s,loc,z,-0.4*cl
! select port nodes
bf,all,port,1
! flag port nodes
hfport,1,tline,,v1,sext,188.36
! specify port options (wave impedance Z=188.36)
nsel,s,loc,z,-0.6*cl
bf,all,js,0,1,0,-1
! define excitation
! ---------- create transmission line port #2 ------------nsel,s,loc,z,0.4*cl
! select port nodes
bf,all,port,2
! flag port nodes
hfport,2,tline,,v2,sext,94.18
! specify port options (wave impedance Z=94.18)
nsel,s,loc,z,0.6*cl
! define excitation
bf,all,js,0,1,0,-2
alls
! ---------- define paths for transmission line ports -----x1=-cw/2 $x2=cw/2 $z1=-0.4*cl $z2=0.4*cl $y1=-ch/2 $y2=ch/2
path,v1,2
! voltage path for port 1
ppath,1,,0,y2,z1
ppath,2,,0,y1,z1
path,v2,2
! voltage path for port 2
ppath,1,,0,y2,z2
ppath,2,,0,y1,z2

78

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Harmonic Sweep Using VT Accelerator

pasave,all
fini
/solu
alls
spswp,1.e9,5.e9,1.e9,0
fini

! save paths in file.path

! run frequency sweep using Variational Technology

Results in Touchstone Format

!

ANSYS S-parameter Data for 2 Ports (Transmission Line).

# GHz S MA R 188.36 94.18.

! Freq
|S11|
<|S11|
1.0000
0.322 142.788
2.0000
0.339 109.259
3.0000
0.329
69.074
4.0000
0.334
36.978
5.0000
0.345
0.070

|S21|
<|S21|
0.945 -52.364
0.935 -107.633
0.932 197.701
0.921 145.595
0.905
95.776

|S12|
<|S12|
0.947 -53.791
0.948 -106.050
0.959 200.509
0.966 146.452
0.964
93.750

|S22|
<|S22|
0.326 -69.987
0.337 -141.146
0.323 141.352
0.332
66.997
0.368
7.834

2.1.2.2. Waveguide Example Problem

Figure 2.1: Element Plot for Waveguide Example

/batch,list
/title, Dielectric Post in a Rectangular Waveguide
/com, Waveguide Dimension: 22.86 x 10.16 mm^2 (Cutoff Frequency: 6.56 GHz)
/com, Dielectric Post: 12 x 10.16 x 6 mm^3 at the center of waveguide
/com,
epsr = 8.2
/com, Frequency Range: 8 - 12 GHz for TE10 mode
/com, PML Parameter: 5 Layers with -50 dB
/com, Mesh Size: Free Space Wavelength/15 at 12 GHz
/com, Numerical Model: IMPD Driven Port; PML Output Port
/com, Solution Target: S11 over frequency range using series expansion method
/prep7
ch=10.16e-3
! waveguide height
cw=22.86e-3
! waveguide width
c=12e-3
! post width
d=6e-3
! post length
epsr=8.2
! Dielectric constant
freq=10.e9
! Analysis frequency
fmesh=12e9
! Mesh frequency
lamda=3.e8/fmesh ! wavelength
h1=lamda/8
! mesh parameter
cl=5*d
! waveguide length
et,1,HF119,1

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

79

Variational Technology
et,2,HF119,1,,,1 ! PML option
mp,murx,1,1.
mp,perx,1,1.
mp,murx,2,1.
mp,perx,2,epsr ! Dielectric
block,-cw/2,cw/2,-ch/2,ch/2,-cl/2,cl/2
block,-c/2,c/2,-ch/2,ch/2,-d/2,d/2
vsbv,1,2,,delete,keep
block,-cw/2,cw/2,-ch/2,ch/2,cl/2,cl/2+lamda/5
vglue,all
esize,h1
type,1
mat,2
vmesh,2
mat,1
vmesh,4
type,2
vmesh,1

! Dielectric region
! Air region
! PML region

! Tangential E is zero on all side walls

nsel,s,loc,y,-ch/2
nsel,a,loc,y,ch/2
nsel,a,loc,x,-cw/2
nsel,a,loc,x,cw/2
nsel,a,loc,z,cl/2+lamda/5
d,all,ax,0.
nsel,s,loc,z,-cl/2 ! locate rectangular port
sf,all,port,1
hfport,1,rect,,te10,impd,cw,ch,1
alls
fini
/solu
alls
spswp,8.e9,12.e9,0.1e9,0 ! run frequency sweep using Variational Technology option
FINISH
/post1
plsyz,file,s1p,s,db,1,1 ! plot magnitude of S11 (dB)
plsyz,file,s1p,s,ang,1,1 ! plot phase angle (Deg)
finish

Figure 2.2: Graph of Phase Angle

80

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Harmonic Sweep Using VT Accelerator

Figure 2.3: Graph of Magnitude

2.1.3. Harmonic Sweep for Structural Analysis with Frequency-Dependent

Material Properties
The harmonic sweep feature of the VT Accelerator module allows you to define the material elastic
properties as frequency-dependent and efficiently compute the frequency response over an entire frequency range. For more information, see the documentation for the TB,ELAS and TB,SDAMP commands.
To use this formulation, use the hysteretic damping formulation with the HROPT command.
If you define the damping ratio (TB,SDAMP) as a linear function of the frequency, the damping exhibits
behavior similar to that of viscous damping. See the Mechanical APDL Theory Reference for more information about the hysteretic and structural damping formulations.
Limitations
The frequency-dependent tables (defined by the TB,ELAS and TB,SDAMP commands) define piecewise
linear functions. These are used to define the stiffness and damping matrices, which in turn are fitted with
a polynomial over the entire frequency range to compute their derivatives as a function of frequency;
therefore, if the piecewise linear approximation of the material properties is too coarse, the results will
be poor.

2.1.3.1. Beam Example

Consider a cantilever beam, with a Young's modulus of 20e6 psi for static condictions and 30e6 for
500Hz.
ETYPE=186
LF = 10
UF = 500
N = 50
/PREP7
!*
ET,1,ETYPE
/com,
/com,
/com,
/com,
/com,

! STARTING FREQUENCY
! ENDING FREQUENCY
! NUMBER OF SUBSTEP

* ==============================================
*
*
Frequency dependent material properties
*
* ==============================================

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

81

Variational Technology

TB,ELASTIC,1,,2

! Elastic data table

TBFIELD , FREQ,1 ! First frequency value

TBDATA,1,20e6,0.3 ! E and m
TBFIELD ,FREQ,500 ! Fifth frequency value
TBDATA,1,30e6,0.3
TB,SDAMP,1, ,1

! damping data table

TBFIELD , FREQ,1 ! First frequency value

TBDATA,1, 0.02
! Damping co.
TBFIELD ,FREQ,500
TBDATA,1, 0.01

! Fifth frequency value

MP,DENS,1,.10
BLOCK,0,10,0,2,0,2
LSEL,S,LOC,X,-.5,0.5
LESIZE,ALL,,,2
LSEL,S,LOC,X,9.5,10.5
LESIZE,ALL,,,2
LSEL,S,LOC,X,2,8
LESIZE,ALL,,,5
VMESH,ALL
FINISH
/SOL
LSEL,S,LOC,X,-.5,0.5
DL,all,,all
KSEL,S,LOC,X,8,12
KSEL,R,LOC,Y,-.5,.5
FK,ALL,FY,1000
ASEL,S,LOC,Y,1.8,2.2
SFA,ALL,,PRES,1000,
allsel
FINI
/com, * ==============================================
/com, *
/com, *
VT Harmonic Analysis
/com, *
/com, * ==============================================
/VT
VTCLR,ALL
VTRFIL
VTFREQ,frq,LF,UF,N
VTRSLT,disp,NODE,U,ALL,0.01,ALL
FINISH
/SOLU
ANTY,HARM
HROUT,OFF
! Print complex displacements as amplitude and phase angle
KBC,1
HROPT,VT,,,,HYST
Solve
FINISH
/show,post
/post26
nsol,10,57,u,y,d1
prvar,10

The graph below shows the difference in the Y displacement at the end of the beam, taking constant
material properties for 0Hz, constant properties for 500 Hz, and variable properties. Note that the first
peak is higher on the variable curve than the other two. This is because of the frequency step used to
create the frequency response.

82

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Harmonic Sweep Using VT Accelerator

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

83

84

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 3: Adaptive Meshing

The ANSYS program provides approximate techniques for estimating mesh discretization error for certain
types of analyses. Using this measure of mesh discretization error, the program can then determine if
a particular mesh is fine enough. If not, the program will automatically refine the mesh so that the
measured error will decrease. This process of automatically evaluating mesh discretization error and
refining the mesh, called adaptive meshing, continues through a series of solutions until the measured
error drops below some user-defined value (or until a user-defined limit on the number of solutions is
reached).
The following adaptive meshing topics are available:
3.1. Prerequisites for Adaptive Meshing
3.2. Employing Adaptive Meshing
3.3. Modifying the Adaptive Meshing Process
3.4. Adaptive Meshing Hints and Comments
3.5. Where to Find Examples
For more information about error-value approximation, see The General Postprocessor (POST1) in the
Basic Analysis Guide.

3.1. Prerequisites for Adaptive Meshing

The ANSYS program includes a macro named ADAPT.MAC for performing adaptive meshing. Your
model must meet certain criteria before you can successfully activate the macro. (In some cases, models
not conforming to the criteria can be adaptively meshed using a modified procedure, as discussed below.)
The requirements include the following:
The standard adaptive meshing procedure is valid only for single-solution linear static structural and
linear steady-state thermal analyses.
Use only one material type, as the error calculations are based in part on average nodal stresses, and
would thus often be invalid at the material interfaces. Also, an element's error energy is affected by
its elastic modulus; therefore, even if the stress discontinuity is the same in two adjoining elements,
their error energy will be different if they have different material properties. Also, avoid abrupt changes
in shell thickness, as such discontinuities cause similar stress-averaging problems.
Use meshable solid model entities. (Avoid characteristics that will cause meshing failure.)
Some element- and material-type limitations apply. For more information, see the documentation for
the PRERR command.

3.2. Employing Adaptive Meshing

The general process for running the adaptive meshing macro follows:
1. As in any linear static structural or steady state thermal analysis, enter the preprocessor (/PREP7 command
or menu path Main Menu> Preprocessor) and specify the element type, real constants, and material
properties (according to the prerequisites).
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

85

Adaptive Meshing
2. Model your system using solid modeling procedures, creating meshable areas or volumes describing the
geometry of your system. It is not necessary to specify element sizes, nor do you need to mesh these
areas and volumes; the ADAPT macro initiates meshing automatically. (If you need to mesh your model
with both area and volume elements, create an ADAPTMSH.MAC subroutine.
3. You can either proceed to SOLUTION (/SOLU or menu path Main Menu> Solution) or remain in PREP7
to specify analysis type, analysis options, loads, and load step options. Apply only solid model loads and
inertia loads (accelerations, rotational accelerations, and rotational velocities) in a single load step. (Finite
element loads, coupling, and constraint equations can be introduced through the ADAPTBC.MAC subroutine. Multiple load steps can be introduced through the ADAPTSOL.MAC subroutine.)
4. If in /PREP7, exit the preprocessor (FINISH). (You can invoke the ADAPT macro from either SOLUTION
or the Begin level.)
5. Invoke the adaptive solution. To do so, use one of these methods:
Command(s): ADAPT
GUI: Main Menu> Solution> Solve> Adaptive Mesh
You can use the ADAPT macro in either a thermal or a structural analysis, but you cannot combine
the two disciplines in one adaptive solution. As the adaptive meshing iterations proceed, element
sizes are adjusted (within the limits set by FACMN and FACMX) to decrease and increase the elemental
error energies until the error in energy norm matches the target value (or until the specified maximum
number of solutions has been used).
After you have invoked the adaptive solution, this macro controls all program operations until the
solution is completed. The ADAPT macro will define element sizes, generate the mesh, solve, evaluate errors, and iterate as necessary till the target value of error in energy norm is met. All these
steps are performed automatically, with no further input required from you.
6. When adaptive meshing has converged, the program automatically turns element shape checking on
(SHPP,ON). It then returns to the SOLUTION phase or to the Begin level, depending on which phase you
were in when you invoked ADAPT. You can then enter /POST1 and postprocess as desired, using
standard techniques.

3.3. Modifying the Adaptive Meshing Process

This section describes how to modify adaptive meshing by employing selective adaptivity, enhancing
the ADAPT macro's functionality via user subroutines, and customizing the ADAPT macro itself via the
UADAPT macro.

3.3.1. Selective Adaptivity

If you know that mesh discretization error (measured as a percentage) is relatively unimportant in some
regions of your model (for instance, in a region of low, slowly-changing stress), you can speed up your
analysis by excluding such regions from the adaptive meshing operations. Also, you might want to exclude regions near singularities caused by concentrated loads. Selecting logic provides a way of handling
such situations.

86

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Modifying the Adaptive Meshing Process

Figure 3.1: Selective Adaptivity

Selective adaptivity can improve the performance of models having concentrated loads.
If you select a set of keypoints, the ADAPT macro will still include all your keypoints (that is, the ADAPT
macro will modify the mesh at both selected and non-selected keypoints), unless you also set KYKPS
= 1 in the ADAPT command (Main Menu> Solution> Solve> Adaptive Mesh).
If you select a set of areas or volumes, the ADAPT macro will adjust element sizes only in those regions
that are in the selected set. You will have to mesh your entire model in PREP7 before executing ADAPT.

3.3.2. Customizing the ADAPT Macro with User Subroutines

The standard ADAPT macro might not always be applicable to your particular analysis needs. For instance,
you might need to mesh both areas and volumes, which is not possible with the standard macro. For
this and other such situations, you can modify the ADAPT macro to suit your analysis needs. By using
a macro to perform the adaptive meshing task, we have intentionally given you access to the logic involved, and have thereby furnished you with the capability for modifying the technique as you might
desire.
Fortunately, you do not always need to change the coding within the ADAPT macro to customize it.
Three specific portions of the macro can be readily modified by means of user subroutines, which are
separate user files that you can create and that will be called by the ADAPT macro. The three features
that can be modified by user subroutines are:
the meshing command sequence
the boundary condition command sequence,
the solution command sequence
The corresponding user subroutine files must be named ADAPTMSH.MAC, ADAPTBC.MAC, and ADAPTSOL.MAC, respectively.

3.3.2.1. Creating a Custom Meshing Subroutine (ADAPTMSH.MAC)

By default, if your model contains one or more selected volumes, the ADAPT macro will mesh only
volumes (no area meshing will be done). If you have no selected volumes, then the ADAPT macro will
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

87

Adaptive Meshing
mesh only areas. If you desire to mesh both volumes and areas, you can create a user subroutine, ADAPTMSH.MAC, to perform all the desired operations. You will need to clear any specially-meshed entities
before remeshing. Such a subroutine might look like this:
C*** Subroutine ADAPTMSH.MAC - Your name - Job Name - Date Created
TYPE,1
! Set element TYPE attribute for area meshing
ACLEAR,3,5,2
! Clear areas and volumes to be meshed by this subroutine
VCLEAR,ALL
AMESH,3,5,2
! Mesh areas 3 and 5 (no other areas will be meshed by ADAPT)
TYPE,2
! Change element type for volume mesh
VMESH,ALL
! Mesh all volumes

Please see the TYPE, ACLEAR, VCLEAR, AMESH, and VMESH command descriptions for more information.
We strongly recommend that you include a comment line (C***) to identify your macro uniquely. This
comment line will be echoed in the job printout, and will provide assurance that the ADAPT macro
has used the correct user subroutine.

3.3.2.2. Creating a Custom Subroutine for Boundary Conditions (ADAPTBC.MAC)

The ADAPT macro clears and remeshes with every new solution loop. As a result, your model's nodes
and elements will be repeatedly changing. This situation generally precludes the use of finite element
loads, DOF coupling, and constraint equations, all of which must be defined in terms of specific nodes
and elements. If you need to include any of these finite-element-supported items, you can do so by
writing a user subroutine, ADAPTBC.MAC. In this subroutine, you can select nodes by their location,
and can then define finite element loads, DOF coupling, and constraint equations for the selected nodes.
A sample ADAPTBC.MAC subroutine follows:
C*** Subroutine ADAPTBC.MAC - Your name - Job Name - Date Created
NSEL,S,LOC,X,0
! Select nodes @ X=0.0
D,ALL,UX,0
! Specify UX=0.0 for all selected nodes
NSEL,S,LOC,Y,0
! Select nodes @ Y=0.0
D,ALL,UY,0
! Specify UY=0.0 for all selected nodes
NSEL,ALL
! Select all nodes

3.3.2.3. Creating a Custom Solution Subroutine (ADAPTSOL.MAC)

The default solution command sequence included in the ADAPT macro is simply:
/SOLU
SOLVE
FINISH

This default sequence will solve only a single load step. You might be able to implement other solution
sequences by incorporating them into the user subroutine ADAPTSOL.MAC.

3.3.2.4. Some Further Comments on Custom Subroutines

You can create the ADAPTMSH.MAC, ADAPTBC.MAC, and ADAPTSOL.MAC subroutine files as you
would any user files.
You can use either the APDL command *CREATE (Utility Menu> Macro> Create Macro), the APDL
command *END, or a third-party text editor. When the ADAPT macro calls the subroutines, the program
will search first through the ANSYS root directory, then through your root directory, and last through
the current directory. Thus, you should take care to ensure that no identically-named files from another
directory will be read in place of your intended files. The comment lines (C***) shown in the example
subroutines above would be echoed in your printout, and would provide one means of checking that
the proper files were used. Additionally, executing /PSEARCH,OFF (Utility Menu> Macro> Macro Search

88

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Adaptive Meshing Hints and Comments

Path) before running the ADAPT macro would restrict the file search to the ANSYS root directory and
your current directory, reducing somewhat the possibility that the wrong file will be used.
No matter where they reside, the subroutines will be accessed only if you set KYMAC = 1 on the ADAPT
command (Main Menu> Solution> Solve> Adaptive Mesh). See the Guide to User-Programmable Features
for more information on user subroutines and the ANSYS Parametric Design Language Guide for more
information on APDL.

3.3.3. Customizing the ADAPT Macro (UADAPT.MAC)

For those cases when you need to modify the ADAPT macro but are unable to do so via separate user
subroutines, you can modify the main body of the macro.
To maintain the integrity of the ADAPT macro, ANSYS does not recommend modifying the ADAPT.MAC
file directly. Instead, use the UADAPT.MAC file (an identical copy of ADAPT.MAC), provided on the
ANSYS installation media.
If you modify the UADAPT.MAC file, it is a good idea to rename the modified file (especially to denote
the specific version that you have created). Afterwards, instead of issuing the command ADAPT, call
the modified adaptive meshing procedure by issuing the new file name.
Be aware that if you use the new file name as an "unknown command," the program will first search
through the high-level directory, then through the login directory, and finally through the working
directory, until the macro is found. If a modified adaptive procedure is to be accessible by a single user,
it makes sense to store the file in a directory no higher than the user's login directory. If the macro file
is stored in such a low-level directory, the file search can be streamlined by calling the macro using the
*USE command (Utility Menu> Macro> Execute Data Block) in place of the unknown command
format.

3.4. Adaptive Meshing Hints and Comments

Use the following hints to enhance your implementation of adaptive meshing:
No initial element sizes are required, but they may be specified to aid the adaptive convergence
when desired. If you specify initial element sizes at keypoints, the ADAPT macro will use these sizes
in its first loop, and will adjust these sizes as appropriate in subsequent loops. To specify initial element
sizes, use one of these methods:
Command(s): KESIZE
GUI: Main Menu> Preprocessor> Meshing> Size Cntrls> ManualSize> Keypoints> All KPs
Main Menu> Preprocessor> Meshing> Size Cntrls> ManualSize> Keypoints> Picked KPs
If you specify line divisions or spacing ratios, the ADAPT macro will use these values in every loop. (That
is, line divisions or spacing ratios that you specify will not be changed by the ADAPT macro.) If you do
not specify mesh divisions of any kind, default element sizing (SMRTSIZE and DESIZE) will be used for
the initial mesh. To specify line divisions or spacing ratios, use one of these methods:
Command(s): LESIZE
GUI: Main Menu> Preprocessor> Meshing> Size Cntrls> ManualSize> Lines> All Lines
Main Menu> Preprocessor> Meshing> Size Cntrls> ManualSize> Lines> Picked Lines
Mapped meshing (all quadrilaterals) is available for 2-D solid and 3-D shell analyses. The benefits of
using mapped area meshing are usually minimal, however.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

89

Adaptive Meshing
Mapped meshing (all hexahedral bricks) is available for 3-D solid analyses. Mapped volume meshing,
if possible for a given model, will probably give superior performance, compared to tetrahedral
meshing.
In general, midside-node elements will perform better than linear elements in an adaptive meshing
operation.
Do not use concentrated loads or other singularities (such as sharp re-entrant corners), because the
ADAPT procedure cannot show energy value convergence with mesh refinement in the vicinity of
these singularities. If a concentrated loading condition is present in your model, replace it with an
equivalent pressure load, applied over a small area. (Or, exclude the region of the concentrated load
from adaptive modification using the select options discussed previously.)
For many problems, it could be preferable to use a number of relatively small, simple regions in place
of fewer large, complicated regions, for best meshing performance.
If the location of maximum response is known or suspected beforehand, then a keypoint should be
placed near that location.
If the ADAPT procedure is used in an interactive run and the ANSYS program aborts abruptly without
issuing the proper error message, then the output file for the adaptive meshing portion of your run
(Jobname.ADPT) should be reviewed to determine the cause.
Similarly, if the ADAPT procedure is used in a batch run, then Jobname.ADPT should be saved and
examined in case of an abrupt abort.
If a model has an excessive curvature in some region, then your model might experience mesh failure.
In this case, use the SIZE field on the KESIZE command (Main Menu> Preprocessor> Meshing>
Size Cntrls> ManualSize> Keypoints> Picked KPs) to define the maximum element edge length
settings at keypoints near the excessive curvature region. FACMX should also be set equal to 1 (in
the ADAPT command) so that element size will not be permitted to grow in the vicinity of the excessive curvature.
You should save the results file (Jobname.RST or Jobname.RTH). In case of a program abort in
the middle of the ADAPT procedure, the results file will still have the entire analysis data from the
previous solution completed by the ADAPT procedure.
You should issue a SAVE (Utility Menu> File> Save as Jobname.db) before starting the adaptive
meshing operation. In case of system abort, Jobname.DB can then be used to restart the analysis.

3.5. Where to Find Examples

The Mechanical APDL Verification Manual describes several analyses that demonstrate adaptive meshing.
The Mechanical APDL Verification Manual consists of test case analyses demonstrating the analysis capabilities of the ANSYS program. While these test cases demonstrate solutions to realistic analysis problems,
the Mechanical APDL Verification Manual does not present them as step-by-step examples with lengthy
data input instructions and printouts. However, most ANSYS users who have at least limited finite element
experience should be able to fill in the missing details by reviewing each test case's finite element
model and input data with accompanying comments.
The Mechanical APDL Verification Manual contains the following adaptive meshing test cases:
VM193 - Adaptive Analysis of 2-D Heat Transfer with Convection

90

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Where to Find Examples

VM205 - Adaptive Analysis of an Elliptic Membrane Under a Uniformly Distributed Load

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

91

92

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 4: Rezoning
In a finite large-deformation analysis, mesh distortion reduces simulation accuracy, causes convergence
difficulties, and can eventually terminate an analysis. Rezoning allows you to repair the distorted mesh
and continue the simulation.
The program's manual rezoning capability allows you to decide when to use rezoning, what region(s)
to rezone, and what remeshing method to use on the selected region(s).
During the rezoning process, the program updates the database as necessary, generates contact elements
if needed, transfers boundary conditions and loads from the original mesh, and maps all solved variables
(node and element solutions) to the new mesh automatically. Afterwards, with equilibrium achieved
based on the mapped variables, you can continue solving using the new mesh.
The following rezoning topics are available:
4.1. Benefits and Limitations of Rezoning
4.2. Rezoning Requirements
4.3. Understanding the Rezoning Process
4.4. Step 1: Determine the Substep to Initiate Rezoning
4.5. Step 2. Initiate Rezoning
4.6. Step 3: Select a Region to Remesh
4.7. Step 4: Perform the Remeshing Operation
4.8. Step 5: Verify Applied Contact Boundaries, Loads, and Boundary Conditions
4.9. Step 6: Automatically Map Variables and Balance Residuals
4.10. Step 7: Perform a Multiframe Restart
4.11. Repeating the Rezoning Process if Necessary
4.12. Postprocessing Rezoning Results
4.13. Rezoning Restrictions
4.14. Rezoning Examples

4.1. Benefits and Limitations of Rezoning

Even when an analysis terminates due to severe mesh distortion, rezoning allows you to continue the
analysis and complete the simulation. You can also use rezoning to improve analysis accuracy and
convergence when the mesh is distorted but does not terminate the analysis.
To illustrate how rezoning works in a case where the analysis terminates, assume that the following
initial mesh and boundary conditions exist:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

93

Rezoning

The simulation terminates at TIME = 0.44. Rezoning begins on the deformed mesh at substep 7 (TIME
= 0.40):

After remeshing the selected region, an acceptable new mesh is ready:

94

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Benefits and Limitations of Rezoning

Based on the new mesh, the simulation concludes successfully at TIME = 1.0:

For a more detailed example, see Rezoning Examples (p. 124).

4.1.1. Rezoning Limitations

The purpose of rezoning is to repair a distorted mesh in order to overcome convergence problems
caused by the distortion.
Rezoning is effective only when the mesh distortion is caused by a large, nonuniform deformation.
Rezoning cannot help if divergence occurs for any other reason such as unstable material, unstable
structures, or numerical instabilities.
Unstable Material
Most nonlinear material models, especially those employing hyperelastic materials, have
their own applicable ranges. When a deformation is too large or a stress state exceeds
the applicable range, the material may become unstable. The instability can manifest
itself as a mesh distortion, but rezoning cannot help in such cases. While it is sometimes
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

95

Rezoning
difficult to determine when material is unstable, you can check the strain values, stress
states, and convergence patterns. A sudden convergence difficulty could mean that
material is no longer stable. The program also issues a warning at the beginning of the
solution indicating when hyperelastic material could be unstable, although such a
warning is very preliminary and applies only to cases involving simple stress states.
Unstable Structures
For some geometries and loads, a deformation may cause a "snap-through," or local
buckling. Such behavior can also manifest itself as a mesh distortion, but one that
rezoning cannot repair. The effect is usually easy to detect by closely checking the deformed region or the load-versus- time (displacement) curve.
Numerical Instabilities
A condition of numerical instability can occur when a problem is nearly overconstrained.
The constraints can include kinematic constraints such as applied displacements, CP,
and CE, and volumetric constraints introduced by fully incompressible material in mixed
u-P elements. In many cases, numerical instability is apparent even in the early stages
of an analysis.
For a successful rezoning, the new mesh must be of a higher quality than the old mesh. If the new
mesh is not better than the original mesh, rezoning cannot improve convergence, and can even worsen
convergence problems.

4.2. Rezoning Requirements

Rezoning is available for 2-D and 3-D analyses. All multiframe restart files must be available.
Following are the supported analysis types, elements, materials, loads, boundary conditions, and other
rezoning requirements:
Support Category
Solid elements

Requirements
PLANE182 -- B-bar method only (KEYOPT(1) = 0).
PLANE183
For both PLANE182 and PLANE183, all stress states (KEYOPT(3)) are allowed: plane strain, plane stress, axisymmetric, and generalized plane
strain. Pure displacement formulation (KEYOPT(6) = 0) or mixed u-P formulation (KEYOPT(6) = 1) is allowed.
SOLID185 -- B-bar method only (KEYOPT(1) = 0). Pure displacement formulation (KEYOPT(6) = 0) or mixed u-p formulation (KEYOPT(6) = 1) is
allowed in a nonlayered configuration (KEYOPT(3) = 0).
SOLID285

Contact elements

TARGE169
CONTA171 and CONTA172 with any of the following valid KEYOPT settings:
KEYOPT (1) = 0

96

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Rezoning Requirements
Support Category

Requirements
KEYOPT (2) = 0, 1, 3, 4
KEYOPT (3) = 0
KEYOPT (4) = 0, 1, 2
KEYOPT (5) = 0, 1, 2, 3, 4
KEYOPT (7) = 0, 1, 2, 3
KEYOPT (8) = 0
KEYOPT (9) = 0, 1, 2, 3, 4
KEYOPT (10) = 0, 2
KEYOPT (11) = 0
KEYOPT (12) = 0, 1, 2, 3, 4, 5, 6
KEYOPT(14) = 0
TARGE170
CONTA173 and CONTA174 with any of the following valid KEYOPT settings:
KEYOPT (1) = 0
KEYOPT (2) = 0, 1
KEYOPT (4) = 0, 2
KEYOPT (5) = 0, 1, 2, 3, 4
KEYOPT (7) = 0, 1, 2, 3
KEYOPT (8) = 0
KEYOPT (9) = 0, 1, 2, 3, 4
KEYOPT (10) = 0, 2
KEYOPT (11) = 0
KEYOPT (12) = 0, 1, 2, 3, 4, 5, 6
KEYOPT(14) = 0

Contact pair
behavior

Rigid-flexible -- Target elements and pilot node cannot be remeshed.

Flexible-flexible contact.
Self-contact.

Materials

All structural materials are supported except for CAST (cast iron), CONCR
(concrete), MPLANE (microplane), SMA (shape memory alloy), and SWELL
(swelling).
For materials associated with user-defined state variables (TB,STATE), it
may be necessary to issue the MAPVAR command.

Analysis types

Static analysis with geometric nonlinearity (NLGEOM,ON and SOLCONTROL,ON).

Loads and
boundary conditions (BCs)

Displacements, forces, pressures, and nodal temperatures (applied via

a BF,TEMP command).
Loads applied on surface-effect elements (such as SURF153 and SURF154)
are not supported.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

97

Rezoning
Support Category

Requirements
Tabular displacements and forces are supported.
Tabular pressures that are functions of time only are supported.

Region to be
remeshed

The selected nodes inside the region must have the same nodal coordinate system.
Boundary nodes can have different nodal coordinate systems.
Elements must be of the same element type, material, element coordinate
system, and real constant.
If two regions with different attributes require remeshing, you must
remesh the regions separately. For more information, see Hints for
Remeshing Multiple Regions (p. 106).

Files

.rdb, .rst, .rxxx, .ldhi, and .cdb.

The conditions specified apply only to the region to be remeshed. No limitations exist for other regions,
although the analysis type itself must support rezoning.

4.3. Understanding the Rezoning Process

The rezoning process involves the following general tasks:
1. Determine the load step and substep at which the region must be remeshed.
2. Initiate the rezoning process.
3. Select the region to remesh.
4. Remesh the region with a mesh of better quality than the distorted mesh in the original domain.
5. Verify boundary conditions, loads, temperatures, and fluid-penetration parameters applied to the
new mesh (from the old, distorted mesh).
6. Automatically map displacements and state variables from the old (distorted) mesh to the new mesh
and rebalance the resulting residual forces.
7. Use the restart capability to continue the analysis.
You can select one or more parts, or regions, of the mesh to repair at the same time. You can also perform
rezoning multiple times in an analysis.
The following topics related to understanding the general rezoning process are available:
4.3.1. Overview of the Rezoning Process Flow
4.3.2. Key Commands Used in Rezoning

4.3.1. Overview of the Rezoning Process Flow

The following flowcharts illustrate the general rezoning process according to the remeshing method
used:
98

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Understanding the Rezoning Process

Figure 4.1: Rezoning Using a Program-Generated New Mesh
Figure 4.2: Rezoning Using a Generic New Mesh Generated by Another Application
Figure 4.3: Rezoning Using Manual Splitting of an Existing Mesh
Some mesh-generation methods are limited to 2-D analyses only.
For information about the commands shown in the flowcharts, see Key Commands Used in Rezoning
(p. 101)or the documentation for a given command in the Command Reference.
Figure 4.1: Rezoning Using a Program-Generated New Mesh
This flowchart shows a process for 2-D rezoning using a new mesh generated internally
by the program (AREMESH and AMESH):

Figure 4.2: Rezoning Using a Generic New Mesh Generated by Another Application
This flowchart shows a process for both 2-D and 3-D rezoning using a generic (.cdb
format) new mesh generated by a third-party application:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

99

Rezoning

Figure 4.3: Rezoning Using Manual Splitting of an Existing Mesh

This flowchart shows a process for 2-D rezoning by splitting an existing mesh:

100

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Understanding the Rezoning Process

4.3.2. Key Commands Used in Rezoning

Following is a description of key commands used in the rezoning process and provides supplemental
information to the rezoning process flowcharts.
Command

Description

Rezoning Comments

/CLEAR

Clears the database

Always clear the database first, before reentering the solution processor (/SOLU) and starting the rezoning process.

REZONE

Initiates rezoning

When you initiate rezoning, the program verifies that the

necessary files (.rdb, .rst, .rxxx, and .ldhi) exist for
the specified substep and rebuilds the data environment
at that substep.
All nodes are updated to the deformed geometry in
preparation for remeshing.

REMESH

Generates or
obtains the new
mesh required
for rezoning

You can remesh more than one region at the same specified
substep during the rezoning process.
After remeshing (REMESH,FINISH), the program generates contact elements (if any) and transfers loads and
boundary conditions automatically.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

101

Rezoning
Command

Description

Rezoning Comments
For more information, see Step 4: Perform the
Remeshing Operation (p. 104).

MAPSOLVE

Maps node and

element solutions

After remeshing, this command maps the solved nodal and

element solutions from the original (distorted) mesh to the
new mesh and achieves equilibrium on the new mesh.

4.4. Step 1: Determine the Substep to Initiate Rezoning

Determining Why the Solution Terminated
To investigate the reason(s) why a solution terminated, you can enter the POST1 general postprocessor (/POST1) to review analysis results.
You can also use the GUI to select a substep for which restart files are available but results are
not saved in a results file, generate the results for the selected substep, then review the results
in POST1.
If the solution terminated because of a mesh distortion, it is a candidate for rezoning.
It is important to select a suitable substep at which to start rezoning. You must select a converged
substep where the restart files are available.
The last converged substep is often a natural choice for initiating rezoning because there is typically
less of a remaining load to apply. However, the last converged substep may have more severely distorted
elements which can cause larger errors when mapping solution variables (MAPSOLVE), in turn requiring
more mapping substeps to balance residual forces or convergence failures. Rezoning from a substep
with an extremely distorted mesh may also reduce the accuracy of the final solution, and can even
cause the automatic transfer of boundary conditions after remeshing (REMESH,FINISH) to fail.
Typically, the best choice is the first, second, or third converged substep preceding the last converged
substep. To determine the best possible substep to initiate rezoning, you may need to enter the POST1
general postprocessor (/POST1) to examine the deformed element shapes, and stress and strain distributions.
If no results data exists for a substep in the results file, you can use program restart features to generate
results data for the substep, and then enter POST1 to examine the deformed shape; afterwards, you
can reenter the solution processor and initiate rezoning as usual.
Hints for selecting the best substep to initiate rezoning:
The substep should have an obvious mesh distortion but should contain no element having an internal
angle too closely approaching, equal to, or exceeding 180 degrees.
The substep should have minimal penetration in contact. Try to avoid substeps where contact status
has changed suddenly from previous substeps.
If an error occurs when the program transfers boundary conditions and loads after remeshing
(REMESH,FINISH) and the old mesh is severely distorted, try the substep preceding the one most recently
chosen.

102

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Step 3: Select a Region to Remesh

The best substep is often not the last converged substep, but rather one of the several preceding the
last one.
If the last few converged substeps are separated by very small time increments and you have already
tried one or more of those substeps unsuccessfully, select a converged substep that is separated from
the others by a larger time increment.
If you are still having difficulty obtaining a mesh of reasonable quality because the old mesh is too
distorted, try the substep preceding the one most recently chosen.
If the mapping operation (MAPSOLVE) fails to converge even if you allow up to 500 substeps, try the
substep preceding the one most recently chosen.
When you have determined the substep at which to initiate rezoning, proceed to Step 2. Initiate
Rezoning (p. 103).

4.5. Step 2. Initiate Rezoning

Rezoning is based on an existing solution, so the initial solution must have terminated.
Because the simulation terminated, the rezoning process must begin with a clean database.
To start rezoning:
1. Clear the database (/CLEAR).
2. Reenter the solution processor (/SOLU).
3. Initiate rezoning, specifying the load step and substep at which rezoning should occur
(REZONE,MANUAL,LDSTEP,SBSTEP).
Proceed to Step 3: Select a Region to Remesh (p. 103).

4.6. Step 3: Select a Region to Remesh

A region that you select (ESEL) for remeshing can contain the entire deformed domain or a portion of
it. A selected region should consist of the same:
Material type
Element type (including the coordinate system and KEYOPT settings)
Thickness (real constant) for plane stress
Nodal coordinate system (except for boundary nodes which can have different nodal coordinate
systems).
A selected region should contain all of the highly distorted elements. It is a good practice to select a
group of elements slightly larger than the target group of distorted elements so that the new mesh
can be assured of a fairly good distribution of nodes on the interface boundary.
If the boundary nodes are distributed too unevenly, the elements attached to the nodes should also
be included. The selected region's boundary can have any shape.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

103

Rezoning
A selected region that is too large may require more processing time and more subsequent remeshings.
If the selected region is too small to contain all of the highly distorted mesh areas, the new model after
rezoning may not be of sufficient quality to achieve convergence.
Using the GUI to Select a Region to Remesh
Select a region to remesh using either of the following methods available via the ANSYS Main Menu:
In the SOLU solution processor: Solution>Manual Rezoning>Select Rezone Elements
In the POST1 postprocessor: General Postproc>Manual Rezoning>Create Rezone Component
Select elements and group them into a component. During rezoning, import the component after
remeshing starts (REMESH,START).
This component can be imported into the program or exported to a third-party tool to create a new
mesh for the selected region.
If more than one region requires rezoning, see Remeshing Multiple Regions at the Same Substep (p. 116).
After you have selected the region(s) to remesh, proceed to Step 4: Perform the Remeshing Operation (p. 104).

4.7. Step 4: Perform the Remeshing Operation

Issue the REMESH,START command to begin the remeshing operation.
When remeshing begins, your analysis temporarily exits the solution processor and enters a special
mode of the PREP7 preprocessor, where a limited number of preprocessing commands are available
for mesh control.
The following topics concerning the remeshing operation are available:
4.7.1. Choosing a Remeshing Method
4.7.2. Mesh Control
4.7.3. Remeshing Multiple Regions at the Same Substep
To exit the special preprocessing mode and reenter the solution processor, issue a REMESH,FINISH
command.
After remeshing, proceed to Step 5: Verify Applied Contact Boundaries, Loads, and Boundary Conditions (p. 116).

4.7.1. Choosing a Remeshing Method

To generate a new mesh for a selected region, you can use the program's native capabilities, import a
generic new mesh from a third-party application, or manually split an existing mesh:
4.7.1.1. Remeshing Using a Program-Generated New Mesh (2-D)
4.7.1.2. Remeshing Using a Generic New Mesh (2-D and 3-D)
4.7.1.3. Remeshing Using Manual Mesh Splitting (2-D)

104

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Step 4: Perform the Remeshing Operation

4.7.1.1. Remeshing Using a Program-Generated New Mesh (2-D)

This remeshing method requires the AREMESH and AMESH commands, as shown in Figure 4.1: Rezoning
Using a Program-Generated New Mesh (p. 99). Remeshing using a program-generated mesh is available
for 2-D analyses only.
The following topics are available if you intend to remesh with a program-generated mesh:
4.7.1.1.1. Creating an Area to Remesh
4.7.1.1.2. Using Nodes From the Old Mesh
4.7.1.1.3. Hints for Remeshing Multiple Regions
4.7.1.1.4. Generating a New Mesh
To study a sample problem, see Example: Rezoning Using a Program-Generated New Mesh (p. 124).

4.7.1.1.1. Creating an Area to Remesh

Based on the selected region, the AREMESH command creates an area to generate the new mesh. The
program validates the selected region and creates an area on that region for the new mesh. The program
maintains compatibility with neighboring regions, and nodes with applied loads and boundary conditions
remain intact.
How the program creates the boundary lines affects the new mesh quality and mesh density on the
area; therefore, specify AREMESH command options carefully. The boundary lines are based on the
element edges on the boundary of the selected region, as follows:
Line combining (AREMESH,0) allows you to redistribute the new nodes on the boundary and provides
more control over the new element size. New elements can be larger or smaller. If the boundary is highly
curved, however, the new element edges may constitute a slightly different boundary.
If line segments are not combined (AREMESH,-1), the new boundary will match the old one. In this case,
you cannot control the positions of old nodes on the boundary (which may result in an unacceptable new
mesh), and elements on the boundary can only be the same size or smaller.
To maintain compatibility, the program does not combine line segments connected to elements outside
the selected remeshing region (even if you specify line combining). Also, two segments are not combined
if an old node is located between them and that node:
Has applied force or displacements, or
Is the starting and ending point of applied pressures, distributed displacements, or contact boundaries.

4.7.1.1.2. Using Nodes From the Old Mesh

It is best to avoid using nodes from the old mesh whenever possible. Retaining old nodes introduces
more constraints when generating the new mesh and makes it more difficult to create a new mesh of
better quality.
If necessary, you can retain some old nodes on the boundaries of the selected region to use on the
new mesh. To do so, select the nodes that you want to keep and group them into a nodal component
named _ndnocmb_rzn (CM,_ndnocmb_rzn,NODE) before issuing the AREMESH command.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

105

Rezoning

4.7.1.1.3. Hints for Remeshing Multiple Regions

When remeshing using a program-generated new mesh, you can remesh multiple regions at a given
substep (referred to as horizontal rezoning) as follows:
1. After starting the remeshing operation (REMESH,START), select a region to remesh.
2. Generate the area for the new mesh (AREMESH).
3. Create the new mesh (AMESH).
4. Select another region (being careful not to overlap regions), generate the area for its new mesh, and
create the new mesh on that region.
Repeat the process for as many regions as you wish, but only after issuing the REMESH,START command
and before issuing a REMESH,FINISH command (as shown in Understanding the Rezoning Process (p. 98)).
When remeshing two regions or areas that connect to each other, it is best to select them as a single
region. If two connected regions must be treated separately, create the mesh for the first region before
remeshing the second one.

4.7.1.1.4. Generating a New Mesh

After generating the area (AREMESH) for the selected region, issue the AMESH command to generate
the new mesh.
Several preprocessing (/PREP7) commands are available to help you create a good mesh on the selected
region. For more information, see Mesh Control (p. 114).
The mesh-control commands are available at any point after issuing a REMESH,START command and
before issuing a REMESH,FINISH command.

4.7.1.2. Remeshing Using a Generic New Mesh (2-D and 3-D)

You can remesh using a generic new mesh created by another application. This remeshing method is
available for both 2-D and 3-D analyses.
As shown in Figure 4.2: Rezoning Using a Generic New Mesh Generated by Another Application (p. 99),
using a generic mesh for the remeshing operation requires the REMESH,READ command, used in place
of the commands required for other meshing methods.
To use a new third-party mesh, the mesh file must have a .cdb file format. The .cdb file must have
mesh information, but an IGES file (geometry information) is not required. Typically, the new .cdb
mesh is generated from a faceted geometry representation of the boundary of the region to be rezoned.
The element types supported for remeshing with a generic new mesh are given in Rezoning Requirements (p. 96).
The following additional topics related to using a generic new mesh are available:
4.7.1.2.1. Using the REMESH Command with a Generic New Mesh
4.7.1.2.2. Requirements for the Generic New Mesh
4.7.1.2.3. Using the REGE and KEEP Remeshing Options
To study a sample problem, see Example: Rezoning Using a Generic New Mesh (p. 127).

106

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Step 4: Perform the Remeshing Operation

4.7.1.2.1. Using the REMESH Command with a Generic New Mesh

The commands used for reading in a generic new mesh are shown in Figure 4.2: Rezoning Using a
Generic New Mesh Generated by Another Application (p. 99).
Because the REMESH command's READ option reads only generic meshes, all properties of the solid
elements in the new mesh are inherited internally from the corresponding underlying solid elements
in the old mesh. The program ignores the solid element properties of the new mesh in the .cdb file
and calculates them internally depending upon their location in the model; therefore, only the NBLOCK
and EBLOCK records of the .cdb file (which define the nodal coordinates and element connectivity,
respectively) are read in after issuing a REMESH,READ command.
You can issue multiple REMESH,READ commands for various parts of the mesh in the same rezoning
problem (referred to as horizontal rezoning). These multiple regions can be isolated or can touch each
other at the boundary, but they cannot overlap. The new mesh, representing multiple regions, can also
reside in a single .cdb file.

4.7.1.2.2. Requirements for the Generic New Mesh

The following general requirements must be met by the generic mesh if it is to be a candidate for a
new mesh for rezoning:
The space that the new mesh space occupies, as well as its topology, must conform to that of the old
mesh.
Locations of concentrated forces in the old mesh (usually a node) must be present in the new mesh.
Limit locations for pressure loads and boundary conditions in the old mesh must be present in the new
mesh.
Limit locations for contact and contact/target element descriptions (in rigid-flex contact and flex-flex
contact, respectively) in the old mesh must be present in the new mesh.
The locations of the limits are marked by single nodes in 2-D analyses and by a line of connected nodes
in 3-D analyses.
It is not necessary to specify loads, boundary conditions, material properties, etc. on the generic mesh's
.cdb file. The program assigns those values to the new mesh from the model automatically and ignores
any specified values. If necessary, you can add new loads later via additional load steps or a restart.
ANSYS, Inc. recommends exporting the deformed old mesh with all discretized boundary information
to a suitable third-party application and, when generating the new mesh, verifying that the node positions
of concentrated loads, contact/target region limits, boundary condition and distributed load limits are
retained. If these key nodes are not retained, you will be unable to proceed with the analysis.
The new mesh is acceptable even if the smoothed boundary geometry of the new mesh does not correspond exactly to the faceted geometry of the old mesh, as shown in Figure 4.4: Boundary Geometry
of a Generic (CDB) New Mesh (p. 108); however, the offset must be very small.
In a 3-D analysis involving contact, the difference of the recovered boundary geometry from the faceted
boundary of the old mesh affects mapping results much more so than in a 2-D analysis. Accurate geometry extraction is therefore essential in 3-D rezoning.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

107

Rezoning
Figure 4.4: Boundary Geometry of a Generic (CDB) New Mesh

If the rezoned part has contact/target elements, the program generates those elements automatically
for the new mesh, depending on whether the underlying old mesh had the same type of contact/target
elements. Isolated rigid target elements in the model remain the same throughout the analysis and
cannot be remeshed; however, all contact and target elements associated with solid elements are candidates for remeshing. While it is possible to read in the new contact/target elements of the new mesh
from the .cdb file, it is faster and more reliable to read in only the remeshed solid elements and allow
the program to generate the new contact/target elements.
The .cdb file of the new mesh must have no line breaks in the NBLOCK and EBLOCK records. Also,
while writing the mesh .cdb file, a block file format is necessary (CDWRITE,,,,,,,Fmat, where Fmat =
BLOCKED).

4.7.1.2.3. Using the REGE and KEEP Remeshing Options

The REMESH command's REGE option regenerates all node and element numbers on the new mesh
using an offset of the highest existing node and element numbers. The KEEP option keeps the similarly
numbered nodes and elements in the new and the old meshes unchanged.
Apply Option = KEEP carefully. It assumes that either the new mesh node and element numbers are
already offset by the maximum node and element number of the old mesh or that the common node
and element numbers in the new mesh and the old mesh match geometrically.
The following figure illustrates how the REMESH command's REGE and KEEP options work:
Figure 4.5: Remeshing Options when Using a Generic (CDB) New Mesh
In this example, a meshed domain with 24 nodes and 15 elements is remeshed using
the REMESH command's REGE (default) option. The .cdb file for the new mesh has
nodes 1 through 16 and element numbers 1 through 9. After remeshing, these node
and element numbers are suitably offset by the maximum node and element numbers
(that is, 15 and 24, respectively) of the old mesh.

108

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Step 4: Perform the Remeshing Operation

The same problem appears in this example. However, the .cdb file for the new mesh
has node numbers defined from 28 through 43 and element numbers defined from 17
through 25. In this case, remeshing occurs using the REMESH command's KEEP option,
so the node and element numbers are not offset.

For more information, see the REMESH command documentation as it applies to remeshing using a
generic mesh created by another application.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

109

Rezoning

4.7.1.3. Remeshing Using Manual Mesh Splitting (2-D)

You can manually split an existing mesh to obtain the solution of a nonlinear analysis which cannot
otherwise converge, or to improve its accuracy. This remeshing method is available in 2-D analyses
only.
Mesh splitting increases the number of degrees of freedom of the model by enriching the existing
mesh. It is a useful option for rezoning if the number of degrees of freedom must be increased in contact
gaps, or if a new program-generated mesh or generic third-party mesh does not fully satisfy your requirements.
As shown in Figure 4.3: Rezoning Using Manual Splitting of an Existing Mesh (p. 100), the mesh-splitting
remeshing method requires the REMESH,SPLIT command, used in place of commands required for
other meshing methods.
The following topics related to manual mesh-splitting are available:
4.7.1.3.1. Understanding Mesh Splitting
4.7.1.3.2. Geometry Details for Mesh Splitting
4.7.1.3.3. Using the REMESH Command for Mesh Splitting
4.7.1.3.4. Mesh Transition Options for Mesh Splitting

4.7.1.3.1. Understanding Mesh Splitting

Splitting occurs on selected solid elements in the mesh. If no solid elements are explicitly selected, the
program splits all solid elements in the mesh. The mesh-splitting capability generally applies to
remeshing cases in 2-D mesh-rezoning problems using the elements listed in Rezoning Requirements (p. 96).
The program splits parent elements into child elements, as follows:
A parent quadrilateral element into four child quadrilateral elements
A parent degenerate element into three child quadrilateral elements
A parent triangular element into four child triangular elements
For more information, see Geometry Details for Mesh Splitting (p. 110).
Child elements inherit all shape characteristics of the parent element. Therefore, if a particular element
is badly distorted and is causing convergence difficulties, simply subdividing the element by splitting
it does not improve convergence.

4.7.1.3.2. Geometry Details for Mesh Splitting

The REMESH command's SPLIT option uses no geometry information; instead, it uses only information
about the mesh itself (that is, nodal connectivity and nodal coordinates).
All child elements automatically inherit all attributes of the parent element from which they were generated. Quadrilateral parent elements are split into four child elements, degenerate parent elements
are split into three quadrilateral child elements, and triangular parent elements are split into four triangular child elements, as shown in the following figures:

110

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Step 4: Perform the Remeshing Operation

Figure 4.6: Splitting of Quadrilateral and Degenerate Linear Elements ( LANE182)
P

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

111

Rezoning
Figure 4.7: Splitting of Quadrilateral, Degenerate and Triangular Quadratic Elements ( LANE183)
P

4.7.1.3.3. Using the REMESH Command for Mesh Splitting

To perform mesh-splitting, select the region to be rezoned (via the GUI or the ESEL command). After
you have selected the target region, issue a REMESH,SPLIT command.
You can issue multiple REMESH,SPLIT commands for various parts of the mesh in the same rezoning
problem (referred to as horizontal rezoning). These multiple zones can overlap or they can be isolated;
however, a large number of overlaps can cause badly shaped transition elements to develop.
If the rezoned part has contact/target elements, the program generates those elements automatically
for the new mesh (according to whether the underlying old mesh had the same type of contact/target
elements). Isolated rigid target elements in the model remain the same throughout the analysis and
cannot be remeshed; however, all contact and target elements associated with solid elements are candidates for remeshing. When you split a solid element that is associated to contact/target elements,
the program deletes these associated contact/target elements. The program generates the correct
112

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Step 4: Perform the Remeshing Operation

contact/target elements for the new child elements automatically at the end of the remeshing operation
(REMESH,FINISH).
Because splitting refinement is mesh-based and not geometry-based, it cannot be undone after it has
occurred. To create a different or new splitting scheme, or to revert to the original mesh, you must
create a new rezoning environment (REZONE,MANUAL,LDSTEP,SBSTEP).
For more information about using the REMESH,SPLIT command, see Mesh Transition Options for Mesh
Splitting (p. 113).

4.7.1.3.4. Mesh Transition Options for Mesh Splitting

The default REMESH,SPLIT command forces the transition elements to be mostly quadrilateral and
minimizes the number of degenerated elements. The command can also be issued as:
REMESH,SPLIT,,,TRAN,QUAD
Issuing the REMESH command using the transition and quadrilateral options helps with convergence
because non-degenerate elements are less prone to locking behavior.
It is possible that transition elements designed in such a way can disturb the localization of the mesh.
The element subdivision that occurs when transitioning from the split zone can traverse several element
layers.
Generating a More Localized Mesh
If you desire a more localized mesh after splitting, issue the following command:
REMESH,SPLIT,,,TRAN,DEGE
In this case, the program creates the degenerate elements in the transition zone, and the split and
the unsplit regions are connected within a single element layer.
The following figure illustrates the options for transition element generation:
Figure 4.8: Transition Element Generation Methods

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

113

Rezoning
Elements selected for splitting (ESEL) are marked as shown.

Default splitting algorithm with all-quadrilateral transition elements and

no degenerate elements.
Regions of non-localization are shown. Either the REMESH,SPLIT command
or the REMESH,SPLIT,,,TRAN,QUAD command can apply here.

Splitting with single-layer transition and degenerate elements.

In this case, the REMESH,SPLIT,,,TRAN,DEGE command applies.

4.7.2. Mesh Control

Creating a better mesh is the key to successful rezoning. The new mesh must be better than the old
mesh; otherwise, rezoning cannot improve convergence, and can even result in nonconvergence if the
new mesh is worse.

114

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Step 4: Perform the Remeshing Operation

Generally, a good mesh has these characteristics:
The element internal angles do not approach 180 degrees or 0 degrees.
The opposite edges or faces of quadrilaterals have less cross angle (that is, they are closer to parallel with
each other).
Boundary nodes are more evenly distributed.
The elements have a better aspect ratio.
For creating a good mesh, satisfactory internal angles are a more important consideration than whether
aspect ratios may be too high or too low. For 2-D meshes, avoid triangle elements as much as possible;
however, if quadrilateral elements have very large internal angles, it is preferable to accept some triangle
elements with more acceptable internal angles instead. For 3-D meshes, try to avoid pyramid elements
altogether, and use as few degenerate shapes as possible.
The quality of the new mesh is fully dependent on the shape of the selected region, neighboring elements, and boundary nodes with applied loads and boundary conditions. It may sometimes be necessary
to edit or modify the new mesh prior to the mapping operation (MAPSOLVE); in such cases, several
preprocessing (/PREP7) commands are available to help you create a good mesh on the selected region:
Figure 4.9: /PREP7 Mesh-Control Commands Available in Rezoning
Type of Mesh Control Desired

Applicable Commands

Element size

AESIZE, DESIZE, ESIZE, KESIZE, LESIZE, and SMRTSIZE

Element shape

MOPT, MSHAPE, and MSHKEY

Element internal
angles

SHPP,MODIFY to reset the shape parameter limits. Specify input values of

11~14 and 17~18 depending on the element type.

Boundary node distribution

Boundary nodes require special attention. Use the LESIZE command to

space the nodes as uniformly as possible without disturbing other characteristics of the mesh.

Refining

AREFINE, EREFINE, KREFINE, LREFINE, NREFINE,

Other

ACLEAR, AMESH, KSCON, LCCALC, MSHMID

If you remesh using a generic new mesh (rather than a programgenerated mesh), the N and EN commands are also available.
The ACLEAR command applies only to the area generated via the
AREMESH command (that is, when remeshing using a program-generated new mesh).

The mesh-control commands are available at any point after issuing a REMESH,START command and
before issuing a REMESH,FINISH command.
Nodes at which force or isolated displacements are applied must remain during remeshing, as do nodes
at the boundaries of distributed displacements, pressures, or contact/target surfaces. The new elements
must have the same attributes as the old elements, such as element type, material, real constant, and
element coordinate system. The program rotates the new nodes to the same angles as the old nodes
in the region or on the boundary, as the case may be.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

115

Rezoning
After creating the new mesh, verify that the new elements cover the entire selected region and are
compatible with neighboring regions. (If another region requires remeshing, you can do so, but do not
issue another REMESH,START command.

4.7.3. Remeshing Multiple Regions at the Same Substep

You can create a new mesh for more than one region in the same deformed domain. Remeshing two
or more regions at the same specified substep is called horizontal rezoning.
You must use horizontal rezoning if you intend to remesh:
Two regions with different materials, element coordinate systems, nodal coordinate systems, or real constants.
The elements on both sides of a contact boundary.
Two groups of highly distorted, noncontiguous elements.
Horizontal rezoning using a program-generated new mesh or a generic new mesh is possible only when
the remeshed areas are isolated from each other or, at most, share an interface; however, the remeshed
areas may not intersect. This restriction does not exist for manual mesh splitting.

4.8. Step 5: Verify Applied Contact Boundaries, Loads, and Boundary

Conditions
The program automatically generates the target/contact elements if the underlying solid elements are
remeshed, then transfers all applied boundary conditions (BCs) and loads to the new mesh. Use plot
or list commands to verify that the newly generated target/contact elements, and the transferred BCs
and loads, are complete and correct.
For further information, the following related topics are available:
4.8.1. Contact Boundaries
4.8.2. Pressure and Contiguous Displacements
4.8.3. Forces and Isolated Applied Displacements
4.8.4. Nodal Temperatures
4.8.5. Other Boundary Conditions and Loads
After the contact boundaries, loads, and boundary conditions have been applied and verified, proceed
to Step 6: Automatically Map Variables and Balance Residuals (p. 118).

4.8.1. Contact Boundaries

Rigid target elements cannot be remeshed. They are automatically transferred from the original mesh
to the new mesh.
Deformable target elements and all contact elements are always attached to solid elements; therefore,
provided that the underlying solid elements are remeshed, the program creates the target/contact
elements automatically after exiting remeshing (REMESH,FINISH). the program passes all specified element
options, real constants, and materials from the old target/contact elements to the new ones automatically.
Verify that the new elements on the contact boundary are complete and correct.

116

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Step 5: Verify Applied Contact Boundaries, Loads, and Boundary Conditions

4.8.2. Pressure and Contiguous Displacements

During remeshing, if you are using a program-generated new mesh, the program maintains the nodes
on the boundary of the region where pressures or contiguous displacements are applied. If you are
using a generic new mesh created by another application, you must maintain the nodes manually.
In 2-D problems, the boundary consists of only the starting and ending points. In 3-D problems, the
boundary can be a line of ordered nodes representing a line bounding a set of facets.
You can redistribute nodes inside the region (between starting and ending points). If you do so, the
program uses linear interpolation to apply the pressures and displacements at the new node locations.
If the starting and ending points are rotated, the new nodes are also rotated with the angle determined
by linear interpolation of the angles of the old nodes. If the original distribution is not linear, the interpolation may introduce a small degree of error, although the error should not be significant if both the
original mesh and the new mesh are sufficiently dense.
The following illustration shows how the displacements are applied before and after remeshing.

dol ni egnhg
C C
C

stnemelpsid
deilppe

C C
C
C

C
hsem dlo
hsem wen

4.8.3. Forces and Isolated Applied Displacements

If a force or isolated (applied only at one node) displacement is applied at a node, the program maintains
that node, its rotation angle (if any), and the applied forces and displacements, during rezoning.

4.8.4. Nodal Temperatures

The program transfers only nodal temperatures applied via the BF command to the new nodes (by interpolation with shape functions) from the old nodes.
If the remeshed region crosses the boundary where one side has old nodes with applied temperatures
and the other side does not, the interpolation may cause different temperature distributions at the new
nodes close to the boundary. Therefore, avoid remeshing the region crossing the boundary and use
horizontal rezoning instead.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

117

Rezoning

4.8.5. Other Boundary Conditions and Loads

Any other loads and boundary conditions (such as acceleration, coupling, and constraint equations) are
not valid in the remeshed region and are lost during rezoning, which may result in a solution that does
not converge or a solution that is very different from the one expected.

4.9. Step 6: Automatically Map Variables and Balance Residuals

Issue the MAPSOLVE command to map the solved node and element solutions automatically from the
original mesh to the new mesh. The mapping operation also introduces extra substeps to balance the
residuals and achieve equilibrium.
Assuming that you intend to continue the solution, MAPSOLVE is required and is the only rezoning
command that you can issue after remeshing (REMESH,FINISH).
For further information, the following mapping topics are available:
4.9.1. Mapping Solution Variables
4.9.2. Balancing Residual Forces
4.9.3. Interpreting Mapped Results
4.9.4. Handling Convergence Difficulties
When the mapping operation has completed, proceed to Step 7: Perform a Multiframe Restart (p. 120).

4.9.1. Mapping Solution Variables

The program maps the solution variables (such as nodal displacements, and element stresses and strains)
from the old mesh to the new mesh at the rezoned substep. The mapped variables must satisfy the
strain-displacement relationships, constitutive laws, and the equilibrium equations. Algorithms inherent
in the mapping scheme handle the first two conditions, but the mapped stress field is usually not in
equilibrium due to the different mesh. Therefore, the out-of-balance forces (the residual forces) must
be balanced. Balancing occurs via additional substeps.

4.9.2. Balancing Residual Forces

Additional substeps are introduced automatically to reduce residuals to zero. During this stage, the
time is increased by only a tiny amount (TIME * 10-6, where TIME = the time of the current load step).
Therefore, you can consider external loading to be unchanged.
The program always attempts to balance the residual forces within one substep. If it cannot, the program
uses bisection logic (automatically, regardless of any manual settings) to use more substeps. A rebalance
factor measures the balanced residuals and acts as the control factor during bisection (unlike a standard
solution where time is the control factor.) A rebalance factor of zero means that no residuals are balanced
yet, and a factor of 1 means that all residuals are balanced and the stress field is in equilibrium.
You can specify the maximum number of substeps--the default is five--to use during mapping (MAPSOLVE). Most problems should achieve balanced residual forces within a few substeps. If contact is included, more substeps may be necessary.
Output File
In the output file, the program presents MAPSOLVE command data as shown:
S O L U T I O N

118

O P T I O N S

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Step 6: Automatically Map Variables and Balance Residuals

PROBLEM DIMENSIONALITY. . .
DEGREES OF FREEDOM. . . . .
ANALYSIS TYPE . . . . . . .
MAPSOLVE FOR REZONING. .
NONLINEAR GEOMETRIC EFFECTS
STRESS-STIFFENING . . . . .
EQUATION SOLVER OPTION. . .
NEWTON-RAPHSON OPTION . . .
GLOBALLY ASSEMBLED MATRIX .
L O A D

.
.
.
.
.
.
.
.
.

. .
UX
. .
. .
. .
. .
. .
. .
. .

. .
UY
. .
. .
. .
. .
. .
. .
. .

S T E P

. . . .
ROTZ
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .

.2-D
.STATIC (STEADY-STATE)
.YES
.ON
.ON
.SPARSE
.PROGRAM CHOSEN
.SYMMETRIC

O P T I O N S

LOAD STEP NUMBER. . . . . . . . . . . . . .

TIME AT END OF THE LOAD STEP. . . . . . . .
AUTOMATIC TIME STEPPING . . . . . . . . . .
INITIAL NUMBER OF SUBSTEPS . . . . . . .
MAXIMUM NUMBER OF SUBSTEPS . . . . . . .
MINIMUM NUMBER OF SUBSTEPS . . . . . . .
MAXIMUM NUMBER OF SUBSTEPS FOR MAPSOLVE.
MAXIMUM NUMBER OF EQUILIBRIUM ITERATIONS. .
STEP CHANGE BOUNDARY CONDITIONS . . . . . .
STRESS-STIFFENING . . . . . . . . . . . . .
TERMINATE ANALYSIS IF NOT CONVERGED . . . .
CONVERGENCE CONTROLS. . . . . . . . . . . .
PRINT OUTPUT CONTROLS . . . . . . . . . . .
DATABASE OUTPUT CONTROLS
ITEM
FREQUENCY
COMPONENT
ALL
ALL

.
.
.
.
.
.
.
.
.
.
.
.
.

.
1
. 1.0000
.
ON
.
10
.
100
.
5
.
50
.
15
.
NO
.
ON
.YES (EXIT)
.USE DEFAULTS
.NO PRINTOUT

Mapping Substeps
The number of substeps used for mapping appears as follows:
*** LOAD STEP
1
*** TIME = 0.399110

SUBSTEP

10 COMPLETED.
REBALANCE FACTOR =

CUM ITER =
1.00000

101

*** ANSYS BINARY FILE STATISTICS

BUFFER SIZE USED= 16384
0.375 MB WRITTEN ON ELEMENT SAVED DATA FILE: RznExample.esav
0.125 MB WRITTEN ON ASSEMBLED MATRIX FILE: RznExample.full
0.375 MB WRITTEN ON RESULTS FILE: RznExample.rst
MAPSOLVE IS DONE SUCCESSFULLY IN 3 SUBSTEPS FOR MANUAL REZONING.

Monitor File
The 6th and 7th columns of the monitor file indicate the rebalance factor (rather than the time) for
each mapping substep, as shown:
SOLUTION HISTORY INFORMATION FOR JOB: RznExample.mntr
13:06:16
LOAD
STEP

1
1
1

SUBSTEP

8
9
10

NO.
NO. TOTL
ATTMP ITER ITER

2
1
1

7
4
4

93
97
101

INCREMENT
REBALANCE
FACTOR
0.35000
0.35000
0.30000

09/20/2004

TOTAL
REBALANCE
FACTOR
0.35000
0.70000
1.0000

VARIAB 1
MONITOR
CPU
5.4300
8.0600
11.260

VARIAB 2
MONITOR
MxDs
-.64291
-.81383
-.93615

VARIAB 3
MONITOR
MxPl
0.78886E-30
0.78886E-30
0.78886E-30

Although multiple substeps may be necessary to balance all residuals, the program generates the restart
file and saves the result in a results file for only the last converged mapping substep. (Only the last
substep gives the balanced solution.) The program ignores any preexisting output specifications (set
via OUTRES or RESCONTROL commands, for example).

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

119

Rezoning

4.9.3. Interpreting Mapped Results

The mapped results are not simply the solution variables interpolated from the old mesh, but balanced
results mapped from the old mesh. If the new mesh topology or density is very different from the old
mesh, the mapped results may also be very different.
The reported stresses and strains represent the total stresses and strains from TIME = 0. However, the
reported displacements are from the rezoned time or substep (when the element coordinates were
updated via the REZONE command).
After the MAPSOLVE command has executed, the new mesh may be somewhat distorted, and the
distortion may be even more significant if the residual forces are large. In such a case, more careful region
selection and better remeshing are necessary.

4.9.4. Handling Convergence Difficulties

If the rezoning process encounters convergence difficulties during the mapping (MAPSOLVE) stage of
the rezoning process, try the following:
Specify a larger maximum number of mapping substeps (although no more than 500) via the MAPSOLVE
command.
Minimize the differences in mesh density and topology between the old and new mesh, especially in the
elements on boundaries.
Rezone from an earlier substep (so that you can start with a slightly less distorted original mesh).
Enhance the quality of the new mesh.

4.10. Step 7: Perform a Multiframe Restart

After successfully mapping (MAPSOLVE) solved variables from the old mesh to the new mesh and
checking the mapped results, continue the solution based on the new mesh via a standard multiframe
restart from the last mapping substep (that is, the substep at which you initiated rezoning plus the
number of substeps required by the mapping operation).
Even if you have rezoned several times in the same domain, you can still perform a standard multiframe
restart.
You can restart from any substep at which the .rxxx file, .rdb file, and .ldhi load history file exist.
You need only specify the substep to restart; the program detects the necessary .rdb file, and the
.rst file to modify, then finds the corresponding load history information from the .ldhi file.
After the restart, the program deletes all .rdb, .rst, and .rxxx files on the substeps subsequent to
the specified restart substep.
This step concludes the rezoning process. At this point, you can repeat the rezoning process if you
wish, or perform postprocessing on the rezoned analysis results.

120

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Postprocessing Rezoning Results

4.11. Repeating the Rezoning Process if Necessary

More than one rezoning may be necessary if a mesh distortion still exists in the domain despite a previous rezoning. You can rezone a domain that you have already rezoned by selecting a region (or regions)
in the same domain at a different time and substep than that of the initial rezoning.
To rezone again, a region that you select for remeshing can be the same region that you selected in a
previous rezoning or a different region. As in the initial rezoning, you can use horizontal rezoning to
select more than one region for remeshing.
Although no theoretical limit on the number of allowable rezonings in the same domain exists, the
program allows up to 99 rezonings in a single job. In practice, two or three rezonings in a given domain
should be sufficient.
To perform another rezoning, simply repeat the rezoning process. No special command is necessary,
and the rezoning process remains the same.

4.11.1. File Structures for Repeated Rezonings

When multiple rezonings occur in the same domain during the same analysis, each creates a different
finite element model. To make restarting and rezoning from any substep possible, the program saves
the .rdb files for the initial model and for each rezoned model. Similarly, different .rst files are created
for each model to postprocess the results.
Following is the file structure after more than one rezoning has occurred:
File type

Rezone 0
(standard
run)

Rezone 1

Rezone 2

... Rezone 11

... Rezone nn

.rdb

.rdb

.rd01

.rd02

...

.rd11

...

.rdnn

.rxxx

.rxxx

.rxxx

.rxxx

...

.rxxx

...

.rxxx

...

.rs11

...

.rsnn

.ldhi

.ldhi
.rst

.rst

.rs01

.rs02

The maximum number of .rdb files and .rst files is 99. The .ldhi load history file contains information for all models created as a result of multiple rezonings.

4.12. Postprocessing Rezoning Results

For rezoning, the POST1 postprocessor is the primary postprocessing tool and most of its capability is
available. Animation is also available via the ANDATA macro. Output from the POST26 time-history
postprocessor (/POST26) is restricted to the data contained in any one results file.
Each time rezoning occurs, the mesh changes, so new .rdb and .rst files are necessary. For example,
if you have employed rezoning twice during the same analysis, the program writes the following results
files: .rst, .rs01, and .rs02, and the following database file: .rdb, .rd01, and .rd02. (For more
information, see File Structures for Repeated Rezonings (p. 121).) Most POST1 postprocessing operations
proceed seamlessly from one file to the next automatically. Therefore, do not delete the .rst or .rdb
file, nor any .rsnn or .rdnn files, until you have completed postprocessing.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

121

Rezoning
Displacement output for the rezoned model reports values as of the most recent rezoning, so the displacement will not seem to be continuous over multiple rezonings.

4.12.1. The Database Postprocessor

Before accessing the POST1 database postprocessor (/POST1), consider how the program should handle
postprocessing data files, as follows:
The program treats all postprocessing data files as a single file automatically (assuming that all necessary
files are present). If you exited the program after the solution phase of the analysis and now want to
perform postprocessing on your rezoning results, issue the /FILNAME command to specify the appropriate
jobname before entering the POST1 processor. This is the preferred method for postprocessing rezoning
results.
If you must examine only one postprocessing data file--because some files are very large or some are
missing, for example--resume (RESUME) the appropriate database to obtain the correct plot, then issue
the FILE command within POST1. (For example, to check the results in postprocessing data file .rs02,
resume database file .rd02.)

4.12.1.1. Listing the Rezoning Results File Summary

After entering the POST1 processor, it is helpful to issue the SET,LIST command.
For rezoning, the output list generated by the command includes an additional column indicating which
postprocessing data file contains each saved substep of each load step. The first use of the new file is
flagged with the word rezone to emphasize the change of file.
Following is an example output list from a SET,LIST command:
*****

INDEX OF DATA SETS ON RESULTS FILE *****

SET
TIME/FREQ
LOAD STEP
SUBSTEP CUMULATIVE
1 0.1000000
1
1
2
2 0.2000000
1
2
4
3 0.3500000
1
3
6
4 0.5500000
1
4
8
5 0.7500000
1
5
10
6 0.7500010
1
6
13
7 0.8750000
1
7
15
8 1.000000
1
8
17
...

FILE SUFFIX
rst
rst
rst
rst
rst
rs01
rezone
rs01
rs01

The output from a SET,LIST command is also useful for simply determining what information is available.
For non-rezoning runs, for example, you can select substeps for further study by load step and substep
number, time, or set number.
Although the program creates multiple postprocessing data files for rezoning, you can consider them
a single file for POST1 processing (assuming that the appropriate jobname is already specified) because
there is no need to access individual files directly. The program automatically detects the specific file
needed according to the SET command issued, after which most of the usual postprocessing command
capability (PLESOL, PRNSOL, etc.) is available.

4.12.1.2. Animating the Rezoning Results

Animation for rezoning is available via the ANDATA macro.
Assuming that the appropriate jobname is already specified, the program proceeds from one postprocessing data file to the next automatically, using the data in each to generate the animation.

122

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Rezoning Restrictions
During the macro's initial scan, the program stores the view location and size of every saved substep.
The program then combines the information to provide one fixed view (by internally setting the /FOCUS
and /DIST commands to fixed values).

4.12.1.3. Using the Results Viewer for Rezoning

The Results Viewer does not support results viewing across all rezoning results files automatically.
When you open one file, you can view only the results data set in that file. To view results in another
file, you must open the specific file manually. To open any .rsxx file in the Results Viewer (for example,
.rs01):
1. Select File>Open Results....
2. Specify the All File Types (*.*) option in the Files of type field.
3. Select the .rsxx results file of interest.

4.12.2. The Time-History Postprocessor

When using the POST26 time-history postprocessor (/POST26) on a model that you have rezoned, all
of the usual information is available. You must open each results file separately, however, because
POST26 works directly with the postprocessing data files and database files.
For each file that you want to open:
1. Reset (RESET) the postprocessor specifications to initial defaults, erase all defined variables, and clear
the data storage space.
2. Resume (RESUME) the appropriate database file. (For example, to check the results in postprocessing
data file .rs02, resume database file .rd02.)
3. Issue a FILE command to open the desired file.
Only the output information from the requested file is available for output at any given time because
elements and nodes that exist in one file do not always exist in another file.

4.13. Rezoning Restrictions

After you initiate the rezoning process (REZONE), and until you map the solved variables from the old
mesh to the new mesh (MAPSOLVE) at the conclusion of the rezoning process, you will lose your
rezoning work if you exit the solution processor (/SOLU) prematurely. This condition applies even if
you save (SAVE) before exiting and then resume (RESUME) after returning to the solution processor.
Issuing a REMESH,READ command (to import a generic new mesh) or a REMESH,SPLIT command (to
split an existing mesh) causes the mesh to lose associativity with its corresponding geometry. It is
therefore not possible to issue commands that perform geometry-based operations. (You can issue
commands that perform mesh-based operations, however.)
When using a generic new mesh created by a third-party application (REMESH,READ):
Mesh regions cannot intersect.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

123

Rezoning
If multiple regions are read in (REMESH,READ) via a single .cdb file, the nodes of these regions must be
encapsulated in a single NBLOCK environment, and the elements must be encapsulated in a single EBLOCK
environment.
For a 2-D analysis, you can issue the following remeshing commands in the same session on multiple
regions:
AREMESH,AMESH (to create a program-generated new mesh)
REMESH,READ (to import a generic new mesh)
REMESH,SPLIT (to split an existing mesh)
For a 3-D analysis, remeshing is possible only by importing a new mesh created by another application
(REMESH,READ).
The POST1 load-case combination commands (such as LCASE, LCDEF, LCOPER, and LCSEL) are not
available.
Initial state loading (INISTATE) is not available for rezoning.
Rezoning is not available in Distributed ANSYS.

4.14. Rezoning Examples

The following example simulations introduce you to rezoning:
4.14.1. Example: Rezoning Using a Program-Generated New Mesh
4.14.2. Example: Rezoning Using a Generic New Mesh
Also see Nonlinear Analysis of a 2-D Hyperelastic Seal Using Rezoning, Ring-Gear Forging Simulation
with Rezoning, and Hot-Rolling Structural Steel Analysis with 3-D Rezoning in the Technology Demonstration Guide.

4.14.1. Example: Rezoning Using a Program-Generated New Mesh

Following is an example simulation involving a sealing assembly problem. The example uses a programgenerated new mesh for the remeshing operation.
A rubber seal has an initial rectangular shape and consists of a hyperelastic material modeled with two
parameters as a Mooney-Rivlin model. A shaft has a circular cross section and is assumed to be rigid.
The shaft moves down vertically. The simulation plots the element strain in the y direction.
Given the initial input, the simulation terminates at substep 10 (TIME = 0.44) because of a mesh distortion.
After rezoning, which occurs at substep 7, the simulation concludes successfully with the new mesh.

124

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Rezoning Examples

4.14.1.1. Initial Input for the Analysis

This input results in a deformed mesh and causes the analysis to terminate:
/batch,list
/filnam,RznExample1
/prep7
/com define parameters
h=20
b=10
el=b/8
xc=17.5
yc=32.99038
rc=15
PilotMove= -13
c10=62.3584129
c01=-37.8485452
dd=1e-4
/com define element types, material, etc.
et,1,182
keyopt,1,3,2
keyopt,1,6,1
et,2,169
et,3,172
keyopt,3,9,0
keyopt,3,10,1
et,4,169
et,5,172
keyopt,5,9,0
keyopt,5,10,1
tb,hyper,1,,2,mooney
tbdata,1,c10,c01,dd
mp,mu,2,0.0
r,3
r,4
/com define geometry
k,1,xc,yc

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

125

Rezoning
k,2,xc,yc,yc
k,3,xc-rc,yc
k,4,0.0,0.0
k,5,2*rc,0.0
rect,0,b,0,h
circle,1,rc,2,3,360,1
/pnum,line,1
lplot
l,4,5
lplot
aplot
/com create solid elements
esize,el
mat,1
type,1
real,1
amesh,1
/com generate the 1st contact pair
mat,2
real,3
type,2
esize,h
lmesh,5,7
*get,PilotID,node,,num,max
PilotID=PilotID+1
nkpt,PilotID,1
tshap, pilo
e,PilotID
type,3
lsel,,,,2,3
nsll,,1
esln,,0
esurf
alls
/com generate the 2nd contact pair
real,4
type,4
lmesh,8
lsel,,,,8
esll
esurf,,reverse
alls
type,5
lsel,,,,1,2
nsll,,1
esln,,0
esurf
alls
/com apply boundary conditions and loads
d,PilotID,ux,0.0
d,PilotID,uy,PilotMove
d,PilotID,rotz,0.0
lsel,,,,4
nsll,,1
d,all,ux,0.0
alls
lsel,,,,8
nsll,,1
d,all,uy,0.0
alls
nlist

126

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Rezoning Examples
elist
dlist
/com check the contact definition
cncheck
finish
/solution
rescontrol,,all,1
pred,off
nlgeom,on
time,1
NSUBST,10,100,5
outres,all,all
solve
finish
/post1
set,1,6
prns,u,comp
prns,s,comp
prns, cont
finish

4.14.1.2. Rezoning Input for the Analysis

This input uses rezoning to remesh the deformed region and allow the analysis to proceed using the
new mesh:
/batch,list
/clear,nostart
/filnam,RznExample1
/solution
rezone,manual,1,6

! specify the substep to rezone

remesh,start
esel,,,,65,128
aremesh
lesize,10,,,8
lesize,31,,,10
lesize,18,,,6
amesh,2
esel,all
nsel,all
remesh,fini

! start remeshing operation

! select region to remesh
! create area for new mesh

elist
dlist

! check the new model, BC and loads

mapsolve,50

! map solutions

! create the new mesh

! finish remeshing operation

finish
/solution
antype,,restart
solve
finish

! restart

4.14.2. Example: Rezoning Using a Generic New Mesh

Following is an example simulation involving a heading assembly problem. The example uses an imported
generic mesh generated by another application for the remeshing operation.
The model represents an axisymmetric hollow hemisphere that pushes down a cylindrical workpiece.
The spherical ball and the grip die are modeled as rigid surfaces. Due to element distortion, the initial
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

127

Rezoning
run stops at t = 0.7875. Rezoning is applied at this time to achieve complete loading. The entire deformed
model at substep 4 is imported into ANSYS ICEM CFD, which generates a new mesh. After reading the
new mesh back in, the program creates the contact automatically when you issue the REMESH,FINISH
command.
The solid element used in the model is PLANE182 (using the B-Bar method with mixed u-P formulation).
CONTA171 and TARGE169 elements are also used. The material model used is a hyperelastic material
with a three-parameter OGDEN option.

4.14.2.1. Initial Input for the Analysis

This is the initial mesh:

This input results in a deformed mesh and causes the analysis to terminate at t = 0.7875 seconds:
/batch, list
/file,RznExample2
/prep7
h=4.6295
b=1.5
el=b/4
xc=0
yc=2.6295
rc=2.5
PilotMove= -yc
! ogden parameters
TB,HYPE,1,1,3,OGDE
TBTEMP,0
TBDATA,1,3.2084E-009,7.281,0.035198,3.0149,6.3712,2.0493
et,1,182
keyopt,1,3,0
keyopt,1,6,1
et,2,169
et,3,171
keyopt,3,9,0
keyopt,3,10,1
et,4,169
et,5,171
keyopt,5,9,0
keyopt,5,10,1
mp,mu,2,0.0
r,3
r,4
k,1,xc,yc
k,2,xc,yc,yc

128

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Rezoning Examples
k,3,rc,yc
k,4,0.0,0.0
k,5,rc+1,0.0
rect,0,b,0,h
circle,1,rc,2,3,90,1
/pnum,line,1
lplot
l,4,5
lplot
aplot
esize,el
mat,1
type,1
real,1
amesh,1
/pnum,elem,1
/pnum,node,1
/com the 1st contact pair
mat,2
real,3
type,2
esize,h
lmesh,5
lsel,,,,5
esll
esurf,,reverse
alls
*get,PilotID,node,,num,max
PilotID=PilotID+1
nkpt,PilotID,1
tshap, pilo
e,PilotID
type,3
lsel,,,,2,3
nsll,,1
esln,,0
esurf
alls
/com the 2nd contact pair
real,4
type,4
lmesh,6
lsel,,,,6
esll
esurf,,reverse
alls
type,5
lsel,,,,1,2
nsll,,1
esln,,0
esurf
alls
d,PilotID,ux,0.0
d,PilotID,uy,PilotMove
d,PilotID,rotz,0.0
lsel,,,,4
nsll,,1
d,all,ux,0.0
alls
lsel,,,,6
nsll,,1
d,all,uy,0.0
alls
/solution
pred,off
rescontrol,,all,1,
eresx,no
nlgeom,on
time,1
NSUBST,10,100,5
outres,all,all
solve

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

129

Rezoning
finish

Following is the total elastic strain along the Y axis at t = 0.7875. The element distortion is apparent
and causes the problem to diverge.

4.14.2.2. Exporting the Distorted Mesh as a CDB File

When the nonlinear analysis stops, reload the database at load step 1 and substep 4. Select all solid
elements and write out to a .cdb file. (Only solid elements can be read in later when you are ready to
generate the new mesh.)
/clear,nostart
/file,RznExample2
/solu
rezone,manual,1,4
eplot
etlist
ESEL,S,TYPE,,1
cdwrite,db,RznExample2,cdb
finish

!
!
!
!
!
!
!

reload the database

enter solution environment
start rezoning at load step1, substep 4
plot the elements
list the element types
select only the elements of type 1(solids)
write out the selected elements to a CDB file

With the deformed mesh corresponding to load step 1, substep 4 is shown next. ANSYS ICEM CFD uses
the boundary segments of this mesh next to generate the new mesh. The nodal discretization at the
boundary remains same for both the old and the new mesh.

130

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Rezoning Examples

Next, the total elastic strain along the Y axis for this mesh is shown. This is one of the state variables
which is transferred to the new mesh when mapping solved node and element solutions from the original mesh to the new mesh (MAPSOLVE).

4.14.2.3. Importing the File into ANSYS ICEM CFD and Generating a New Mesh
At this stage of the rezoning process, start ANSYS ICEM CFD and read in the .cdb file. (Reminder: As
indicated in Exporting the Distorted Mesh as a CDB File (p. 130), only solid elements can be read in.)
Generate the new .cdb file as follows:
1. Import the .cdb file in ANSYS ICEM CFD as mesh (File Menu> Import Mesh> From Ansys).
2. Extract triangulated (STL) geometry from the mesh (Edit Menu> Mesh to Facets)

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

131

Rezoning
3. Set the global maximum element size in the order of the element size that you require (Mesh> Global
Mesh Setup> Global Mesh Size> Max Element).
4. Build the topology (Geometry> Repair geometry> Build topology)
5. Select the Respect line elements and Protect given line elements options (Mesh> Global Mesh Setup>
Shell Meshing Params).
6. Compute the new mesh (Mesh> Compute mesh> Surface mesh only> Mesh type: All Quad > Compute).
7. Select the Solve Options tab and write the input file. Do not include the bar elements.
8. Rename the new input file as a .cdb file.
The new mesh obtained from ANSYS ICEM CFD is shown here. Notice that the boundary discretization
remains the same as that of the old mesh.

4.14.2.4. Rezoning Using the New CDB Mesh

Continue rezoning with the new mesh (.cdb file) and restart the analysis, as follows:
/clear,nostart
/file,RznExample2
/solu
rezone,manual,1,4
remesh,start
remesh,read,RznExample2,cdb,rege
remesh,finish
mapsolve,500,pause
finish

!
!
!
!
!
!
!

reload the database

enter solution environment
start rezoning from load step 1, substep 4
start remeshing
read in the new mesh (CDB file)
finish remeshing, autogenerate contacts
do state variable mapping and equilibriation

After the MAPSOLVE command has executed (mapping the solved node and element solutions from
the original mesh to the new mesh), the total elastic strains along Y for the new mesh appears. Notice
that some expected nodal realignment has occurred in the new mesh.

132

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Rezoning Examples

Restart the problem. The solution to progresses to t = 1s.

/clear,nostart
/file,RznExample2
/solu
antype,,restart
solve
finish

!
!
!
!

reload the database

enter solution environment
multiframe restart
solve the problem

Allow the analysis to complete. Following is a plot of the total elastic strain along the Y direction:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

133

134

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 5: 2-D to 3-D Analysis

It is sometimes necessary to extend 2-D nonlinear finite element results to a corresponding extruded
3-D body such that the solution can continue based on the 3-D model.
For example, in a tire analysis, the 2-D axisymmetric tire-inflation results must be extended to an extruded
3-D tire model before a footprint analysis can be performed.
The 2-D to 3-D analysis capability extrudes (EEXTRUDE) a 2-D deformed mesh to a new 3-D mesh. The
program updates the database as necessary, generates contact elements if needed, and transfers
boundary conditions, loads and nodal temperatures from the 2-D mesh to the extruded 3-D mesh. All
solved variables (node and element solutions) are then mapped (MAP2DTO3D) to the new 3-D mesh
automatically.
The following topics related to 2-D to 3-D analysis are available:
5.1. Benefits of 2-D to 3-D Analysis
5.2. Requirements for a 2-D to 3-D Analysis
5.3. Overview of the 2-D to 3-D Analysis Process
5.4. Performing a 2-D to 3-D Analysis
5.5. 2-D to 3-D Analysis Restrictions

5.1. Benefits of 2-D to 3-D Analysis

Mapping solution variables from a 2-D mesh to an extruded 3-D mesh allows you to perform a multistage
nonlinear analysis. For example, in a tire analysis, 2-D axisymmetric inflation results are required for a
subsequent 3-D footprint analysis.
In effect, you can continue with a 3-D analysis on the extruded body after mapping solution results
from the 2-D body.

5.2. Requirements for a 2-D to 3-D Analysis

All multiframe restart files must be available.
Following are the 2-D elements and the materials, loads, and boundary conditions that can be used for
2-D to 3-D analysis:
Support Category
Solid elements

Requirements
PLANE182 -- Plane strain and axisymmetric states of stress only
PLANE183 -- Plane strain and axisymmetric states of stress only
Enhanced strain formulation is not supported.

Contact elements

TARGE169
CONTA171 and CONTA172 with any of the following valid KEYOPT settings:
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

135

2-D to 3-D Analysis

Support Category

Requirements
KEYOPT (1) = 0
KEYOPT (2) = 0, 1, 3, 4
KEYOPT (3) = 0
KEYOPT (4) = 0, 1, 2
KEYOPT (5) = 0, 1, 2, 3, 4
KEYOPT (7) = 0, 1, 2, 3
KEYOPT (8) = 0
KEYOPT (9) = 0, 1, 2, 3, 4
KEYOPT (10) = 0, 2
KEYOPT (11) = 0
KEYOPT (12) = 0, 1, 2, 3, 4, 5, 6
KEYOPT(14) = 0
TARGE170
CONTA173 and CONTA174 with any of the following valid KEYOPT settings:
KEYOPT (1) = 0
KEYOPT (2) = 0, 1
KEYOPT (4) = 0, 2
KEYOPT (5) = 0, 1, 2, 3, 4
KEYOPT (7) = 0, 1, 2, 3
KEYOPT (8) = 0
KEYOPT (9) = 0, 1, 2, 3, 4
KEYOPT (10) = 0, 2
KEYOPT (11) = 0
KEYOPT (12) = 0, 1, 2, 3, 4, 5, 6
KEYOPT(14) = 0

Contact pair
behavior

Rigid-flexible -- 2-D target elements (line) are extruded in 3-D as tangent

planes or as annuli in axisymmetric extrusion. The extrusion is always
planar in plane strain extrusion.
Pilot nodes in 2-D are retained as pilot nodes in 3-D.
Flexible-flexible contact.
Self-contact.

Materials

All structural materials are supported except for CAST (cast iron), CONCR
(concrete), MPLANE (microplane), SMA (shape memory alloy), and SWELL
(swelling).
For materials associated with user-defined state variables (TB,STATE), it
may be necessary to issue the MAPVAR command. The defined size of
state variables must be such that full 3-D descriptions of stress and strain
tensors can be accommodated (that is, the defined stress and strain
tensors must be able to possess six independent components).

136

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Overview of the 2-D to 3-D Analysis Process

Support Category
Loads and
boundary conditions (BCs)

Requirements
Displacements, forces, pressures, and nodal temperatures (BF,TEMP).
Loads applied on surface-effect elements (such as SURF153) are not
supported.
Tabular displacements and forces are supported.
Tabular pressures that are functions of time only are supported.

Files

.rdb, .rst, .rxxx, .ldhi, .ist, and .cdb.

The EEXTRUDE command documentation shows the topological mapping of 2-D elements (solid, contact,
and target) into specific 3-D elements.

5.3. Overview of the 2-D to 3-D Analysis Process

The 2-D to 3-D analysis process involves the following general tasks:
1. Determine the load step and substep in the 2-D model at which extrusion should occur.
2. Initiate the 2-D to 3-D analysis process.
3. Perform the extrusion and map boundary conditions and load.
4. Map displacements and solution variables automatically from the 2-D mesh to the new 3-D mesh.
5. Continue the 3-D analysis using the data in the .ist files (INISTATE,READ) along with new applied
loads and boundary conditions (if applicable) on the deformed mesh.
The entire 2-D mesh is used for the extrusion process; therefore, only one MAP2DTO3D command
block can be used at a given loadstep and substep.
The following topics related to understanding the general 2-D to 3-D analysis process are available:
5.3.1. Overview of the 2-D to 3-D Analysis Process Flow
5.3.2. Key Commands Used in 2-D to 3-D Analysis

5.3.1. Overview of the 2-D to 3-D Analysis Process Flow

The following flowchart illustrates the general 2-D to 3-D analysis process:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

137

2-D to 3-D Analysis

Figure 5.1: 2-D to 3-D Process Flow

For information about the commands shown in the flowchart, see Key Commands Used in 2-D to 3-D
Analysis (p. 138)or the documentation for a given command in the Command Reference.

5.3.2. Key Commands Used in 2-D to 3-D Analysis

Following is a description of key commands used in the 2-D to 3-D analysis process and provides supplemental information to the 2-D to 3-D analysis process flowchart.
Command

Description

2-D to 3-D Analysis Comments

/CLEAR

Clears the database

Always clear the database first, before reentering the solution processor (/SOLU) and starting the 2-D to 3-D analysis
process.

MAP2DTO3D,START
Initiates the
2-D to 3-D
analysis

When you initiate 2-D to 3-D analysis, the program verifies

that the necessary files (.rdb, .rst, .rxxx, and .ldhi)
exist for the specified substep and rebuilds the data environment at that substep.
All nodes are updated to the deformed geometry in
preparation for extrusion.

138

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Performing a 2-D to 3-D Analysis

Command

Description

2-D to 3-D Analysis Comments

To prevent the loads and boundary conditions from
being generated on the 3-D model, specify the
MAP2DTO3D commands NOBC option.

EEXTRUDE

Extrudes
the 2-D
mesh to a
3-D mesh.
Deletes the
2-D rigid
targets and
regenerates
the 3-D rigid targets.

The entire 2-D mesh is selected automatically for extrusion. Partial mesh extrusion is not supported.
In the extrusion phase of a 2-D to 3-D analysis, the
command generates the 3-D contact elements. The
command also regenerates the loads, boundary conditions, and temperatures for the extruded 3-D mesh.

Maps the
loads, temperatures,
and boundary conditions from
the 2-D
mesh to
the corresponding 3D mesh.
Maps
MAP2DTO3D,FIN- node and
element soluISH
tions

After extrusion, this command maps the solved nodal and

element solutions from the 2-D (distorted) mesh to the
extruded 3-D mesh.

5.4. Performing a 2-D to 3-D Analysis

The following steps describe the general process for performing a 2-D to 3-D analysis:
5.4.1. Step 1: Determine the Substep to Initiate
5.4.2. Step 2: Initiate the 2-D to 3-D Analysis
5.4.3. Step 3: Extrude the 2-D Mesh to the New 3-D Mesh
5.4.4. Step 4: Map Solution Variables from 2-D to 3-D Mesh
5.4.5. Step 5: Perform an Initial-State-Based 3-D Analysis

5.4.1. Step 1: Determine the Substep to Initiate

It is typical in a 2-D to 3-D analysis to use the results at the end of the 2-D analysis as the starting point
for the 3-D extrusion. You can, however, choose any substep to initiate the 2-D to 3-D analysis process
if the restart files for that substep exist.
When you have determined a suitable substep, proceed to Step 2: Initiate the 2-D to 3-D Analysis (p. 140).

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

139

2-D to 3-D Analysis

5.4.2. Step 2: Initiate the 2-D to 3-D Analysis

2-D to 3-D analysis is based on a new model with a higher dimension (3-D), extruded from a specific
substep of a lower dimensional (2-D) model solution; therefore, the 2-D to 3-D analysis process must
begin with a clean database.
To initiate a 2-D to 3-D analysis:
1. Clear the database (/CLEAR,NOSTART).
2. Reenter the solution processor (/SOLU).
3. Initiate 2-D to 3-D analysis, specifying the load step and substep at which 2-D to 3-D analysis should
occur (MAP2DTO3D,START,LDSTEP,SBSTEP or MAP2DTO3D,START,LDSTEP,SBSTEP,NOBC).
Proceed to Step 3: Extrude the 2-D Mesh to the New 3-D Mesh (p. 140).

5.4.3. Step 3: Extrude the 2-D Mesh to the New 3-D Mesh
Extrude the 2-D mesh to the 3-D mesh (EEXTRUDE).
The following figures illustrate how extrusion generates 3-D meshes from a plane strain and an
axisymmetric 2-D mesh, respectively:
Figure 5.2: 2-D Plane Strain to 3-D Solid Extrusion

Figure 5.3: Axisymmetric to 3-D Solid Extrusion

Extrusion preserves the 2-D mesh topology on the generating plane and replicates it through the depth
of the extruded body.

140

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

2-D to 3-D Analysis Restrictions

Contacts, Targets, and Boundary Conditions
The extrusion process generates the 3-D contact and target elements with the solid elements. All 2D rigid target elements are deleted, and corresponding 3-D rigid target elements are generated.
When you issue the MAP2DTO3D,START,LDSTEP,SBSTEP command followed by the EEXTRUDE
command, boundary conditions, pressure loads, applied nodal forces, applied nodal displacements,
and applied nodal temperatures are transferred automatically from the 2-D mesh to the corresponding
entities in the extruded 3-D mesh. (If you specify the MAP2DTO3D commands NOBC option, none
of those attributes except for the nodal temperatures transfer to the extruded 3-D mesh.)
After extrusion, proceed to Step 4: Map Solution Variables from 2-D to 3-D Mesh (p. 141).

5.4.4. Step 4: Map Solution Variables from 2-D to 3-D Mesh

Issue the MAP2DTO3D,FINISH command to map the solved node and element solutions automatically
from the original 2-D mesh to the new 3-D mesh.
No residual balancing is done and only the solution variables are mapped from the 2-D mesh to the
extruded 3-D mesh.
After mapping solution variables to the 3-D mesh, proceed to Step 5: Perform an Initial-State-Based 3D Analysis (p. 141).

5.4.5. Step 5: Perform an Initial-State-Based 3-D Analysis

After successfully mapping the solved variables from the old 2-D mesh to the new 3-D mesh, you can
continue your 3-D analysis by reading in the 2-D to 3-D results as initial state conditions (INISTATE,READ)
and adding new loads for the 3-D analysis.
For more information, see Initial State in the Basic Analysis Guide.

5.5. 2-D to 3-D Analysis Restrictions

The following limitations apply to a 2-D to 3-D analysis:
You cannot use remeshing commands within the 2-D to 3-D analysis environment (after MAP2DTO3D,START
and before MAP2DTO3D,FINISH). The incompatible commands include REMESH,READ, REMESH,SPLIT,
and AREMESH.
Axisymmetric extrusion with zero offset of nodes about the Y axis is not supported.
2-D plane stress (KEYOPT(3) = 0 or KEYOPT(3) = 3) and generalized plane strain (KEYOPT(3) = 5) elements
are not supported.
Layered or reinforced solid elements are not supported.
Enhanced strain element technology is not supported.
2-D to 3-D analysis is not available in Distributed ANSYS.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

141

142

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 6: Cyclic Symmetry Analysis

A component or assembly is cyclically symmetric if it has a correspondence in form or arrangement of
parts (that is, repetitive patterns) centered around an axis.
A fan wheel is a typical example of a cyclically symmetric structure, as is a spur gear. Another example
is this model of a hydro rotor:
Figure 6.1: Hydro Rotor -- Model of a Cyclically Symmetric Structure

The following cyclic symmetry analysis topics are available:

6.1. Understanding Cyclic Symmetry Analysis
6.2. Cyclic Modeling
6.3. Solving a Cyclic Symmetry Analysis
6.4. Postprocessing a Cyclic Symmetry Analysis
6.5. Limitations in a Cyclic Symmetry Analysis
6.6. Example Modal Cyclic Symmetry Analysis
6.7. Example Buckling Cyclic Symmetry Analysis
6.8. Example Harmonic Cyclic Symmetry Analysis
6.9. Example Magnetic Cyclic Symmetry Analysis

6.1. Understanding Cyclic Symmetry Analysis

If a structure exhibits cyclic symmetry, you can perform an automated static, modal, harmonic, or
buckling analysis on the structure.
The following topics help you to understand cyclic symmetry analysis:
6.1.1. How the Program Automates a Cyclic Symmetry Analysis
6.1.2. Commands Used in a Cyclic Symmetry Analysis

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

143

Cyclic Symmetry Analysis

6.1.1. How the Program Automates a Cyclic Symmetry Analysis

A cyclic symmetry analysis conserves time and CPU resources and allows you to view analysis results
on the entire structure. The program automates cyclic symmetry analysis by:
Solving for the behavior of a single symmetric sector (part of a circular component or assembly)
Using the single-sector solution to construct the response behavior of the full circular component or assembly (as a postprocessing step).
For example, by analyzing a single 10 sector of a 36-blade turbine wheel assembly, you can obtain the
complete 360 model solution via simple postprocessing calculations. Using twice the usual number of
degrees of freedom (DOFs) in this case, the single sector represents a 1/18th part of the model.

6.1.2. Commands Used in a Cyclic Symmetry Analysis

The most important command in an automated cyclic symmetry analysis is CYCLIC, which initiates a
cyclic analysis and configures the database accordingly. The command automatically detects cyclic
symmetry model information such as edge components, the number of sectors, the sector angles, and
the corresponding cyclic coordinate system.
The ANTYPE command specifies the analysis type (for example, static, modal or buckling), and the
SOLVE command obtains the cyclic solution.
Other cyclic-specific commands include:
CYCOPT for specifying solution options
/CYCEXPAND for graphically expanding displacements, stresses and strains of a cyclically symmetric
model
CYCPHASE for determining minimum and maximum possible result values from frequency couplets during
postprocessing (/POST1).
Depending upon the type of cyclic symmetry analysis that you want to perform and your specific needs,
it may be necessary to issue other commands. For example:
In a prestressed modal cyclic symmetry analysis, you must issue the PSTRES,ON command during the
static portion of the analysis to apply the prestress effects. (The PSTRES command is not needed if the
linear perturbation analysis procedure is used.)
During postprocessing, you may want to issue the ANCYC command to apply a traveling wave animation
to your cyclic model.
The sections of this document describing various cyclic symmetry analyses mention such commands
as necessary. For more information, see Solving a Cyclic Symmetry Analysis (p. 150).

6.2. Cyclic Modeling

This section describes how to set up a cyclic sector model, discusses important considerations for edge
component pairs, and shows how to verify the cyclically symmetric model.
The following cyclic modeling topics are available:
6.2.1.The Basic Sector

144

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Cyclic Modeling
6.2.2. Edge Component Pairs
6.2.3. Model Verification (Preprocessing)

6.2.1. The Basic Sector

A cyclic symmetry analysis requires that you model a single sector, called the basic sector. A proper
basic sector represents one part of a pattern that, if repeated N times in cylindrical coordinate space,
yields the complete model, as shown:
Figure 6.2: A Basic Sector in a Cyclically Symmetric Structure

Define a basic sector model that is cyclically symmetric in any global or user-defined cylindrical coordinate
system. (For information about creating a model, see the Modeling and Meshing Guide.)
The angle (in degrees) spanned by the basic sector should be such that N = 360, where N is an integer.
The basic sector can consist of meshed or unmeshed geometry. The program allows user-defined
coupling and constraint equations on nodes that are not on the low or high edges of the cyclic sector.
(For more information about the cyclic sector's low and high edges, see Edge Component Pairs (p. 146).)
If meshed, the basic sector may have matching (as shown in Figure 6.3: Basic Sector Definition (p. 145))
or unmatched lower and higher angle edges. Matching means that corresponding nodes exist on each
edge, offset geometrically by the sector angle . The edges may be of any shape and need not be "flat"
in cylindrical coordinate space. For more information, see Identical vs. Dissimilar Edge Node Patterns (p. 149).
Figure 6.3: Basic Sector Definition

Low Component Nodes

High Component Nodes

Z
Y
Sector

angle

X
CSYS = 1

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

145

Cyclic Symmetry Analysis

6.2.2. Edge Component Pairs

The cyclic sector has two edges that align along the surfaces of cyclic symmetry. The edge having the
algebraically lower in the R-Theta (cylindrical) coordinate system is called the low edge and the one
having the higher is called the high edge. The angle between the two successive surfaces of cyclic
symmetry is called the sector angle.
When setting up a cyclic symmetry analysis, the CYCLIC command defines edge components automatically, assigning them a default root name of CYCLIC.
Optionally, you can use the CYCLIC command to define the edges and the component names manually.
If you do so, you must specify a root name for the sector low- and high-edge components (line, area,
or node components). A root name that you specify can contain up to 11 characters. The naming convention for each low- and high-edge component pair is either of the following:
name_mxxl, name_mxxh
(potentially matched node patterns)
name_uxxl, name_uxxh
(potentially unmatched node patterns)
The name value is the default (CYCLIC) or specified root name, and xx is the component pair ID
number (sequential, starting at 01).

6.2.2.1. CYCOPT Auto Detection Tolerance Adjustments for Difficult Cases

If the CYCLIC command fails to auto-detect the edges of your cyclic sector, it may be possible to adjust
1 or 2 tolerances to remedy the situation by adjusting the ANGTOL and/or FACETOL values of the
CYCOPT command. The most effective way to correct auto-detection is usually by changing the ANGTOL
value; however, for more difficult cases from FEA models you may need to change FACETOL to achieve
auto detection.
When using the CYCOPT command, face tolerance, or FACETOL, automatically defaults to 15 degrees.
Face tolerance applies only to auto-detection from node/element models that are already meshed, and
are not solid models. The default face tolerance will accommodate most models, unless they include
extreme angles or complex model geometry, both of which can cause surface nodes to be excluded.
This problem, and possible solutions are illustrated in this section using the cyclic model shown in
Figure 6.4: Full Cyclic Model (p. 147).

146

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Cyclic Modeling
Figure 6.4: Full Cyclic Model

Figure 6.5: Cyclic Sector (p. 147) shows a cyclic sector from the model above. As you can see, this sector
model is leaning heavily in the direction of theta. The low and high boundaries in this figure have been
auto detected correctly using the default FACETOL value of 15 deg. This successful auto detection of
these boundaries can be seen in Figure 6.6: Successful Auto Detection with Default FACETOL = 15
Deg (p. 148).
Figure 6.5: Cyclic Sector

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

147

Cyclic Symmetry Analysis

Figure 6.6: Successful Auto Detection with Default FACETOL = 15 Deg

If you reset the FACETOL values in the figure above from 15 degrees to 30 degrees, auto detection
fails because the logic treats the left and bottom sides of the sector as a single face, and the top and
right sides as a second large face. This is because the bottom-left edge and top-right edge of this
model both contain a dihedral angle greater than 150 degrees (180-30), but less than 165 degrees (18015). These large angles add to the possibility of auto-detection errors.
If you start with a model tilted more than the sector in Figure 6.6: Successful Auto Detection with Default
FACETOL = 15 Deg (p. 148), it may fail at the default tolerance of 15 degrees. This may require you to
reduce FACETOL to 10 or even 5 degrees to get a successful result.
A FACETOL value that is set too low can also result in failure. A FACETOL value that is too low can
cause edges not to be detected along element boundaries on smooth surfaces. The valid range of FACETOL is model and mesh dependent, and may be dictated by a single edge shared between 2 elements.
Whenever auto detect fails for an element model, save the node groups for each element face that you
were working with as node components NPF_001, NPF_002, etc. for diagnostic purposes. Each node
component should represent exactly one face of the cyclic sector (cyclic boundary or not).
In Figure 6.7: Auto Detection Failure Due to Large Face Tolerance (p. 149) , NPF_001 is clearly too large.
As you can see from the nodes, the group has leaked across the lower left edge. This indicates that
FACETOL is too large for the given dihedral angle.

148

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Cyclic Modeling
Figure 6.7: Auto Detection Failure Due to Large Face Tolerance

6.2.2.2. Identical vs. Dissimilar Edge Node Patterns

Automated Matching
The AMESH and VMESH commands perform automated matching. All other meshing-operation
commands (for example, VSWEEP) do not.
If you specify a meshing operation other than AMESH or VMESH, ensure that node and element
face patterns match, if desired. The CYCLIC command output indicates whether each edge-component pair has or can produce a matching node pair.
To ensure the most accurate solution, it is preferable to have identical node and element face patterns
on the low and high edges of the cyclic sector. If you issue the CYCLIC command before meshing the
cyclic sector (via the AMESH or VMESH command only), the mesh will have identical node and element
face patterns on the low and high edges if possible. In this case, all entities must be meshed together
using one meshing command.
The program allows dissimilar node patterns on the low and high edges of the cyclic sector, useful
when you have only finite-element meshes for your model but not the geometry data necessary to
remesh it to obtain identical node patterns. In such cases, it is possible to obtain solution (SOLVE) results,
although perhaps at the expense of accuracy. A warning message appears because results may be degraded near the cyclic sector edges.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

149

Cyclic Symmetry Analysis

6.2.2.3. Unmatched Nodes on Edge-Component Pairs

Unmatched nodes on the low- and high-edge components produce approximate cyclic symmetry
solutions (as compared to matched-node cases). The program uses an unmatched-node algorithm
(similar to that of the CEINTF command) to connect dissimilar meshes.
In unmatched cases, the eigenvector shapes exhibit discontinuity across segment boundaries when
expanded (via the /CYCEXPAND command). The discontinuity is an expected behavior; in the expansion
process, the low edge of sector 2 lies adjacent to the high edge of sector 1, and so on throughout the
full 360.
For information about expanding the solution results of a cyclic symmetry analysis, see Expanding the
Cyclic Symmetry Solution (p. 167).

6.2.2.4. Identifying Matching Node Pairs

To identify matching node pairs, you can issue a *STATUS command to list the cyclic parameter array
Name_xref_n (where Name is the root name of the low- and high-edge components specified via the
CYCLIC command).
The cyclic parameter array is generated internally during element plotting with cyclic expansion activated
(/CYCEXPAND,,ON).
In the cyclic parameter array listing, the matching node pairs appear as a pair of node numbers with
the low-edge node number having a negative value.

6.2.3. Model Verification (Preprocessing)

If the CYCLIC command's default automatic detection capability accepts your model for cyclic analysis,
the program will have already verified the following two essential conditions for a cyclic analysis:
When your model rotates by the cyclic angle about the local Z axis of the cyclic coordinate system, the
edges identified as "low" occupy the same space as those identified by "high" prior to the rotation.
The cyclic angle divides evenly into 360.
If you specify edge components and cyclic quantities manually, you must verify the two conditions
yourself.

6.3. Solving a Cyclic Symmetry Analysis

The program solves for the full cyclically symmetric model using the basic sector model that you have
set up during preprocessing with the appropriate boundary conditions, loading, and any coupling and
constraint equations. (For more information, see Cyclic Modeling (p. 144).)
This section provides specific information for obtaining the solution to various types of cyclic symmetry
analyses and covers the following topics:
6.3.1. Understanding the Solution Architecture
6.3.2. Solving a Static Cyclic Symmetry Analysis
6.3.3. Solving a Modal Cyclic Symmetry Analysis
6.3.4. Solving a Linear Buckling Cyclic Symmetry Analysis
6.3.5. Solving a Harmonic Cyclic Symmetry Analysis
6.3.6. Solving a Magnetic Cyclic Symmetry Analysis
6.3.7. Database Considerations After Obtaining the Solution
150

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Solving a Cyclic Symmetry Analysis

6.3.8. Model Verification (Solution)

6.3.1. Understanding the Solution Architecture

At the solution (SOLVE) stage of a cyclic symmetry analysis, the program applies the appropriate cyclic
symmetry boundary conditions for each nodal-diameter solution requested (via the CYCOPT command)
and solves. The program performs each nodal-diameter solution as a separate load step.
The following solution architecture topics are available for cyclic symmetry analysis:
6.3.1.1.The Duplicate Sector
6.3.1.2. Coupling and Constraint Equations (CEs)
6.3.1.3. Non-Cyclically Symmetric Loading

6.3.1.1. The Duplicate Sector

The architecture of the cyclic symmetry solution process depends upon how the compatibility and
equilibrium conditions of the cyclic sector are enforced in the matrix-solution process. The two most
common solution methods are Duplicate Sector and Complex Hermitian. For faster performance, the
program uses the Duplicate Sector method.
During the solution stage, the program generates a duplicate sector of elements at the same geometric
location as the basic sector. (Duplicate sector creation occurs automatically and transparently.) The
program applies all loading, boundary conditions, and coupling and constraint equations present on
the basic sector to the duplicate sector.

6.3.1.2. Coupling and Constraint Equations (CEs)

The program enforces cyclic symmetry compatibility conditions for each nodal-diameter solution via
coupling and/or constraint equations (CEs) connecting the nodes on the low- and high-edge components
on the basic and duplicate sectors. The program deletes the coupling and/or constraint equations after
each nodal-diameter solution, preserving any internal coupling and constraint equations that you may
have defined on the basic sector for subsequent analyses. The constraint equations for edge-component
nodes have the form shown in Equation 6.1 (p. 152).

Note
Internal CEs cannot be listed or viewed after a cyclic symmetry solution.
During the solution stage of a cyclic symmetry analysis, the program automatically rotates the nodal
coordinate systems of all nodes on the low and high sector edges to be parallel with the cyclic coordinate
system.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

151

Cyclic Symmetry Analysis

Figure 6.8: Connecting Low and High Edges of Basic and Duplicate Sectors

k
k

k
k

wo vf
o
wo v e

h gir f
h g ir e

(6.1)

where,
k = Harmonic index -- (0,1,2,,N / 2) when N is even, (0,1,2,,(N-1) / 2) when N is odd. (N is an integer
representing the number of sectors in 360.)
= Sector angle (2 / N)
U = Vector of displacement and rotational degrees of freedom
UALow represents the basic sector low side edge
UAHigh represents the basic sector high side edge
UBLow represents the duplicate sector low side edge
UBHigh represents the duplicate sector high side edge
The equation is a function of harmonic index k generating different sets of constraint equations for
each harmonic index. Therefore, for each harmonic index solution requested, the program creates the
appropriate constraint equations automatically, connects the edge-component nodes on basic sector
A and duplicate sector B, and solves.
Constraint equations that tie together the low and high edges of your model are generated from the
low- and high-edge components, and nowhere else. You should verify that automatically detected
components are in the correct locations and that you are able to account for all components; to do so,
you can list (CMLIST) or plot (CMPLOT) the components.

6.3.1.3. Non-Cyclically Symmetric Loading

A load is non-cyclic when it varies between sectors and involves at least one harmonic index greater
than zero. Support is available for cyclic analyses that have some combination of cyclic and non-cyclic
loading.
For non-cyclic loading, the program considers the arbitrary forces acting on the full system as the sum
of a finite number of spatial Fourier harmonics. The program analyzes the structure for each spatial
harmonic index by applying constraint equations between the basic sector and duplicate sector. For
each spatial harmonic of force, the program solves a corresponding equation, then expands and sums
152

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Solving a Cyclic Symmetry Analysis

the calculated harmonics of the response to give the response for each substructure. For more information, see Cyclic Symmetry Transformations in the Mechanical APDL Theory Reference.
Table 6.1: Valid Non-Cyclically Symmetric Loads
Non-Cyclic
Load Type

Commands

Loads

Comments

Nodal Loads

F, K

FX, FY, and FZ constraints follow sector-restricted loading

(CYCOPT,LDSECT,n where n >
0).

HFLOW is blocked for sectorrestricted loading but can follow afterwards.

Surface
Loads

PRES follows sector-restricted

loading (CYCOPT,LDSECT,n
where n > 0).

SF,
SFA,
SFE,
SFL

All other nodal loads are

blocked for sector-restriced
loading.
CONV is blocked for sector-restriced loading but can follow
afterwards.
All other surface loads are
blocked for sector-restriced
loading.

Inertia Loads

ACEL,
DOMEGA,
CMDOMEGA,
CMOMEGA,
OMEGA

Apply to all sectors. (Not affected by sector-restriction via

CYCOPT,LDSECT,n where n >
0.)

May require harmonic index 0

and/or 1 only.

Default action in global X, Y,

and Z on all sectors.

6.3.1.3.1. Specifying Non-Cyclic Loading

Specify non-cyclically symmetric loading via the CYCOPT command' s LDSECT (load-on-sector) value.
A value greater than 0 (the default, indicating that the loads are identical on all sectors) restricts subsequently defined force loads and surface loads to the specified sector. The restriction remains in effect
until you change or reset it.
When non-cyclic loading applies, the program creates or modifies the required SECTOR tabular
boundary condition (BC) data to apply on the appropriate sector. Therefore, it is not necessary to manipulate tables for situations where the applied BC is not a function of other tabular BC variables such
as TIME, X, Y, Z, and so on.
If a SECTOR-varying table exists on an entity-BC combination (for example, node 17 FZ) and you enter
another value for the same entity-BC combination (perhaps specifying a different sector on which to
apply the load), the following conditions occur:
The program modifies the existing table to accommodate the new specification.
The table cannot reference any other independent variable (for example, TEMP). You must manually define
any BC table requiring more than one independent variable.
If a table exists for an entity-BC combination and you enter another table for the same entity-BC combination, but the table does not reference SECTOR, the new table reference replaces the existing one.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

153

Cyclic Symmetry Analysis

During preprocessing, all tabular BCs referencing SECTOR list table names only. During solution or
postprocessing, all tabular BCs referencing SECTOR list per sector as they would be applied when solving
(SOLVE).
Any tabular data X, Y, or Z variation applied to a cyclic model may not be applied in the same manner
in which such a variation would occur for an equivalent full model (the exception being a variation in
the axial direction). For example, if a tabular value of a nodal force is applied as function of the tabular
variable Y, the program applies it to the designated cyclic sectors using values based upon the Y values
of the basic sector only.
A given high-edge node is usually the same location in the structure as the corresponding low-edge
node of the adjacent sector; therefore, it is necessary to apply constraints consistently. Note that inconsistent constraints are impossible to satisfy if the solution remains cyclic. The results can be unpredictable.
If a high (or low)-edge DOF has a constant (non-tabular) constraint, and the corresponding low (or
high)-edge DOF is unconstrained, the program copies the constraint to the opposite edge. If a high (or
low)-edge DOF has a tabular constraint, and the corresponding low (or high)-edge DOF is unconstrained,
the program stops the solution with an error message. If a high-low corresponding pair of DOF are both
constrained in any manner, the program assumes that you have specified constraints in a consistent
manner.
One warning is issued the first time this is done for a given SOLVE operation.
Because edge nodes are rotated into the cyclic coordinate system during solution, any applied displacements or forces on sector edges will be in the cyclic coordinate system.
Example 6.1: Non-Cyclic Loading via Automatically-Defined Tabular Load
CYCOPT,LDSECT,1 ! LOADS ON SECTOR 1 ONLY
SFL,ALL,PRES,10000
For DOF constraints, force loads, and body forces, any non-tabular load is cyclic. Any
tabular load that does not reference the variable SECTOR is cyclic. The program assumes
any tabular load referencing SECTOR to be non-cyclic (although it could be identical on
all sectors).
Example 6.2: Non-Cyclic Loading via User-Defined Tabular Load
*DIM,S1PRES,TABLE,5,1,1,SECTOR
*SET,S1PRES(1,0,1),1,2,3,4,5
*SET,S1PRES(1,1,1),10000,0,0,0,0

! PRESSURE ON SECTOR 1 ONLY

SFL,ALL,PRES,%S1PRES%
When combined with other independent variables, SECTOR can be in positions 1, 2, or
3 only. Other independent variables operate as they do for non-cyclic data. (Think of X,
Y, and Z as ghost coordinates, behaving as though all sectors have been modeled with
actual nodes and elements.)
Example 6.3: Deleting a Sector Load
CYCOPT,LDSECT,3

154

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Solving a Cyclic Symmetry Analysis

F,10,UX,value
3

! Apply a load (
value) on node 10 at sector

...
FDELE,10,UX

! Delete the load on node 10 at sector 3

To delete a previously applied load on a specified sector, issue an FDELE command.

6.3.1.3.2. Commands Affected by Non-Cyclic Loading

When non-cyclic loading applies, the following commands are affected: D, DA, DK, DL, FK, SF, SFA,
SFE, SFL.
The commands do not allow a table name in any value position. You must manually define any
boundary condition or loading table requiring more than one independent variable (for example, SECTOR
and some other variable).
The SF and SFE commands do not allow nodal pressure values to vary over an element face.

6.3.1.3.3. Plotting and Listing Non-Cyclic Boundary Conditions

You can plot non-cyclic boundary conditions (BCs) on the sector on which the BC (F, D, SF) is applied.
By expanding the cyclic sector model plot to the full 360 degrees (via the /CYCEXPAND command),
you can view a BC on the sector on which it is applied.
Issue BC-listing commands FLIST, DLIST, and SFLIST to list non-cyclic BCs. The list indicates the value
of the BC and the sector on which it is applied.

6.3.1.3.4. Graphically Picking Non-Cyclic Boundary Conditions

You can use graphical picking via the GUI to apply non-cyclic BCs on any sector. The graphical picking
option is available after expanding the cyclic model (/CYCEXPAND). Applicable BCs are:
Surface pressure (SF, SFL, SFA)
Force (F, FK)
BCs applied by graphical picking ignore the current CYCOPT,LDSECT setting when cyclic expansion
(/CYCEXPAND) is active. When cyclic expansion is not active, BCs are applied to the sector specified by
CYCOPT,LDSECT (or all sectors if CYCOPT,LDSECT,ALL).
The mathematical characteristics of a cyclic symmetry solution require that displacement BCs (D, DK,
DL, DA) apply to all sectors.

6.3.2. Solving a Static Cyclic Symmetry Analysis

For cyclically symmetric loading, support is available for linear static and large-deflection nonlinear
static solution options. (Cyclically symmetric loading implies any load applied on the cyclic sector representing a loading pattern that is repetitive at sector angle increments around the 360 structure.)
The following flowchart illustrates the process involved in a static (linear or large-deflection) cyclic
symmetry analysis with cyclic loading.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

155

Cyclic Symmetry Analysis

Figure 6.9: Process Flow for a Static Cyclic Symmetry Analysis (Cyclic Loading)

Only a harmonic index zero solution is valid for a static solution with cyclic loading.
For non-cyclically symmetric loading, the program supports linear static analysis only. The following
flowchart illustrates the process involved in a static cyclic analysis with non-cyclic loading.
Figure 6.10: Process Flow for a Static Cyclic Symmetry Analysis (Non-Cyclic Loading)

For more information, see Non-Cyclically Symmetric Loading (p. 152).

156

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Solving a Cyclic Symmetry Analysis

6.3.3. Solving a Modal Cyclic Symmetry Analysis

This section describes harmonic indices in relation to modal cyclic symmetry analyses and provides information necessary for solving several types of modal analyses. The following pages cover these topics:
Understanding Harmonic Index and Nodal Diameter (p. 157)
Solving a Stress-Free Modal Analysis (p. 158)
Solving a Prestressed Modal Analysis (p. 159)
Solving a Large-Deflection Prestressed Modal Analysis (p. 160)

6.3.3.1. Understanding Harmonic Index and Nodal Diameter

To understand the process involved in a modal cyclic symmetry analysis, it is necessary to understand
the concepts of harmonic indices and nodal diameters.
The nodal diameter refers to the appearance of a simple geometry (for example, a disk) vibrating in a
certain mode. Most mode shapes contain lines of zero out-of-plane displacement which cross the entire
disk, as shown in these examples:
Figure 6.11: Examples of Nodal Diameters (i)

For a complicated structure exhibiting cyclic symmetry (for example, a turbine wheel), lines of zero
displacement may not be observable in a mode shape.
The harmonic index is an integer that determines the variation in the value of a single DOF at points
spaced at a circumferential angle equal to the sector angle. For a harmonic index equal to nodal diameter d, the function cos(d*) describes the variation. This definition allows a varying number of waves
to exist around the circumference for a given harmonic index, provided that the DOF at points separated
by the sector angle vary by cos(d*). For example, a harmonic index of 0 and a 60 sector produce
modes with 0, 6, 12, ... , 6N waves around the circumference.
The nodal diameter is the same as the harmonic index in only some cases. The solution of a given
harmonic index may contain modes of more than one nodal diameter.
The following equation represents the relationship between the harmonic index k and nodal diameter
d for a model consisting of N sectors:

(6.2)

k xm d
where m = 0, 1, 2, 3, ...,
For example, if a model has seven sectors (N = 7) and the specified harmonic index k = 2, the program
solves for nodal diameters 2, 5, 9, 12, 16, 19, 23, ....
The following table illustrates Equation 6.2 (p. 157), showing how the harmonic index, nodal diameter
and number of sectors relate to one another:
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

157

Cyclic Symmetry Analysis

Harmonic Index
(k)

Nodal Diameter (d)

2N

2N

...

N-1

N+1

2N - 1

2N + 1

...

N-2

N+2

2N - 2

2N + 2

...

N-3

N+3

2N - 3

2N + 3

...

N-4

N+4

2N - 4

2N + 4

...

...

...

...

...

...

...

...

N/2

N/2

3N / 2

3N / 2

5N / 2

...

(N - 1) /
2

(N + 1) /
2

(3N - 1)
/2

(3N + 1)
/2

(5N - 1)
/2

...

N/2
(N
is
even)
(N - 1) /
2
(N
is
odd)

Note
To avoid confusion, be aware that in some references mode refers to harmonic index as
defined here and nodal diameter describes the actual number of observable waves around
the structure.
Harmonic Index in an Electromagnetic Analysis For electromagnetic analyses, only the EVEN and
ODD harmonic index settings (see the CYCOPT command) are valid (for symmetric and antisymmetric
solutions, respectively).
Using VT Accelerator You can use the Variational Technology Accelerator (VT Accelerator) to speed
up the solve time needed to sweep over the range of values of the harmonic index. To activate VT Accelerator, issue CYCOPT, VTSOL prior to solving. You can use VT Accelerator only with matched node
pattern sectors in a modal cyclic symmetry analysis. Linear perturbation modal analysis does not support
VT Accelerator. You will see the most significant speed up for models with a large number of sectors
and/or a large number of eigenvalues. The benefit of using VT Accelerator is realized only for five or
more Harmonic Indices (HI). In addition, the level of performance improvement realized with VT Accelerator may also be dependent up on the problem. Using less than five sectors prevents a solution and
displays an error message.
The Variational Technology acceleration method is available for the perturbed modal cyclic symmetric
analysis.

6.3.3.2. Solving a Stress-Free Modal Analysis

The following flowchart illustrates the process involved in a stress-free modal cyclic symmetry analysis.

158

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Solving a Cyclic Symmetry Analysis

Figure 6.12: Process Flow for a Stress-Free Modal Cyclic Symmetry Analysis

A modal cyclic symmetry analysis allows only cyclically symmetric applied boundary constraints and
applied loads. Applied loads are not used in a modal analysis (which is a free-vibration problem). Eigensolutions are performed, looping on the number of harmonic indices specified (via the CYCOPT
command) at each load step.

6.3.3.3. Solving a Prestressed Modal Analysis

The process for a prestressed modal cyclic symmetry analysis is essentially the same as that for a stressfree case, except that a static solution is necessary to calculate the prestress in the basic sector. The
prestress state of the sector may be from a linear static or a large-deflection nonlinear static analysis.
The following flowchart illustrates the process involved in a prestressed modal cyclic symmetry analysis.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

159

Cyclic Symmetry Analysis

Figure 6.13: Process Flow for a Prestressed Modal Cyclic Symmetry Analysis

The modal cyclic symmetry solution occurs after the static cyclic symmetry solution. The modal solution
uses the same low- and high-edge components defined in the static cyclic analysis stage (via the CYCLIC
command). The analysis yields the eigenvectors of the structure in the prestressed state.

6.3.3.4. Solving a Large-Deflection Prestressed Modal Analysis

Geometric nonlinearity occurs when the deflections are large enough to cause significant changes in
the geometry of the structure. In such cases, the equations of equilibrium must account for the deformed
configuration. (For more information, see the Mechanical APDL Theory Reference.) When a nonlinearity
is present, the program uses an iterative process to obtain the solution.
To calculate the frequencies and mode shapes of a deformed structure, you can perform a prestressed
modal analysis (using the linear perturbation solution method) of cyclic structuresafter first performing
a large-deflection (NLGEOM,ON) static analysis.
The prestress effects are automatically accounted for in the linear perturbation modal analysis procedure.
To obtain the cyclic symmetry modal solutions of a deformed structure, follow these steps:
1. Perform a nonlinear static solution with the prestress load. Use the RESCONTROL command to
define the necessary restart files.
2. Restart the previous static solution from the desired load step and substep.
3. Issue the PERTURB command to define the analysis type, material behavior to be used, contact
status (ContKey = CURRENT, STICKING, or BONDED) and load values to be retained from the previous
static solution (LoadControl = ALLKEEP, INERKEEP, PARKEEP, or NOKEEP).
4. Modify the behavior of individual contact pairs, as needed, using the CNKMOD command.

160

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Solving a Cyclic Symmetry Analysis

5. Issue the SOLVE, ELFORM command to regenerate the matrices.
6. Issue the MODOPT and MXPAND commands to specify the modal analysis option.
7. Issue the SOLVE command to perform the eigensolution.
8. Postprocess the results from the Jobname.RSTP file.
The flowchart below illustrates the process involved in a large-deflection prestressed modal cyclic
symmetry analysis.
Figure 6.14: Process Flow for a Large-Deflection Prestressed Modal Cyclic Symmetry Analysis

Prestressed Modal

SOLUTION
Perform a static restart with perturbation:
ANTYPE,STATIC,RESTART,,PERTURB

PREP7
Define the basic sector

Perform a large-deflection:
NLGEOM,ON

Define the options for the perturbed

modal analysis: PERTURB,MODAL
Reform the element matrices:
SOLVE,ELFORM

Define modal options: MODOPT

Specify cyclic symmetry:
CYCLIC
Define cyclic options: CYCOPT
SOLUTION
Define static analysis:
ANTYPE, STATIC

Define static loads

Solve modal analysis: SOLVE

POST1
Specify the result file for the linear
perturbation: FILE,,RSTP

Enable file writing for the restart

step: RESCONTROL,define
Read the results: SET
Define cyclic option
CYCOPT

Obtain the prestressed static

solution: SOLVE

Calculate max/min results: CYCPHASE

Display results on full model: /CYCEXPAND

For detailed information about the linear perturbation analysis procedure, see Linear Perturbation
Analysis in the Structural Analysis Guide. In addition, Example 9.4: Contact Status Control in a Linear
Perturbation Modal Analysis in that same document provides a complete example input listing for a
linear perturbation cyclic symmetry modal analysis.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

161

Cyclic Symmetry Analysis

6.3.3.4.1. Solving a Large-Deflection Prestressed Modal Analysis with VT Accelerator

Specify the VT Accelerator method (CYCOPT,VTSOL) in the block solution to solve a linear perturbed
cyclic symmetry modal analysis with the VT Accelerator method.
/solu
csys,0
antype,0
! Perform Static analysis.
nlgeom,on
! Include large deformation effects
rescontrol,define,all,1 ! Enable the file writing in multiframe restart
nsub,10,10,10
! Number of substeps = 10
solve
finish
/solu
antype,static,restart,,,perturb ! Perform a static restart with perturb
! from the last substep of the previous static solve
perturb,modal,,,allkeep ! Set the analysis options for perturbed modal analysis
solve,elform
outres,all,all
modopt,lanb,10
mxpand,10
cycopt,hindx,0,6,1
cycopt,vtsol,yes
solve
finish

! Solve harmonic indices ranging from 0 through 6

! Turn on the VT accelerator method

6.3.4. Solving a Linear Buckling Cyclic Symmetry Analysis

The process for a linear buckling analysis is essentially the same as that for a prestressed modal cyclic
symmetry analysis, with the exception that buckling options (ANTYPE,BUCKLE and BUCOPT,LANB) are
necessary to calculate buckling loads and the corresponding buckled mode shapes. The following
flowchart illustrates the process involved in an eigenvalue buckling cyclic symmetry analysis.
Figure 6.15: Process Flow for a Linear Buckling Cyclic Symmetry Analysis

You can specify the SET command's ORDER option to sort the harmonic index results in ascending order
of buckling load multipliers.
162

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Solving a Cyclic Symmetry Analysis

6.3.5. Solving a Harmonic Cyclic Symmetry Analysis

Cyclic or non-cyclic loading may be applied in a full harmonic cyclic analysis.
The flowchart below illustrates the process involved in a harmonic cyclic symmetry analysis with noncyclic loading.
Figure 6.16: Process Flow for a Harmonic Cyclic Symmetry Analysis (Non-Cyclic Loading)

For more information, see Non-Cyclically Symmetric Loading (p. 152)

6.3.6. Solving a Magnetic Cyclic Symmetry Analysis

Most magnetic analysis problems can be defined with flux parallel and/or flux normal boundary conditions. With problems such as electrical machines, however, cyclic boundary conditions best represent
the periodic nature of the structure and excitation, and have the advantage of being able to use a less
computation-intensive partial model, rather than a full model.
You can analyze only one sector of the full model to take advantage of this kind of symmetry. The full
model consists of as many sectors as the number of poles. In Example Magnetic Cyclic Symmetry Analysis (p. 190), the number of sectors is two; the analysis can be done on a half model.
The cyclic boundary condition is between matching degrees of freedom on corresponding symmetry
faces. The studied sector is bounded by two faces called the low edge and high edge, respectively. In

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

163

Cyclic Symmetry Analysis

Figure 6.27: Two-Phase Electric Machine Half Model (p. 191), the low edge face is the y = 0, x >= 0
plane; the high edge is the y = 0, x <= 0 plane.
The simplest case is when the node-matching interface of the low edge is the same as the high edge.
In this case, for every node, there is one and only one matching node on the high edge; moreover the
pertinent geometry and connectivity are the same. In this case, the cyclic boundary condition for the
edge formulation could be formulated as
Az(low entity) = - Az(high entity)
This is an anti-symmetric condition which is called ODD symmetry.
The analysis could be carried out on a 360/p sector (where p is the number of poles), in which case the
cyclic condition would be:
Az(low entity) = + Az(high entity)
which is a symmetric condition called EVEN symmetry.
In Example Magnetic Cyclic Symmetry Analysis (p. 190), the ODD model is smaller and thus more practical. For some problems, depending on geometry and excitation, EVEN symmetry may be more practical.
The program supports both ODD and EVEN cyclic symmetry.
In a more general case, the mesh on the low and high end may be different. In this case, more general
cyclic symmetry conditions can be established by interpolation on the pertinent faces. The program
handles this process automatically via the CYCLIC command.
The geometry of the low- and high-end cyclic faces may be more general than a simple plane surface.
Thus, for example, a skewed slot of an electric machine may constitute the cyclic sector modeled.
Cyclic Modeling (p. 144) discusses cyclic modeling in detail.
The following restrictions apply to a magnetic cyclic symmetry analysis:
Cyclic conditions can be restricted to specific degrees of freedom (DOFs) via the DOF option. DOF
restrictions may be useful, for example, in cases involving circuit-/voltage-fed solenoidal edge elements
.
Multiphysics coupling must use the same EVEN/ODD condition.
Circuit coupling is not supported for cyclic symmetry except solenoidal SOLID97.
Harmonic and transient analyses are not supported.
By default, plotting displays the partial solution only. To see the full model solution, issue the /CYCEXPAND command. For magnetic cyclic symmetry, the /CYCEXPAND command produces contour
plots but not vector plots.
Figure 6.9: Process Flow for a Static Cyclic Symmetry Analysis (Cyclic Loading) (p. 156) shows the process
for a cyclic symmetry analysis. The process is virtually identical for a magnetic cyclic symmetry analysis;
simply disregard the step for a large-deflection solution.
Magnetic cyclic boundary conditions can be applied to the following element types:
Nodal magnetic vector potential elements: PLANE13, PLANE53, SOLID97

164

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Solving a Cyclic Symmetry Analysis

Magnetic scalar potential elements: SOLID5, SOLID96, SOLID98
Electrostatic elements: PLANE121, SOLID122, SOLID123

6.3.7. Database Considerations After Obtaining the Solution

At the conclusion of the cyclic symmetry solution, exit the solution processor via the FINISH command.
If you intend to exit the program at this point (before postprocessing), save the database (file.db).
The saved database allows you to perform postprocessing on the analysis results at a later time.

6.3.8. Model Verification (Solution)

The cyclic solution reports the number of constraint equations generated for each harmonic index
solution and information about how they were created. The information should match what you already
know about the analysis model; if not, try to determine the reason for the discrepancy. The following
extracts (from a batch output file or an interactive output window) are typical:
NUMBER OF CONSTRAINT EQUATIONS GENERATED =
124
(USING THE MATCHED NODES ALGORITHM -MAX NODE LOCATION ERROR NEAR ZERO)

Meaning: 124 constraint equations are created, used, and then deleted to enforce cyclic symmetry
conditions between the low- and high-edge nodes. Every node on the low edge is precisely matched
to a corresponding node on the high edge, representing the best possible situation.
NUMBER OF CONSTRAINT EQUATIONS GENERATED =
124
(USING THE MATCHED NODES ALGORITHM -MAX NODE LOCATION ERROR = 0.73906E-02)

Meaning: 124 constraint equations are created, used, and then deleted to enforce cyclic symmetry
conditions between the low- and high-edge nodes. Every node on the low edge is matched to a corresponding node on the high edge within the current tolerance setting, but not all matches are precise.
The largest position mismatch is 0.0073906.
NUMBER OF CONSTRAINT EQUATIONS GENERATED =
(USING THE UNMATCHED NODES ALGORITHM)

504

Meaning: 504 constraint equations are created, used, and then deleted to enforce cyclic symmetry
conditions between the low- and high-edge nodes. At least one node on the low edge does not match
any node on the high edge within the current tolerance setting, so the program uses the unmatched
nodes algorithm.

6.3.8.1. Solving a Prestressed Harmonic Cyclic Symmetry Analysis

The process for solving a prestressed harmonic cyclic symmetry analysis is essentially the same as a
stress free case, except that a static solution is necessary to calculate the prestress in the basic sector.
The prestress state of the sector may be from a linear static or a large-deflection nonlinear static analysis.
The linear perturbed harmonic cyclic symmetric analysis is supported for the following methods: AUTO,
FULL, VT (see HROPT command).
Cyclic or non-cyclic loading may be applied in a prestressed full harmonic cyclic analysis.
The flowchart below illustrates the process of a harmonic cyclic symmetry analysis with non-cyclic
loading.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

165

Cyclic Symmetry Analysis

Figure 6.17: Process Flow for a Prestressed Harmonic Cyclic Symmetry Analysis

Prestressed Modal

SOLUTION
Perform a static restart with perturbation:
ANTYPE,STATIC,RESTART,,PERTURB

PREP7
Define the basic sector

Define the options for the perturbed

harmonic analysis: PERTURB,HARMONIC

Reform the element matrices:

SOLVE,ELFORM

Perform a large-deflection:
NLGEOM,ON

Specify cyclic symmetry:

CYCLIC

Define tabular loads or body loads

Define frequency range and number of

substeps:
HARFRQ
NSUB

SOLUTION
Define static analysis:
ANTYPE, STATIC

Define harmonic options: HROPT

Define static loads
Define cyclic options: CYCOPT
Enable file writing for the restart
step: RESCONTROL,define
Solve harmonic analysis: SOLVE
Define cyclic option
CYCOPT

Obtain the prestressed static

solution: SOLVE

POST1
Specify the result file for the linear
perturbation: FILE,,RSTP

Read the results: SET

Display results on full model:

/CYCEXPAND
PLNSOL

6.4. Postprocessing a Cyclic Symmetry Analysis

This section describes how to perform postprocessing on the solution results obtained from a cyclic
symmetry analysis.
The following topics are available for postprocessing a cyclic symmetry analysis:
6.4.1. Real and Imaginary Solution Components
6.4.2. Expanding the Cyclic Symmetry Solution
6.4.3. Phase Sweep of Repeated Eigenvector Shapes

166

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Postprocessing a Cyclic Symmetry Analysis

If you exited the program after obtaining the cyclic symmetry solution, use the database (file.db)
that you saved for postprocessing. For more information, see Database Considerations After Obtaining
the Solution (p. 165).

6.4.1. Real and Imaginary Solution Components

A cyclic symmetry solution typically has multiple load step results depending upon the harmonic index
solutions requested.
The real (basic sector) and imaginary (duplicate sector) parts of the solution reside in the results file.
However, the solution does not yet represent the actual displacements, stresses, or reaction forces for
any part of the actual structure.
Listing or plotting the sector results causes the program to issue a warning message such as PLNSOL is
displaying the unprocessed real and imaginary parts of this cyclic symmetry solution. Furthermore, the basic
and duplicate sectors overplot each other if displayed, providing yet another indication that a problem
exists.
To transform the real and imaginary cyclic symmetry solution results to the actual structure solution,
three postprocessing (/POST1) commands are available:
EXPAND
CYCPHASE

Note
The CYCPHASE command uses full model graphics (/GRAPHICS,FULL) to compute peak
values. Because of this, there may be slight differences between max/min values obtained
with CYCPHASE, and those obtained via /CYCEXPAND (/GRAPHICS,POWER).
For information about /CYCEXPAND and EXPAND command usage, see Expanding the Cyclic Symmetry
Solution (p. 167). For information about CYCPHASE command usage, see Phase Sweep of Repeated Eigenvector Shapes (p. 169).

6.4.2. Expanding the Cyclic Symmetry Solution

This section describes the capabilities of the /CYCEXPAND and EXPAND commands and explains their
differences. Use the commands to expand the solution results of your cyclic symmetry analysis to the
full model.

6.4.2.1. Using the /CYCEXPAND Command

The /CYCEXPAND command is the preferred operation for viewing expanded cyclic symmetry results.
The command does not modify the geometry, nodal displacements or element stresses stored in the
database. Issue the command to expand your basic sector model and obtain the full 360 model displacement, stress, or strain response.
After the expansion, you can plot (PLESOL or PLNSOL) or print (PRNSOL) the results. Other commands
(such as NSEL and NSORT) continue to operate on the unprocessed real and imaginary parts of the
solution.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

167

Cyclic Symmetry Analysis

Using the cyclic symmetry solution of the basic and duplicate sectors (illustrated in Figure 6.8: Connecting
Low and High Edges of Basic and Duplicate Sectors (p. 152)), the /CYCEXPAND command combines the
solutions from the two sectors by performing computations on the selected load step (specified via the
SET command) to combine the results of the two sectors. The program uses the following response
equation for the full structure or assembly:

(6.3)

where,
Uj = Response of the full structure or assembly (displacement, stress, or strain) for sector number j
UA = Basic sector solution
UB = Duplicate sector solution
j = Sector number for response expansion -- j = 1,2,3,,N
k = Harmonic index -- (0,1,2,,N / 2) when N is even, (0,1,2,,(N-1) / 2) when N is odd. (N is an integer
representing the number of sectors in 360.)
= Sector angle (2/N)
The equation applies when expanding the static cyclic solution as well as the modal cyclic eigenvector
solution.
If cyclic expansion via the /CYCEXPAND command is active for a static or harmonic cyclic symmetry
analysis, the PLNSOL and PRNSOL commands have summation of all required harmonic index solutions
by default. (You can override the default behavior if necessary.)
In a static or harmonic cyclic symmetry analysis with non-cyclic loading, all applicable harmonic index
solutions are computed and saved in the results file as load step results.
A SET,LIST command lists the range of load step numbers in the group containing each solution. Each
load step post data header contains the first, last, and count of load steps from the given SOLVE command, as shown:
SET
1
2
3
4
5
6
...

***** INDEX OF DATA SETS ON RESULTS FILE *****

TIME/FREQ LOAD STEP SUBSTEP CUMULATIVE HRM-INDEX
1.0000
1
1
1
0
2.0000
2
1
2
1
3.0000
3
1
3
2
1.0000
1
1
1
0
2.0000
2
1
2
1
3.0000
3
1
3
2

GROUP
1-3
1-3
1-3
4-6
4-6
4-6

The SET command establishes which SOLVE load step group should display. Summation via /CYCEXPAND
is automatic (although you can override the default behavior). Plots and printed output show the
summation status.
With /CYCEXPAND turned on, the results are expanded at each load step and then combined to plot
the full solution as a complete sum. For example, in a four sector model where harmonic index results
0 through 3 are available in the results file, the plot command PLNSOL will display the results as STEP=1
THRU=3 COMPLETE SUM.
Accumulation occurs at the first applicable PLNSOL or PRNSOL command. After accumulation, the last
load step number of the current group becomes the new current load step number.

168

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Postprocessing a Cyclic Symmetry Analysis

6.4.2.1.1. Applying a Traveling Wave Animation to the Cyclic Model

After you have completed a modal cyclic symmetry analysis, you can apply an animated traveling wave
to the cyclic model by issuing the ANCYC command (which uses /CYCEXPAND functionality). The
traveling wave capability applies only to modal cyclic symmetry analyses. For more information, see
the description of the ANCYC command in the Command Reference.
Figure 6.18: Traveling Wave Animation Example (p. 169) illustrates the ANCYC command's effect. To view
the input file used to create the model shown, see Example Modal Cyclic Symmetry Analysis (p. 172).
The following demo is presented as an animated GIF. Please view online if you are reading the PDF version
of the help. Interface names and other components shown in the demo may differ from those in the released
product.
Figure 6.18: Traveling Wave Animation Example

6.4.2.2. Using the EXPAND Command

The EXPAND command offers an alternate method for displaying the results of a modal cyclic symmetry
analysis. It is a specification command that causes a SET operation to transform and expand the data
it is reading before storing it in the database. If you request two or more sector repetitions, the command
creates additional nodes and elements to provide space for the extra results.
After the real-space results are stored in the database, you can plot (PLESOL or PLNSOL), print (PRNSOL)
or process them as you would those for a non-cyclic model, a convenience if you want to process the
results in a manner unsupported by the /CYCEXPAND command. The database can become very large,
however, negating the inherent model-size advantage of a cyclic symmetry analysis.

Caution
Do not confuse the EXPAND command with /EXPAND.

6.4.3. Phase Sweep of Repeated Eigenvector Shapes

In a modal cyclic symmetry analysis, repeated eigenfrequencies are obtained at solutions corresponding
to harmonic indices k, greater than 0 and less than N/2. The repeated modes are a consequence of the
cyclically symmetric geometry of the structure or assembly being modeled by the cyclic sector.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

169

Cyclic Symmetry Analysis

The eigenvector shapes corresponding to the repeated eigenfrequencies are non-unique. That is, for
the repeated eigenfrequencies fi = fi+1, the mode shapes corresponding to fi and fi+1 can be linearly
combined to obtain a mode shape that is also a valid mode shape solution for the frequencies fi and
fi+1. A valid linear combination of the eigenvectors is:

(6.4)

where,
c1 and c2 = Arbitrary constants
Ui and Ui+1 = Eigenvectors corresponding to fi and fi+1, respectively
The orientation of the combined mode shape U will be along a nodal diametral line that is neither
along that of Ui nor Ui+1. Because the full structure may have stress-raising features (such as bolt holes),
determining the eigenvector orientation that causes the most severe stresses, strains, or displacements
on the structure or assembly is critical.
To determine the peak value of stress, strain or displacement in the full structure or assembly, it is necessary to calculate U at all possible angular orientations in the range of 0 through 360. In the general postprocessor, the CYCPHASE command performs the computational task.
Because c1 and c2 are arbitrary constants, the CYCPHASE calculation rewrites Equation 6.4 (p. 170) as
follows:

(6.5)

Using the cyclic symmetry expansion of Equation 6.3 (p. 168) in Equation 6.5 (p. 170), the simplified phasesweep equation that operates on the cyclic sector solution (rather on the computation-intensive fullstructure expression in Equation 6.5 (p. 170)) is:

(6.6)

A phase sweep using the CYCPHASE command provides information about the peak values of stress,
strain and/or displacement components and the corresponding phase angle values. Using the phase
angle value further, you can expand the mode shape at that phase angle to construct the eigenvector
shape that produces the peak stress, strain and/or displacement. The expansion expression with the
phase angle used by the /CYCEXPAND command is:

(6.7)

where,
j = 1,2,3,...,N
Example:
To determine the eigenvector orientation that causes the highest equivalent stress, perform a phase
sweep on the stress via the CYCPHASE,STRESS command. Obtain a summary of the phase sweep via
the CYCPHASE,STAT command to determine the value of at which maximum equivalent stress occurred.

170

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Limitations in a Cyclic Symmetry Analysis

You can shift the mode shape to that angle via the /CYCEXPAND,,PHASEANG command and plot the
expanded mode shape via the PLNSOL,S,EQV command.

Note
The CYCPHASE command uses full model graphics (/GRAPHICS,FULL) to compute peak
values. Because of this, there may be slight differences between max/min values obtained
with CYCPHASE, and those obtained via /CYCEXPAND (/GRAPHICS,POWER).

6.5. Limitations in a Cyclic Symmetry Analysis

This section describes limitations in a cyclic symmetry analysis
The following topics are available:
6.5.1. Solver Limitations
6.5.2. /CYCEXPAND Limitations

6.5.1. Solver Limitations

Cyclic symmetry solution does not work for elements that have thermal and hydrostatic pressure degrees
of freedom.
The following structural element types are not supported:
Axisymmetric elements (such as PLANE83)
Generalized axisymmetric elements (such as SOLID272)
Multipoint constraint elements (MPC184)
Layered composite elements (such as SOLID185 with KEYOPT(3)=1)
Thermal surface effect elements (such as SURF151)
Reinforced elements (such as REINF263)
User define element (USER300)
Shear/Twist panel elements (SHELL28)
Reinforced concrete solid elements (SOLID65)
Gasket Elements (INTER192, INTER193, INTER194, INTER195)
Superelement (MATRIX50)
Generalized plane strain option (such as PLANE183 with KEYOPT(3)=5)
Cyclic symmetry modal analyses currently support only the Block Lanczos, PCG Lanczos, and Supernode
methods.
The linear perturbation procedure is not supported for cyclic symmetry buckling analyses.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

171

Cyclic Symmetry Analysis

Non-cyclic loading cannot be applied in the base analysis when performing a linear perturbation modal
and/or harmonic analysis.
Birth and death of elements (EKILL and EALIVE) is not supported in a cyclic symmetry solution.
Non-linear (NLGEOM) cyclic symmetry analyses with non-cyclic loading are not supported.

6.5.2. /CYCEXPAND Limitations

Only PLNSOL, PRNSOL (plot and print of nodal solution), and PLESOL (plot of element solution) are
supported after /CYCEXPAND. All other post processing commands operate on unprocessed real and
imaginary parts of solution.
Post processing a nodal solution on a selected set of nodes (for example, on nodal components defined
by CM) is not supported after /CYCEXPAND.
Load case operations (LCOPER) are not supported after /CYCEXPAND.
Post processing contact element status (PRESOL,CONT) is not supported after /CYCEXPAND.
The /CYCEXPAND command does not work with PGR files.
The /CYCEXPAND command is incompatible with the /ESHAPE,1 or /ESHAPE,FAC command.
The /PBC (plot boundary conditions) command is deactivated when cyclic expansion is active (/CYCEXPAND,,ON). To view BCs on the basic sector, deactivate cyclic expansion (/CYCEXPAND,,OFF) and issue
this command: /PBC,ALL,,1
In cases involving non-cyclically symmetric loading, all boundary conditions (BCs) will plot when /CYCEXPAND is active. When examining the plot, it is not possible to distinguish between BCs that exist on every
sector and those that exist (via a table referencing SECTOR) only on some sectors or which vary in value.
(Only the basic sector is plotted.)
For magnetic cyclic symmetry analyses, the /CYCEXPAND command produces contour plots, but not
vector plots.
The AVRES command with Opt=FULL is not supported after the /CYCEXPAND command is issued.

6.6. Example Modal Cyclic Symmetry Analysis

This example modal cyclic symmetry analysis presents a simplified ring-strut-ring structure used in many
rotating-machinery applications.
6.6.1. Problem Description
6.6.2. Problem Specifications
6.6.3. Input File for the Analysis
6.6.4. Analysis Steps

6.6.1. Problem Description

The component is a simplified fan inlet case for a military aircraft engine. As part of the design process
for the assembly, you must determine the vibration characteristics (natural frequencies and mode
shapes) of the inlet case.

172

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Modal Cyclic Symmetry Analysis

6.6.2. Problem Specifications

The geometric properties for this analysis are as follows:

The material properties for this analysis are as follows:

Young's modulus (E) = 10e6
Poisson's ratio () = 0.3
Density = 1e-4
All applicable degrees of freedom are used for the cyclic symmetry edge-component pairs. The first six
mode shapes for all applicable harmonic indices are requested.

6.6.3. Input File for the Analysis

Use this input file (named cyc_symm.inp) to perform the example modal cyclic symmetry analysis.
The file contains the complete geometry, material properties and solution options for the finite element
model.
! Modal Cyclic Symmetry Analysis Example
! Ring-Strut-Ring Configuration

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

173

Cyclic Symmetry Analysis

! STEP #1
! Start an interactive session
! STEP #2
! Read in this input file: cyc_symm.inp
finish
/clear
r1=5
r2=10
d1=2
nsect=24
alpha_deg=360/nsect
alpha_rad=2*acos(-1)/nsect
/view,1,1,1,2
/plopts,minm,0
/plopts,date,0
/pnum,real,1
/number,1
/prep7
csys,1
k,1,0,0,0
k,2,0,0,d1
k,3,r1,0,0
k,4,r1,0,d1
l,3,4
arotat,1,,,,,,1,2,alpha_deg/2
k,7,r2,0,0
k,8,r2,0,d1
l,7,8
arotat,5,,,,,,1,2,alpha_deg/2
arotat,2,,,,,,1,2,alpha_deg/2
arotat,6,,,,,,1,2,alpha_deg/2
a,5,6,10,9
mshkey,1
et,1,181
r,1,0.20
r,2,0.1
mp,ex,1,10e6
mp,prxy,1,0.3
mp,dens,1,1e-4
esize,0.5
asel,,,,1,4
aatt,,1
asel,,,,5
aatt,,2
allsel
finish
/solution
antype,modal
modopt,lanb,6
mxpand,6,,,yes
dk,5,uz,0
finish
aplot
/prep7
/eof
! STEP #3
! Configure the database for a cyclic symmetry analysis
cyclic
! STEP #4
! Mesh the areas

174

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Modal Cyclic Symmetry Analysis

amesh,all
! STEP #5
! Turn on cyclic symmetry graphical expansion
/cycexpand,,on
! STEP #6
! Plot the elements
eplot
! STEP #7
! List the cyclic status
cyclic,status
! STEP #8
! List the cyclic solution option settings
cycopt,status
! STEP #9
! Solve the modal cyclic symmetry analysis
/solution
solve
! STEP #10
! Specify global cylindrical as the results coordinate system
/post1
rsys,1
! STEP #11
! Read results for "load step 1 - substep 4 - harmonic index 0"
set,2,6
! STEP #12
! Plot the tangential displacement contour
plns,u,y
! STEP #13
! Read results for "load step 13 - substep 1 - harmonic index 12"
set,13,1
! STEP #14
! Plot the tangential displacement contour
plns,u,y
! STEP #15
! Read results for "load step 2 - substep 5 - harmonic index 1"
set,2,5
! STEP #16
! Plot the tangential displacement contour
plns,u,y

6.6.4. Analysis Steps

The following table describes the input listing and the steps involved in the example modal cyclic
symmetry analysis in more detail.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

175

Cyclic Symmetry Analysis

Step

Description

Command

1.

Start an interactive session.

---

2.

Read the input file: cyc_symm.inp

/INPUT,CYC_SYMM.INP

3.

Specify a cyclic symmetry analysis and configure the database

accordingly.

CYCLIC

4.

Mesh the areas.

AMESH,ALL

5.

Activate cyclic symmetry graphical expansion.

/CYCEXPAND,,ON

6.

Plot the elements.

EPLOT

7.

List the cyclic status.

CYCLIC,STATUS

8.

List the cyclic solution option settings.

CYCOPT,STATUS

9.

Solve the modal cyclic symmetry analysis.

/SOLU
SOLVE

10.

Specify the global cylindrical coordinate system.

/POST1
RSYS,1

11.

Read results for load step 1 - substep 4 - harmonic index 0.

SET,2,6

12.

Plot the tangential displacement contour.

PLNSOL,U,Y

Executing this step causes the struts of the assembly to bend in

phase.
13.

Read results for load step 13 - substep 1 - harmonic index 12.

SET,13,1

14.

Plot the tangential displacement contour.

PLNSOL,U,Y

Executing this step causes the struts of the assembly to bend out
of phase.
15.

Read results for load step 2 - substep 5 - harmonic index 1.

SET,2,5

16.

Plot the tangential displacement contour.

PLNSOL,U,Y

---This step completes the example modal cyclic symmetry analysis.

Your results should match those shown in Figure 6.19: Example
Modal Cyclic Symmetry Analysis Results (p. 177).
The results of your analysis should match those shown here:

176

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Buckling Cyclic Symmetry Analysis

Figure 6.19: Example Modal Cyclic Symmetry Analysis Results

Note
Mode shape values may vary slightly depending on your computer system.
To view a traveling wave animation of your model, issue the ANCYC,24,,0.1 command. For more information, see Applying a Traveling Wave Animation to the Cyclic Model (p. 169).

6.7. Example Buckling Cyclic Symmetry Analysis

This example buckling cyclic symmetry analysis presents a simplified ring-strut-ring structure used in
many rotating-machinery applications.
6.7.1. Problem Description
6.7.2. Problem Specifications
6.7.3. Input File for the Analysis
6.7.4. Analysis Steps
6.7.5. Solve For Critical Strut Temperature at Load Factor = 1.0

6.7.1. Problem Description

The object is a simplified structure that experiences a thermal load emanating outward from the center.
The inner ring is kept at a constant 600 F temperature and the outer ring is kept at a constant 0 F. A
linear eigenvalue buckling analysis determines when the struts will buckle as the temperature in the
struts increases from 100 F to 500 F.

6.7.2. Problem Specifications

The geometric properties for this analysis are as follows:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

177

Cyclic Symmetry Analysis

The material properties for this analysis are as follows:

Poisson's ratio () = 0.3
Density = 1e-4
Coefficient of thermal expansion () = 5e-5
Young's modulus (E) = 10e6 (at 0 F)
Young's modulus (E) = 4e6 (at 600 F)
The Young's modulus value varies linearly between 0 and 600 F. All applicable degrees of freedom
(DOFs) are used for the cyclic symmetry edge-component pairs. The first six mode shapes for all applicable
harmonic indices are requested.

6.7.3. Input File for the Analysis

Use this input file (named buck_cyc_sym.inp) to perform the buckling cyclic symmetry analysis
example. The file contains the complete geometry, material properties and solution options for the finite
element model.

178

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Buckling Cyclic Symmetry Analysis

! Cyclic Symmetry Buckling Example

! Ring-Strut-Ring Configuration
! STEP #1
! Start an interactive session
! STEP #2
! Read in the input file: buck_cyc_sym.inp
r1=5
r2=15
d1=4
nsect=6
alpha_deg=360/nsect
alpha_rad=2*acos(-1)/nsect
/view,1,1,1,2
/plopts,minm,0
/plopts,date,0
/pnum,real,1
/number,1
/prep7
csys,1
k,1,0,0,0
k,2,0,0,d1
k,3,r1,0,0
k,4,r1,0,d1
l,3,4
arotat,1,,,,,,1,2,alpha_deg/2
k,7,r2,0,0
k,8,r2,0,d1
l,7,8
arotat,5,,,,,,1,2,alpha_deg/2
arotat,2,,,,,,1,2,alpha_deg/2
arotat,6,,,,,,1,2,alpha_deg/2
a,5,6,10,9
mshkey,1
et,1,181
r,1,0.20
r,2,0.1
mptemp,1,0
mptemp,2,600
mpdata,ex,1,1,10e6
mpdata,ex,1,2,4e6
mp,prxy,1,0.3,0.0
mp,dens,1,1e-4
mp,alpx,1,5e-5
tref,0
esize,1.0
asel,,loc,x,r1
bfa,all,temp,600
asel,a,loc,x,r2
aatt,,1
asel,inve
bfa,all,temp,100
aatt,,2
allsel
amesh,all
lsel,,loc,z,d1/2
lsel,r,loc,y,alpha_deg/2
ksll
nslk
nrotate,all
dk,all,uz,0
dk,all,uy,0
allsel
finish
aplot
/prep7

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

179

Cyclic Symmetry Analysis

/eof
! STEP #3
! Configure the database for a cyclic symmetry analysis
cyclic
! STEP #4
! Turn on cyclic symmetry graphical expansion
/cycexpand,,on
! STEP #5
! Plot the elements
eplot
! STEP #6
! List the cyclic status
cyclic,status
! STEP #7
! List the cyclic solution option settings
cycopt,status
! STEP #8
! Specify static analysis type with prestress effects
/solution
antype,static
pstres,on
! STEP #9
! Solve the prestress static analysis
solve
! STEP #10
! Specify buckling analysis type
finish
/solution
antype,buckle
! STEP #11
! Specify buckling analysis options
bucopt,lanb,3
! STEP #12
! Specify mode expansion options
mxpand,3,,,yes
! STEP #13
! Solve the buckling analysis
solve
! STEP #14
! Read results for the smallest load factor
finish
/post1
set,first,,,,,,,order
! STEP #15
! Plot the buckled mode shape
plnsol,u,sum

180

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Buckling Cyclic Symmetry Analysis

6.7.4. Analysis Steps

The following table describes the input listing and the steps involved in the example buckling cyclic
symmetry analysis in more detail.
Step

Description

Command

1.

Start an interactive session.

---

2.

Read in the input file: buck_cyc_sym.inp

/INPUT,buck_cyc_sym.inp

3.

Specify a cyclic symmetry analysis and configure the database

accordingly.

CYCLIC

4.

Activate cyclic symmetry graphical expansion.

/CYCEXPAND,,ON

5.

Plot the elements.

EPLOT

6.

List the cyclic status.

CYCLIC,STATUS

7.

List the cyclic solution option settings.

CYCOPT,STATUS

8.

Specify a static analysis type with prestress effects.

9.

Solve the prestress static analysis.

10.

Specify a buckling analysis type.

11.

Specify buckling analysis options

BUCOPT, LANB, 3

12.

Specify mode expansion options.

MXPAND, 3, , , YES

13.

Solve the buckling analysis.

SOLVE

14.

Read the results from the smallest load factor. (This should correspond to the smallest frequency.)

15.

Plot the buckled mode shape.

/SOLU
ANTYPE,STATIC
PSTRES,
ON
SOLVE
FINISH
/SOLU
ANTYPE,BUCKLE

FINISH
/POST1
SET, FIRST ,
, , , , , , ORDER
PLNSOL, U, SUM

---This step completes the example buckling cyclic symmetry analysis. Your results should match those shown in Figure 6.20: Example Buckling Cyclic Symmetry Analysis Results (p. 182).
The results of your analysis should match those shown here:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

181

Cyclic Symmetry Analysis

Figure 6.20: Example Buckling Cyclic Symmetry Analysis Results

6.7.5. Solve For Critical Strut Temperature at Load Factor = 1.0

You can automatically solve for the critical strut temperature by iterating on the variable loads until
the eigenvalue becomes 1.0 (or nearly 1.0 within some tolerance). The iterations ensure that the eigenvalue solution does not factor the stress stiffness matrix from the constant loads. The following flowchart
illustrates the process:
Figure 6.21: Buckling Cyclic Symmetry Results: Load Factor Iterations

Use the /PREP7 portion of the previous input file (buck_cyc_sym.inp) to construct your model.
After defining the model parameters--but before activating cyclic symmetry--define the arrays and the
programming operations, as follows:
*dim,Tstrut,array,10
*dim,Tfact,array,10
*do,I,1,10
/prep7

182

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Buckling Cyclic Symmetry Analysis

*if,I,eq,1,then
Tstrut(I)=100
*else
Tstrut(I)=Tstrut(I-1)*Tfact(I-1)
cyclic,undouble
*endif
asel,,real,,2
bfa,all,temp,Tstrut(I)
allsel

After you have defined the iterative parameters, proceed with the cyclic symmetry portion of the analysis:
cyclic
/cycexpand,,on
eplot
cyclic,status
cycopt,status
/solution
antype,static
pstres,on
solve
finish
/solution
antype,buckle
bucopt,lanb,3
mxpand,3,,,yes
solve
finish
/post1
set,first,,,,,,,order
plnsol,u,sum
*get,loadmult,active,,set,freq
Tfact(I)=loadmult
*enddo

The program then plots the data to determine the critical strut temperature:
*dim,data,table,10,2
data(0,1)=1
data(0,2)=2
*do,I,1,10
data(I,0)=I
data(I,1)=Tstrut(I)
data(I,2)=Tfact(I)
*enddo
/AXLAB,X,Strut Temperature
/AXLAB,Y,Load Factor
/GROPT,DIVX,5
/GROPT,DIVY,5
/XRANGE,100,200
/YRANGE,0.9,1.4
/GTHK,CURVE,1
/GMARKER,1,3
*VPLOT,data(1,1),data(1,2)

The eigenvalues (frequencies) calculated for the buckling analysis represent the buckling load factors.
The eigenvalues represent load factors for all applied loads.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

183

Cyclic Symmetry Analysis

The iteration strategy yields the following results:
Table 6.2: Buckling Cyclic Symmetry: Load Factor Iteration Results
Iteration

T (Strut)

Load Factor

100.00

1.3039

130.39

1.1845

154.44

1.0972

169.45

1.0461

177.27

1.0206

180.91

1.0089

182.52

1.0038

183.21

1.0016

183.50

1.0007

10

183.62

1.0003

A graph of the results shows the convergence at Load Factor = 1.0:

Figure 6.22: Buckling Cyclic Symmetry Results: Load Factor Results Graph

6.8. Example Harmonic Cyclic Symmetry Analysis

This section introduces the harmonic cyclic symmetry analysis capability with an example problem. The
example presents a simplified ring-strut-ring structure used in many rotating-machinery applications.
6.8.1. Problem Description
6.8.2. Problem Specifications
6.8.3. Input File for the Analysis
6.8.4. Analysis Steps

184

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Harmonic Cyclic Symmetry Analysis

6.8.1. Problem Description

The component is a simplified fan inlet case for a military aircraft engine. As part of the design process
for the assembly, the harmonic response characteristics of the inlet case may be investigated, as shown
in this example.

6.8.2. Problem Specifications

The geometric and material properties used for the harmonic cyclic analysis are the same as used in
Example Modal Cyclic Symmetry Analysis (p. 172).
All applicable degrees of freedom are used for the cyclic symmetry edge-component pairs. The harmonic
response to a pressure load applied on the assembly is computed.

6.8.3. Input File for the Analysis

Use the input file given below (named harm_cyc_symm.inp) to perform the example modal cyclic
symmetry analysis. The file contains the complete geometry, material properties and solution options
for the finite element model.
! Harmonic Cyclic Symmetry Analysis Example ! Ring-Strut-Ring Configuration
! STEP #1
! Start an interactive session
! STEP #2
! Read in this input file: harm_cyc_symm.inp
finish
/clear
r1=5
r2=10
d1=2
nsect=24
alpha_deg=360/nsect
alpha_rad=2*acos(-1)/nsect
/view,1,1,1,2
/plopts,minm,0
/plopts,date,0
/pnum,real,1
/number,1
/prep7
csys,1
k,1,0,0,0
k,2,0,0,d1
k,3,r1,0,0
k,4,r1,0,d1
l,3,4
arotat,1,,,,,,1,2,alpha_deg/2
k,7,r2,0,0
k,8,r2,0,d1
l,7,8
arotat,5,,,,,,1,2,alpha_deg/2
arotat,2,,,,,,1,2,alpha_deg/2
arotat,6,,,,,,1,2,alpha_deg/2
a,5,6,10,9
mshkey,1
et,1,181
r,1,0.20
r,2,0.1
mp,ex,1,10e6

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

185

Cyclic Symmetry Analysis

mp,prxy,1,0.3
mp,dens,1,1e-4
esize,0.5
asel,,,,1,4
aatt,,1
asel,,,,5
aatt,,2
allsel
dk,5,all,0
aplot
/prep7
!/eof
! STEP #3
! Configure the database for a cyclic symmetry analysis
cyclic
! STEP #4
! Mesh the areas
amesh,all
! STEP #5
! Turn on cyclic symmetry graphical expansion
/cycexpand,,on
! STEP #6
! Plot the elements
eplot
! STEP #7
! List the cyclic status
cyclic,status
! STEP #8
! List the cyclic solution option settings
cycopt,status
fini
! STEP #9
! Solve the harmonic cyclic symmetry analysis
/solution
! Specify harmonic analysis option and frequency sweep values
antype,harmic
hropt,full
harfrq,860,878
nsub,9
! Apply pressure load on the outer surface area
asel,s,area,,2
asel,a,area,,4
! Loading to be applied on sector number 3
cycopt,ldsect,3
sfa,all,,pres,-124
asel,all
solve
fini
! STEP #10
! Go to database results post processor
/post1

186

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Harmonic Cyclic Symmetry Analysis

set,list
! STEP #11
! Read results for "substep 3

- excitation freq. of 866 HZ"

set,,3
plns,u,sum,1
fini
! STEP #12
! Go to time/freq. history postprocessor
/post26
! STEP #13
! Store X and Y disp. of node #34 on sector #3"
nsol,2,34,U,x,UX_34,3
nsol,3,34,U,y,UY_34,3
! STEP #14
! Plot UX and UY disp. of node #34 as a function of freq.
plvar,2,3

6.8.4. Analysis Steps

The following table describes the input listing and the steps involved in the example modal cyclic
symmetry analysis in detail.
Step

Step Description

Description

1.

Start an interactive session.

---

2.

Read the input file: harm_cyc_symm.inp.

/INPUT,HARM,CYC_SYMM.INP

3.

Specify a cyclic symmetry analysis and configure CYCLIC

the database accordingly.

4.

Mesh the areas.

AMESH,ALL

5.

Activate cyclic symmetry graphical expansion.

/CYCEXPAND,,ON

6.

Plot the elements. Figure 6.23: Element Plot

Showing Pressure Load on Sector 3 (p. 188)
shows an element plot showing pressure load
on sector 3.

EPLOT

7.

List the cyclic status.

CYCLIC,STATUS

8.

List the cyclic solution option settings.

CYCOPT,STATUS

9.

Solve the modal cyclic symmetry analysis.

/SOLU
SOLVE

10.

Enter database results postprocessor.

/POST1

11.

Read results for sub step 3 Frequency = 866.

SET,1,3

12.

Plot the displacement sum contour. Figure 6.24: Contour Plot of Displacement Sum at
Frequency of 866 HZ (p. 189) shows a contour
plot of displacement sum at frequency = 866
HZ.

PLNSOL,U,SUM

13.

Enter time/frequency history postprocessor.

/POST26

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

187

Cyclic Symmetry Analysis

14.

Store nodal data from results file for node 34.

NSOL,2,34,u,x,UX_34,3
NSOL,3,34,u,y,UY_34,3

15.

Plot frequency versus displacement. Plot at node PLVAR,2,3

34. Figure 6.25: Displacement Plot as a Function
of Excitation Frequency (p. 190) shows the displacement plot as a function of the excitation
frequency.

The results of your analysis should match those shown below:

Figure 6.23: Element Plot Showing Pressure Load on Sector 3

188

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Harmonic Cyclic Symmetry Analysis

Figure 6.24: Contour Plot of Displacement Sum at Frequency of 866 HZ

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

189

Cyclic Symmetry Analysis

Figure 6.25: Displacement Plot as a Function of Excitation Frequency

6.9. Example Magnetic Cyclic Symmetry Analysis

This magnetic cyclic symmetry analysis uses a model of a simplified electrical machine where the
model size can be reduced via cyclic boundary conditions.
6.9.1. Problem Description
6.9.2. Problem Specifications
6.9.3. Input file for the Analysis

6.9.1. Problem Description

Figure 6.26: Two-Phase Electric Machine Full Model (p. 191) shows a typical example, the full model of
a 2-phase electrical machine.
In the full model, flux parallel boundary conditions can be formulated at the outer surface of the stator
frame. If only phase A were excited, the magnetic flux would point in the y direction at x=0 plane; flux
parallel condition could be formulated at the x=0 plane, allowing an analysis on a half model in the
x>=0 space. Similarly, if only phase B were excited, the magnetic flux would have only x component
on the y=0 plane; again, flux parallel could be applied to a half model in the y>=0 plane.
Typically, however, both coils are excited, and no flux parallel conditions could be formulated over the
x=0 or y=0 planes. However, due to the cyclic nature of the field, the field pattern repeats itself after
180 degrees. In particular, on the y=0 plane:

190

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Magnetic Cyclic Symmetry Analysis

By(x) = By(-x)
A similar pattern can be observed in Figure 6.27: Two-Phase Electric Machine Half Model (p. 191), where
the flux lines (equi vector potential lines) are plotted:
Az(x) = - Az(-x)
In this example, the field has a two pole pattern. In general, there are 2p poles; the repetition would
take place after 180/p degrees.
Figure 6.26: Two-Phase Electric Machine Full Model
y

Cf eshp liog

roto

gf litnegnt xulp

Ee eshp liog

rott
f
C
x

C
f

Ce eshp liog

f
pg rie
Ef eshp liog

Figure 6.27: Two-Phase Electric Machine Half Model

 

d  

d w 
 d

6.9.2. Problem Specifications

The material properties for this analysis are as follows:
Iron relative permeability: 1000
Iron electrical resistivity: 9.579E-8
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

191

Cyclic Symmetry Analysis

Aluminum relative permeability: 1.0
Aluminum electrical resistivity: 2.65E-8
Copper relative permeability: 1.0
Copper electrical resistivity: 1.74E-8

6.9.3. Input file for the Analysis

Use this input file to perform the example magnetic cyclic symmetry analysis. This file contains the
complete geometry, material properties, and solution options for the finite element model. Magnetic
cyclic symmetry commands of particular interest are preceded by the comment:
!!!

Apply Cylic

/title,Cyclic Symmetry Model for EMAG Analysis (Dual Coils with Iron Yoke)
/com
/com ***** Quarter Symmetry Model Expanded to Half Then to Full *****
/com
/com
/com
/nopr
/out,scratch
!!! Setup Model Parameters

_geomgen=1
p=1

!
!
!
!
!

alpha=22.5/p
beta=alpha+(45/p)
gamma=beta+(22.5/p)
r1=3
r2=4.5
r3=5
r4=7
r5=11
ncoil=(4*p)
i1=1
i2=2

Use for number of quarter sectors

(i.e. 1 = 1 90deg sector, 2 = 2 sectors in 90deg)
angle up to the end of first coil
angle from coil1 to coil2
angle from beginning of coil2 to end of sector

*dim,alpha1,,ncoil
*dim,alpha2,,ncoil
*dim,current,,ncoil
*dim,coilname,string,ncoil
coilname(1)
coilname(2)
coilname(3)
coilname(4)

=
=
=
=

'coil1'
'coil2'
'coil3'
'coil4'

*do,i,1,ncoil
alpha1(i) = -alpha + (i-1)*(90/p)
alpha2(i) = alpha + (i-1)*(90/p)
*enddo
ii=0
*do,i,1,p
ii = ii + 1
current(ii) = i2
ii = ii + 1
current(ii) = i1

192

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Magnetic Cyclic Symmetry Analysis

ii = ii + 1
current(ii) = -i2
ii = ii + 1
current(ii) = -i1
*enddo

/prep7

ET,1,13,4

! Use PLANE13 Elements (DOFset = UX,UY,TEMP,AZ)

!!! Setup Model using Parameters

PCIRC,r1, ,0,alpha,
PCIRC,r1, ,0,beta
PCIRC,r1, ,0,gamma
PCIRC,r2, ,0,alpha
PCIRC,r2, ,0,beta
PCIRC,r2, ,0,gamma
PCIRC,r3, ,0,alpha
PCIRC,r3, ,0,beta
PCIRC,r3, ,0,gamma
PCIRC,r4, ,0,alpha
PCIRC,r4, ,0,beta
PCIRC,r4, ,0,gamma
PCIRC,r5, ,0,alpha
PCIRC,r5, ,0,beta
PCIRC,r5, ,0,gamma
AOVLAP,ALL
!!! Setup Material Properties
! IRON
MP,MURX,1,1000
MP,RSVX,1,9.579E-8
! AL
MP,MURX,2,1
MP,RSVX,2,2.65E-8
! Copper
MP,MURX,3,1
MP,RSVX,3,1.74E-8
! Air
MP,MURX,4,1
MP,RSVX,4,0
!!! Setup Components and Atributes
! Iron Core
CSYS,1
ASEL,S,LOC,X,0,r1
CM,Inner_Iron,AREA
AATT,1,,1,

! Enter Cylindrical Mode

! Al Core
ASEL,S,LOC,X,r1,r2
CM,Outer_AL,AREA
AATT,2,,1,
! Air Gap
ASEL,S,LOC,X,r2,r3
CM,AIR,AREA
AATT,4,,1
! Coil 1
ASEL,S,LOC,X,r3,r4
ASEL,R,LOC,Y,0,alpha
CM,COIL1,AREA

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

193

Cyclic Symmetry Analysis

AATT,3,,1
! Coil 2
ASEL,S,LOC,X,r3,r4
ASEL,R,LOC,Y,beta,gamma
CM,COIL2,AREA
AATT,3,,1
! Iron Yoke
ASEL,S,LOC,X,r3,r4
ASEL,R,LOC,Y,alpha,beta
ASEL,A,LOC,X,r4,r5
CM,YOKE,AREA
AATT,1,,1
ALLSEL
CSYS,0

! Enter Cartesian Mode

!!! Setup and Mesh Model

MSHKEY,1
CSYS,1
LSEL,S,LOC,Y,0
LSEL,A,LOC,Y,gamma
LESIZE,ALL,,,6,,1,,,1,
CMSEL,S,Inner_Iron
AMESH,ALL
CMSEL,S,Outer_AL
AMESH,ALL
CMSEL,S,Air
AMESH,ALL
CMSEL,S,Coil1
AMESH,ALL
CMSEL,S,Coil2
AMESH,ALL
CMSEL,S,Yoke
AMESH,ALL
ALLSEL
CSYS,0
!!! Reflect Model across X-axis
!! Create HALF model from QUARTER model
arsym,x,all

/prep7
save,magtest,db
arsym,y,all
nummrg,all

! save half model for cyclic

! create full model reflecting on y axis

csys,1
nsel,s,loc,x,r5
CM,extnode,NODE
! Apply BFE Current loads to each coil
*do,i,1,ncoil
asel,s,loc,x,r3,r4
asel,r,loc,y,alpha1(i),alpha2(i)
esla,s
cm,coilname(i),element
bfe,all,js,,,,current(i)
*enddo
csys,0
allsel
cmsel,s,extnode
d,all,az,0

194

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Magnetic Cyclic Symmetry Analysis

d,all,temp,25
allsel
FINISH
/solu
/out,scratch
antype,static
allsel
solve
FINISH
/post1

!!! Plot Out Result Plots

plvect,b,,,,VECT,ELEM,ON,0
FINISH
parsav,all
/clear,nostart
resume,magtest,db
parres,new

! Resume half Model

!! Delete Bottom half of model and all loading attatched to bottom nodes
/prep7
allsel
nummrg,all
csys,1
nsel,s,loc,x,r5
D,all,az,0 ! AZ = 0 on outside nodes of arc
D,all,temp,25
!! Define Coils on Half Model
! Coil 1
ASEL,S,LOC,X,r3,r4
ASEL,R,LOC,Y,0,alpha
esla,s
CM,COIL1,ELEMENT
! Coil 2
ASEL,S,LOC,X,r3,r4
ASEL,R,LOC,Y,beta,(180-beta)
esla,s
CM,COIL2,ELEMENT
! Coil 3
ASEL,S,LOC,X,r3,r4
ASEL,R,LOC,Y,(180-alpha),180
esla,s
CM,COIL3,ELEMENT
!! Apply bfe loads to Half Model coils

cmsel,s,COIL1
bfe,all,js,,,,i2
cmsel,s,COIL2
bfe,all,js,,,,i1
cmsel,s,COIL3
bfe,all,js,,,,(-i2)
!!!

Apply cyclic - create cyclic model with two sectors

allsel
csys,0
cyclic,2
/solution

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

195

Cyclic Symmetry Analysis

cycopt,hindex,odd
solve
FINISH

! Odd Symmetry for half model

/post1
/out

!!! Plot Out Result Plots

/vscale,1,1,1
plvect,b,,,,VECT,ELEM,ON,0

! See figure for B field plot.

!!! Plot Out Contour Line Plot of Equipotentials

plf2d
FINISH

Figure 6.28: Vector Plot of Cyclic Flux Density (B) - Half Model

Figure 6.29: Contour Line Plot of Equipotentials

196

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 7: Rotating Structure Analysis

In a dynamic analysis involving a rotating structure, ANSYS can take inertia effects into account. The
following topics related to rotating structure analysis are available:
7.1. Understanding Rotating Structure Dynamics
7.2. Using a Stationary Reference Frame
7.3. Using a Rotating Reference Frame
7.4. Choosing the Appropriate Reference Frame Option
7.5. Example Campbell Diagram Analysis
7.6. Example Coriolis Analysis
7.7. Example Unbalance Harmonic Analysis
For additional information, see Coriolis Matrix and Coriolis Force in a Rotating Reference Frame in the
Mechanical APDL Theory Reference.

7.1. Understanding Rotating Structure Dynamics

You can observe inertia effects, applied via the CORIOLIS command, in either a stationary reference
frame or a rotating reference frame. In both cases, you specify angular velocity by issuing an OMEGA
or CMOMEGA command.
The dynamic equation incorporating the effect of rotation is given by
+
+
+

where [M], [C] and [K] are the structural mass, damping, and stiffness matrices, respectively.
[Kc] is the spin softening matrix due to the rotation of the structure. It changes the apparent stiffness
of the structure in a rotating reference frame (described in Rotating Structures in the Mechanical APDL
Theory Reference).
[G] is a damping matrix contribution due to the rotation of the structure. It is usually called Coriolis
matrix in a rotating reference frame, and gyroscopic matrix in a stationary reference frame (described
in Rotating Structures in the Mechanical APDL Theory Reference).
{F} is the external force vector in the stationary reference frame. In a rotating reference frame, it is the
sum of the external force and the effect of the angular rotational velocity force (as described in Acceleration Effect in the Mechanical APDL Theory Reference).
Without the inertia effect applied via the CORIOLIS command, ANSYS does not generate the [G] matrix,
and the usual effect of the angular rotation velocity specified by the OMEGA or CMOMEGA command
applies (as described in Acceleration Effect in the Mechanical APDL Theory Reference). An exception exists,
however, involving a nonlinear transient analysis using element MASS21; in this case, the inertia effect
due to rotation applied via an IC command (or a D command over an incremental time) is included
without having to issue the CORIOLIS and OMEGA or CMOMEGA commands.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

197

Rotating Structure Analysis

7.2. Using a Stationary Reference Frame

The primary application for a stationary (rather than a rotating) frame of reference is in the field of rotordynamics where a rotating structure (rotor) is modeled along with a stationary support structure.
Examples of such an application include a gas turbine engine rotor-stator assembly or an electric turbo
generator, where the rotor spins inside a specially designed housing.
The rotating part of the structure to be modeled must be axisymmetric. The gyroscopic damping matrix
generated is valid only for a linear analysis.
ANSYS computes the displacement field with respect to the global coordinate system (CORIOLIS,Option
= ON,,,RefFrame = ON), referred to as the stationary reference frame.
Elements Supported
Elements that are part of the rotating structure generate the gyroscopic matrix that arises due to the
rotational angular velocity. The gyroscopic matrix is available for the elements listed in the notes section
of the CORIOLIS command.
For a beam element, the angular velocity vector is aligned along the length and the point mass is
aligned along one of the principal axes. The rotating structure must be axisymmetric about the spin
axis.
For SHELL281 and other triangular-shaped elements with midside nodes, modeling a shell structure
with the gyroscopic matrix turned on (CORIOLIS,ON, , ,ON) may yield anomalies with the QR damp eigensolver. This is especially true when only a limited number of modes are extracted. In this case, use
the damped eigensolver (MODOPT, DAMP).
Analysis Types Supported
The following analysis types support rotating structure analysis using a stationary reference frame:
Modal (ANTYPE,MODAL)
Transient (ANTYPE,TRANS)
Harmonic (ANTYPE,HARMIC)
For transient and harmonic analyses, the mode-superposition method (TRNOPT, MSUP, or HROPT,MSUP)
is supported for instances where the gyroscopic matrix does not need updating (see below). For the
mode-superposition method, only the QR Damp mode-extraction method (MODOPT,QRDAMP) is supported.
For a varying rotational velocity, mode superposition analysis (transient or harmonic) is not supported,
since the modal gyroscopic matrix is not updated. This is especially true for cases where:
an unbalance or asynchronous rotating force exists in a harmonic analysis (SYNCHRO command)
a start-up or stop simulation is performed in a transient analysis (use the KBC command to ramp the rotational velocity within one loadstep).
To include unbalance or general asynchronous rotating forces in a harmonic analysis, use the SYNCHRO
command.

198

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using a Stationary Reference Frame

For a transient analysis involving a rotating structure with a stationary reference frame, support for a
start or stop simulation is available. Issue the KBC command to ramp the rotational velocity.
For a prestressed analysis that includes gyroscopic effects, issue the CORIOLIS, ON,,,ON command in
the static prestress portion of the analysis.
Postprocessing
Besides general results, the following specific outputs are available:
Campbell diagram (PRCAMP and PLCAMP) see 8.2.1

Note
For a prestressed structure, set the Campbell key (CAMPBELL,ON) in the first solution pass.
Doing so allows a Campbell diagram analysis.

Orbits (PRORB and PLORB) see 8.2.3

Animation of the whirl (ANHARM)

7.2.1. Campbell Diagram

In a modal analysis with multiple load steps corresponding to different angular velocities , a Campbell
diagram (PLCAMP or PRCAMP) shows the evolution of the natural frequencies.
ANSYS determines eigenfrequencies at each load step. The plot showing the variation of eigenfrequency
with respect to rotational speed may not be readily apparent. For example, if the gyroscopic effect is
significant on an eigenmode, its frequency tends to split so much that it crosses the other frequency
curves as the speed increases. For more information, see Generating a Successful Campbell Diagram
below.
Critical Speeds
The PRCAMP command also prints out the critical speeds for a rotating synchronous (unbalanced) or
asynchronous force. The critical speeds correspond to the intersection points between frequency curves
and the added line F=s. (where s represents SLOPE > 0 as specified via PRCAMP). Because the critical
speeds are determined graphically, their accuracy depends upon the quality of the Campbell diagram.
To retrieve and store critical speeds as parameters, use the *GET command.
Whirls and Stability
As eigenfrequencies split with increasing spin velocity, ANSYS identifies forward (FW) and backward
(BW) whirls, and unstable frequencies. To obtain more information to help you determine how a particular frequency becomes unstable, issue the PLCAMP or PRCAMP command and specify a stability value
(STABVAL) of 1. You can also view the logarithmic decrements by specifying STABVAL = 2. For more

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

199

Rotating Structure Analysis

information about complex eigenvalues and corresponding logarithmic decrements, see Complex Eigensolutions in the Mechanical APDL Theory Reference.

Note
For a rotating structure meshed in shell elements lying in a plane perpendicular to the rotational velocity axis - such as a thin disk - the whirl effects are not plotted or printed by the
PRCAMP or PLCAMP commands. However, they can be visualized using the ANHARM
command.
To retrieve and store frequencies and whirls as parameters, use the *GET command.
Prestressed Structure
For a prestressed structure, set the Campbell key (CAMPBELL,ON) in the static solution portion of the
analysis. Doing so modifies the result file so that it can accommodate a subsequent Campbell diagram
analysis. In this case, static and modal solutions are calculated alternately and only the modal solutions
are retained.
Generating a Successful Campbell Diagram
To help you obtain a good Campbell diagram plot or printout, the sorting option is active by default
(PLCAMP,ON or PRCAMP,ON). ANSYS compares complex mode shapes and pairs similar mode shapes.
(Because eigenmodes at zero velocity are real modes, ANSYS does not pair them with complex modes.)
If the plot is unsatisfactory even with sorting enabled, try the following:
Start the Campbell analysis with a non-zero rotational velocity.
Modes at zero rotational velocity are real modes and may be difficult to pair with complex modes
obtained at non-zero rotational velocity.
Increase the number of load steps.
It helps if the mode shapes change significantly as the spin velocity increases.
Change the frequency window.
To do so, use the shift option (PLCAMP,,,FREQB or PRCAMP,,,FREQB). It helps if some modes fall
outside the default frequency window.
Overcoming Memory Problems
To run the Campbell analysis (PRCAMP or PLCAMP), the scratch memory needed may be important as
complex mode shapes are read from the result file for two consecutive load steps. If your computer
has insufficient scratch memory, try the following:
Decrease the number of extracted modes (MODOPT,,NMODE)
Generate the result file for a reduced set of selected nodes (for example, nodes on the axis of rotation).
Issue OUTRES,ALL,NONE and then OUTRES,Item,Freq,Cname where Item=NSOL, Freq=ALL and Cname
is the name of a node-based component.

200

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using a Stationary Reference Frame

For the sorting process and whirl calculation to be successful, the set of selected nodes must represent
the dynamics of the structure. In general, nodes on the spin axis contribute to the bending mode
shapes that are needed in the Campbell analysis.
Example Analysis
For an example of a rotating structure analysis using a stationary reference frame, see Example Campbell
Diagram Analysis (p. 205).

7.2.2. Harmonic Analysis for Unbalance or General Rotating Asynchronous

Forces
Some forces may rotate synchronously (for example, unbalance) or asynchronously with the structure.
In such cases, use the SYNCHRO command to update the amplitude of the rotational velocity vector
with the frequency of excitation at each frequency step of the harmonic analysis.
Forces are defined as static (F), as shown in this example where X is the assumed spin axis:
Force

Real (VALUE)

Imaginary (VALUE2)

FY

F0cos

-F0sin

FZ

-F0sin

-F0cos

where:
F0 is the amplitude of the force. For unbalance, the amplitude is equal to the mass times
the distance of the unbalance mass to the spin axis.
is the phase of the force, needed only when several such forces, each with a different
relative phase, are defined.
If the forces are caused by an unbalance mass, multiplication of the amplitude of the static forces (F)
by the square of the spin velocity is unnecessary. ANSYS performs the calculation automatically at each
frequency step.
Because the rotational velocity commands (OMEGA and CMOMEGA) define only the orientation of the
spin axis, a harmonic analysis using the SYNCHRO command requires that you define the frequency
of excitation (HARFRQ) instead. For example, if the frequency of excitation is f, then:
= 2f/RATIO
where:
is the new magnitude of the rotational velocity vector used to calculate the gyroscopic
matrices.
RATIO is the ratio between the frequency of excitation and the frequency of the rotational velocity of the structure, as specified via the SYNCHRO command. If no RATIO
value is specified, an unbalance force is assumed; in all other cases, a general rotating
force is assumed.
Example Analysis

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

201

Rotating Structure Analysis

For an example of a harmonic analysis for unbalance forces, see Example Unbalance Harmonic Analysis (p. 210).

7.2.3. Orbits
When a structure is rotating about an axis and undergoes vibration motion, the trajectory of a node
executed around the axis is generally an ellipse designated as a whirl orbit.
In a local coordinate system xyz where x is the spin axis, the ellipse at node I is defined by semi-major
axis A, semi-minor axis B, and phase (PSI), as shown:

e
s

Angle (PHI) defines the initial position of the node (at t = 0). To compare the phases
of two nodes of the structure, you can examine the sum + .
Values YMAX and ZMAX are the maximum displacements along y and z axes, respectively.
You can print out the A, B, PSI, PHI, YMAX, and ZMAX values via a PRORB (print orbits)
command. Angles are in degrees and within the range of -180 through +180. The position
vector of local axis y in the global coordinate system is printed out along with the elliptical orbit characteristics. You can also animate the orbit (ANHARM) for further examination. For a typical usage example of these commands, see Example Unbalance Harmonic
Analysis (p. 210).
To retrieve and store orbits characteristics as parameters, use the *GET command after issuing the
PRORB command.

7.3. Using a Rotating Reference Frame

The primary application for a rotating (rather than a stationary) frame of reference is in the field of
flexible body dynamics where, generally, the structure has no stationary parts and the entire structure
is rotating. Analyses of this type, therefore, consider only the Coriolis force.

Note
The gyroscopic effect is not included in the dynamics equations expressed in a rotating reference frame. Therefore, if the structure contains a part with large inertia - such as a large
disk - the results obtained in the rotating reference frame may not compare well with stationary reference frame results.

202

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using a Rotating Reference Frame

ANSYS computes the displacement field with respect to the coordinate system attached to the structure
and rotating with it at the specified angular velocity (CORIOLIS,Option = ON,,,RefFrame = OFF).
Elements Supported
The Coriolis matrix and forces are available for the structural elements listed in the notes section of the
CORIOLIS command.
Analysis Types Supported
The following analysis types support rotating structure analysis using a rotating reference frame:
Static (ANTYPE,STATIC)
Inertia effects are forces computed by multiplying the Coriolis damping matrix by the velocity of the
structure.
If you issue the CORIOLIS command in a prestressed analysis, ANSYS does not take the Coriolis force
into account in the static portion of the analysis.
In a large-deflection prestressed analysis (NLGEOM,ON and PSTRES,ON), ANSYS generates the Coriolis
matrix and uses it in the subsequent prestressed modal, harmonic, or transient analysis.
In a small-deflection prestressed analysis (PSTRES,ON only), ANSYS does not generate the Coriolis
matrix but still takes the Coriolis force into account in the subsequent prestressed modal, harmonic, or
transient analysis.
Modal (ANTYPE,MODAL)
Support is also available for prestressed modal analysis.
Transient (ANTYPE,TRANS)
Harmonic (ANTYPE,HARMIC)
Spin Softening
In a dynamic analysis, the Coriolis matrix and the spin-softening matrix contribute to the gyroscopic
moment in the rotating reference frame; therefore, ANSYS includes the spin-softening effect by default
in dynamic analyses whenever you apply the Coriolis effect in the rotating reference frame (CORIOLIS,ON).
Supercritical Spin Softening
As shown by equations (3-77) through (3-79) in the Mechanical APDL Theory Reference,
the diagonal coefficients in the stiffness matrix become negative when the rotational
velocity is larger than the resonant frequency.
In such cases, the solver may be unable to properly handle the negative definite stiffness
matrix. Additional details follow:
In a static (ANTYPE,STATIC), a full transient (ANTYPE,TRANS with TRNOPT,FULL), or a full
harmonic (ANTYPE,HARM with TRNOPT,FULL) analysis, the spin-softening effect is more
accurately accounted for by large deflections (NLGEOM,ON). If the stiffness matrix becomes
negative definite, ANSYS issues a warning message about the negative pivot.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

203

Rotating Structure Analysis

In a modal analysis (ANTYPE,MODAL), apply a negative shift (MODOPT,,, FREQB) to extract
the possible negative eigenfrequencies.
If negative frequencies exist, mode-superposition transient and harmonic analyses are not
supported .
Coriolis Effect in a Nonlinear Transient Analysis
In a nonlinear transient analysis with large deflection effects (NLGEOM, ON ), rotation
motion imparted through either the IC command or the D command contributes the
Coriolis effect as part of the nonlinear transient algorithm. The CORIOLIS command
should not be activated in this case. However, beam elements (BEAM188 and BEAM189),
and pipe elements (PIPE288 and PIPE289 ) may produce approximate results when simulating Coriolis effect as above, due to the approximations involved in their inertia calculations.
Campbell Diagram
Because natural frequencies are subject to sudden changes around critical speeds in a
rotating frame, ANSYS, Inc. recommends using a stationary reference frame to create a
Campbell diagram (PRCAMP or PLCAMP).
Example Analysis
For examples of a rotating structure analysis using a rotating reference frame, see Example Coriolis
Analysis (p. 207), and Example: Piezoelectric Analysis with Coriolis Effect in the Coupled-Field Analysis
Guide.

7.4. Choosing the Appropriate Reference Frame Option

The rotating and stationary reference frame approaches have their benefits and limitations. Use this
table to choose the best option for your application:
Reference Frame Considerations
Stationary Reference Frame

Rotating Reference Frame

Not applicable to a static analysis (ANTYPE,STATIC).

In a static analysis, a Coriolis force vector is given

by

represents the nodal velocity vector

where
(specified via the IC command).
You can generate Campbell plots for computing rotor Campbell plots are not applicable for computing
critical speeds.
rotor critical speeds.
Structure must be axisymmetric about the spin axis.

Structure need not be axisymmetric about the

spin axis.

Rotating structure can be part of a stationary structure in an analysis model (such as a gas turbine engine rotor-stator assembly). The stationary structure
and supports (such as bearings) need not be
axisymmetric.

Rotating structure must be the only part of an

analysis model (such as a gas turbine engine rotor).

204

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Campbell Diagram Analysis

Reference Frame Considerations
Stationary Reference Frame

Rotating Reference Frame

Supports more than one rotating structure spinning

at different rotational speeds about different axes of
rotation (such as a multi-spool gas turbine engine).

Supports only a single rotating structure (such as

a single-spool gas turbine engine).

See the CORIOLIS command for the list of elements

supported in the Stationary Reference Frame.

See the CORIOLIS command for the list of elements supported in the Rotating Reference Frame.

Natural Frequencies
Natural frequencies differ according to the reference frame type. In most cases, natural frequencies are
known in a stationary reference frame through analytical expressions or experiment, for example. ANSYS
therefore recommends using the stationary reference frame for modal analyses.

7.5. Example Campbell Diagram Analysis

Following is a modal analysis of a rotating structure using a stationary reference frame. The analysis
generates a Campbell diagram (PLCAMP).

7.5.1. Problem Description

The model is a simply supported beam spinning at up to 30,000 rd/s.

7.5.2. Problem Specifications

The geometric properties for this analysis are as follows:
Length: 8m
Diameter: 0.2m
The material properties for this analysis are as follows:
Young's modulus (E) = 2e+11 N/m2
Poisson's ratio () = 0.3
Density = 7800 kg/m3

7.5.3. Input for the Analysis

Use this input file to perform the example modal analysis of a rotating structure using a stationary reference frame.
//batch,list
/title, Spinning Simply Supported Beam
!* Parameters
lx=8
! length
dia=0.2
! diameter
/PREP7
! ----ET,1,16
R,1, dia, dia/2
MP,EX,1,2e+11
MP,DENS,1,7800
MP,PRXY,1,0.3
n,1
n,9,lx
fill,1,9
e,1,2

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

205

Rotating Structure Analysis

egen,8,1,-1
d,1,uy,,, ,,uz
! simply supported left end
d,9,uy,,, ,,uz
! simply supported right end
d,all,ux
! supress axial motion
d,all,rotx
! supress torsion
finish
/SOLU
! ----antype,modal
! Use the QRDAMP eigensolver, request 8 modes,
!
get complex eigensolutions
modopt,qrdamp,8,,,on
! Reuse the block Lanczos eigenvectors from the first loadstep
qrdopt,on
! Write 8 modes to the result file, calculate
!
element results
mxpand,8,,,yes
! Apply Coriolis effect and specify
!
stationary reference frame
coriolis,on,,,on
! Solve 2 loadsteps with rotational velocity
omega,0.
solve
omega,30000.
solve
finish
/POST1
! ----! Plot Campbell Diagram
plcamp
! Print Campbell Diagram
prcamp
finish

7.5.4. Analysis Steps

The following table describes the input listing and the general process involved in the example analysis
in more detail:
Step
1.

Description

ANSYS Command(s)

Set parameters.

lx=8
dia=0.2

2.

Define nodes, elements, and material properties.

ET,
R,
MP,...
N,
E,

3.

Set boundary conditions.

D,...

4.

Set the analysis type (modal in this case). Use the QRDAMP eigensolver, request 8 modes, and specify complex eigensolutions.

ANTYPE,MODAL
MODOPT,QRDAMP,8,,,ON
MXPAND,...

5.

206

Include the Coriolis effect in a stationary reference frame.

CORIOLIS,ON,,,ON

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Coriolis Analysis

Step

Description

6.

ANSYS Command(s)

For each load step, define the rotation velocity, and then solve. OMEGA,
SOLVE

7.

Plot and print the Campbell diagram.

PLCAMP
PRCAMP

The results of your analysis should be similar to those shown here:

The printout (PRCAMP) should yield the following data:

PRINT CAMPBELL DIAGRAM
Sorting : ON
X axis unit : rd/s

***** FREQUENCIES (Hz) FROM CAMPBELL (sorting on) *****

Spin(rd/s)
1
2
3
4
5
6
7
8

BW
FW
BW
FW
BW
FW
BW
FW

0.000

30000.000

6.207
6.207
24.750
24.750
55.461
55.461
98.248
98.248

4.639
8.305
18.547
33.027
41.735
73.701
74.337
129.852

7.6. Example Coriolis Analysis

Following is an example modal analysis with Coriolis force applied (CORIOLIS) in a rotating reference
frame.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

207

Rotating Structure Analysis

7.6.1. Problem Description

The model is a thick cylindrical shell rotating about its center axis at an angular velocity = 50 Hz.

7.6.2. Problem Specifications

The geometric properties for this analysis are as follows:
Length: 0.254m
Radius: 0.09525m
Thickness: 0.0381m.
The material properties for this analysis are as follows:
Young's modulus (E) = 2.07 x 1011 N/m2
Poisson's ratio () = 0.28
Density = 7.86 x 103 Kg/m3

7.6.3. Input for the Analysis

Use this input file to perform the example modal analysis of a rotating structure using a rotating reference
frame.
/batch,list
/PREP7
!*
et,1,185
keyopt,1,2,2
mp,ex,1,2.07e11
mp,nuxy,1,0.28
mp,dens,1,7860
!* Parameters
thick1 = 0.03810
thick = thick1
hthick = thick/2.0
radius = 0.09525
inradius = radius - hthick
outradius = radius + hthick
length = 0.254
!*
cylind,inradius,outradius,0,length,0,90
cylind,inradius,outradius,0,length,90,180
cylind,inradius,outradius,0,length,180,270
cylind,inradius,outradius,0,length,270,360
nummrg,kp

208

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Coriolis Analysis

esize,thick
vmesh,all
nsel,all
finish
!*
/SOLU
antype,modal
modopt,qrdamp,20
omega,0,0,314.16,1
!
! Apply Coriolis effect and use default
!
rotating reference frame
coriolis,on
!
solve
finish

7.6.4. Analysis Steps

The following table describes the input listing and the general process involved in the example analysis
in more detail.
Step
1.

Description

ANSYS Command(s)

Define elements and material properties.

ET,
MP,

2.

Set parameters.

thick1 = 0.03810
thick = thick1
hthick = thick/2.0
radius = 0.09525
inradius = radius
- hthick
outradius = radius + hthick
length = 0.254

3.

Set boundary conditions.

CYLIND,...
NUMMRG,...
ESIZE,...
VMESH,...
NSEL,...

4.

Set the analysis type (modal in this case) and use the QRDAMP ANTYPE,MODAL
eigensolver. Get 20 complex eigensolutions.
MODOPT,QRDAMP,20

6.

Define the angular rotation velocity and the number of modes OMEGA,
expanded.
MXPAND,
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

209

Rotating Structure Analysis

Step

Description

ANSYS Command(s)

5.

Include the Coriolis effect in a rotating reference frame.

CORIOLIS,ON

7.

Solve.

SOLVE

The non-zero complex frequencies should match the following:

With Coriolis
Without Coriolis
Effect Applied
Effect Applied
---------------------------------------------------------0.0000000
2661.9384
2700.8924
j
0.0000000
2740.4150
2700.8924
j
0.0000000
3043.0948
3078.1392
j
0.0000000
3113.5872
3078.1392
j
0.0000000
6342.7190
6362.7716
j
0.0000000
6365.8292
6362.7716
j
0.0000000
6382.8876
6366.7082
j
0.0000000
6593.8155
6626.0560
j
0.0000000
6658.4573
6626.0560
j
0.0000000
7183.0372
7216.0329
j

7.7. Example Unbalance Harmonic Analysis

Following is an example harmonic analysis with unbalance force. It illustrates the use of the SYNCHRO
command and the following postprocessing capabilities:
Orbits plotting (PLORB)
Whirl animation (ANHARM)

7.7.1. Problem Description

The structure is a two-spool rotor on symmetric bearings. Both spools have two rigid disks. The inner
spool rotates at up to 14,000 RPM and the outer spool rotates 1.5 times faster.

Disks are not visible in the plot because they are MASS21 elements.

7.7.2. Problem Specifications

The unbalance is located on the second disk of the inner spool and harmonic response is calculated.
Outputs are as follows:
210

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Unbalance Harmonic Analysis

Amplitude at nodes 7 and 12 as a function of the frequency
Orbit plot at a given frequency
Animation of the whirl at a given frequency

7.7.3. Input for the Analysis

Use this input file to perform the example unbalance harmonic analysis of rotating structure using a
stationary reference frame.
/batch,list
/title, twin spools - unbalance (inner spool) response
/PREP7
mp,EX ,1,2.1e+11
mp,DENS,1,7800
mp,PRXY,1,0.3
! shaft
et,1,188,,,2
sectype,1,beam,csolid
secdata,0.01524,32
sectype,2,beam,ctube
secdata,0.0254,0.03048,32
! disks
et,2,21
r,3,10.51,10.51,10.51,8.59e-2,4.295e-2,4.295e-2
r,4,7.01 ,7.01 ,7.01 ,4.29e-2,2.145e-2,2.145e-2
r,5,3.5 ,3.5 ,3.5 ,2.71e-2,1.355e-2,1.355e-2
r,6,7.01 ,7.01 ,7.01 ,6.78e-2,3.390e-2,3.390e-2
! bearings
et,3,214,,1
r,7 ,2.63e+7 ,2.63e+7
r,8 ,1.75e+7 ,1.75e+7
r,9 ,0.875e+7,0.875e+7
r,10,1.75e+7 ,1.75e+7
! nodes
n,1
n,2 ,0.0762
n,3 ,0.1524
n,4 ,0.2413
n,5 ,0.32385
n,6 ,0.4064
n,7 ,0.4572
n,8 ,0.508
n,9 ,0.1524
n,10,0.2032
n,11,0.2794
n,12,0.3556
n,13,0.4064
! bearings second nodes
n,101,
,0.05
n,108,0.508 ,0.05
n,109,0.1524,0.05
! components elements
type,1
secn,1
e,1,2
egen,7,1,1
type,2
real,3
e,2
real,6
e,7
cm,inSpool,elem
type,1
secn,2
e,9,10
egen,4,1,10
type,2
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

211

Rotating Structure Analysis

real,4
e,10
real,5
e,12
esel,u,,,inSpool
cm,outSpool,elem
allsel
! bearings
type,3
real,7
e,1,101
real,8
e,9,109
real,9
e,6,13
real,10
e,8,108
! boundary conditions
d,all,ux,,,,,rotx
d,101,all
d,108,all
d,109,all
! unbalance forces (eccentric mass * radius)
f0 = 70e-6
f,7,fy,f0
f,7,fz,,-f0
fini
/SOLU
antype,harmic
synchro,,inSpool
nsubst,500
harfrq,,14000/60 ! implicitly defines OMEGA for Coriolis calculation
kbc,1
dmprat,0.01
cmomega,inSpool,100.
cmomega,outSpool,150.
coriolis,on,,,on
solve
fini
! output: amplitude at nodes 7 and 12 as a function of the frequency
/POST26
nsol,2,7,U,Y,UY
nsol,3,7,U,Z,UZ
realvar,4,2,,,UYR
realvar,5,3,,,UZR
prod,6,4,4,,UYR_2
prod,7,5,5,,UZR_2
add,8,6,7,,UYR_2+UZR_2
sqrt,9,8,,,AMPL7
!
nsol,2,12,U,Y,UY
nsol,3,12,U,Z,UZ
realvar,4,2,,,UYR
realvar,5,3,,,UZR
prod,6,4,4,,UYR_2
prod,7,5,5,,UZR_2
add,8,6,7,,UYR_2+UZR_2
sqrt,10,8,,,AMPL12
!
/gropt,logy,1
/yrange,1.e-7,1.e-3
plvar,9,10
fini
! output: orbit plot at the given frequency
/POST1
set,1,262
/view,,1,1,1
plorb
! output: animation of the whirl at the given frequency
SET,1,500
!reset for subsequent post processing
eshape,1

212

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Unbalance Harmonic Analysis

/gline,,-1
plnsol,u,sum
anharm

7.7.4. Analysis Steps

The following table describes the input listing and the general process involved in the example analysis
in more detail.
Step
1.

Description

ANSYS Command(s)

Define material properties.

MP,EX,1,2.1e+11
MP,DENS,1,7800
MP,PRXY,1,0.3

2.

Define element types, sections, real and nodes.

ET,
SECTYPE,
SECDATA,
R,
N,

3.

Define first component named inSpool.

TYPE,1
SECNUM,1
E,1,2
EGEN,7,1,1
TYPE,2
REAL,3
E,2
REAL,6
E,7
CM,inSpool,ELEM

4.

Define second component named outSpool.

TYPE,1
SECNUM,2
E,9,10
EGEN,4,1,10
TYPE,2
REAL,4

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

213

Rotating Structure Analysis

Step

Description

ANSYS Command(s)
E,10
REAL,5
E,12
ESEL,u,,,inSpool
CM,outSpool,ELEM
ALLSEL

6.

Define bearing elements.

TYPE,3
REAL,7
E,1,101
REAL,8
E,9,109
REAL,9
E,6,13
REAL,10
E,8,108

5.

Set boundary conditions.

D,...

7.

Define the unbalance forces (eccentric mass * radius)

at node 7.

f0 = 70e-6
F,7,FY,f0
F,7,FZ,,-f0

8.

Set the solution options.

ANTYPE,HARMIC

Harmonic analysis

SYNCHRO,,inSpool

Unbalance on component inSpool

NSUBST,500

500 substeps

HARFRQ,,14000/60

Frequency at end of range is 14000/60 Hz

KBC,1

Step loading

DMPRAT,0.01

Damping ratio is 1%

CMOMEGA,inSpool,100.

Rotational velocity of component inSpool

CMOMEGA,outSpool,150.

Rotational velocity of component outSpool

CORIOLIS,ON,,,ON

214

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Unbalance Harmonic Analysis

Step

Description

ANSYS Command(s)

Coriolis force in stationary reference frame

SOLVE

Note
The rotational velocities (CMOMEGA) are
not applied in the usual way. Rather, ANSYS considers only their direction cosines
and the velocity ratio between spools. For
more information, see the documentation
for the SYNCHRO command.
9.

First output (in POST26).

/POST26

The maximum amplitude of the displacement of nodes

7 (AMPL7) is calculated in variable 9.

NSOL,2,7,U,Y,UY
NSOL,3,7,U,Z,UZ

The maximum amplitude of the displacement of nodes

12 (AMPL12) is calculated in variable 10.

REALVAR,4,2,,,UYR

Set logY scale.

REALVAR,5,3,,,UZR

Set a specific scale range in Y.

PROD,6,4,4,,UYR_2

Plot variables 9 and 10.

PROD,7,5,5,,UZR_2
ADD,8,6,7,,UYR_2+UZR_2
SQRT,9,8,,,AMPL7
!
NSOL,2,12,U,Y,UY
NSOL,3,12,U,Z,UZ
REALVAR,4,2,,,UYR
REALVAR,5,3,,,UZR
PROD,6,4,4,,UYR_2
PROD,7,5,5,,UZR_2
ADD,8,6,7,,UYR_2+UZR_2
SQRT,10,8,,,AMPL12
!
/GROPT,LOGY,1
/YRANGE,1.e-7,1.e-3

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

215

Rotating Structure Analysis

Step

Description

ANSYS Command(s)
PLVAR,9,10

10.

/POST1

Read load step 1 and substep 262 from results file.

SET,1,262

Change the view.

/VIEW,,1,1,1

Plot the orbits at each rotating node.

PLORB

Third output.

/ESHAPE,1

Display takes dimensions into account.

/GLINE,,-1

No element outline.

PLNSOL,U,SUM

Display the displacements as contours.

11.

Second output (in POST1).

ANHARM

Animate the displays (defined by the last set command

and last display command).
The outputs of your analysis should match those shown here:

You can obtain the two critical frequencies (at which the amplitudes are largest) via
PRCAMP with SLOPE = 1.0.

216

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Unbalance Harmonic Analysis

Orbits are represented in different colors. Orbits from the inner spool appear in sky blue,
and from the outer spool in purple. Spool lines appear in dark blue.
The following demo is presented as an animated GIF. Please view online if you are reading the PDF version
of the help. Interface names and other components shown in the demo may differ from those in the released
product.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

217

Rotating Structure Analysis

The animation of the whirls shown here is the third output resulting from the example
harmonic analysis with unbalance.

218

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 8: Submodeling
Submodeling is a finite element technique that you can use to obtain more accurate results in a particular region of a model. A finite element mesh may be too coarse to produce satisfactory results in a
given region of interest. The results away from this region, however, may be satisfactory.
Reanalyzing the entire model using a greater mesh refinement in order to obtain more accurate results
in one particular region is time-consuming and costly. Instead, you can use submodeling to generate
an independent, more finely meshed model of only the region (submodel) of interest and then analyze
it.
The following submodeling topics are available:
8.1. Understanding Submodeling
8.2. Using Submodeling
8.3. Example Submodeling Analysis Input
8.4. Shell-to-Solid Submodels
8.5. Where to Find Examples

8.1. Understanding Submodeling

In finite element analysis, the finite element mesh is sometimes too coarse to produce satisfactory results
in a specific region of interest, such as a stress concentration region in a stress analysis as shown in
Figure 8.1: Submodeling of a Pulley (p. 219). The figure illustrates how to deal with the problem by using
submodeling to create a finer mesh on the region (submodel) of interest.
Figure 8.1: Submodeling of a Pulley

Submodeling of a pulley hub and spokes: (a) coarsely meshed model, and (b) finely
meshed submodel (shown superimposed over coarse model)
Submodeling is also known as the cut-boundary displacement method (also known as the specified
boundary displacement method). The cut boundary is the boundary of the submodel which represents

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

219

Submodeling
a cut through the coarse model. Displacements calculated on the cut boundary of the coarse model
are specified as boundary conditions for the submodel.
Submodeling is based on St. Venant's principle, which states that if an actual distribution of forces is
replaced by a statically equivalent system, the distribution of stress and strain is altered only near the
regions of load application. The principle implies that stress concentration effects are localized around
the concentration; therefore, if the boundaries of the submodel are far enough away from the stress
concentration, reasonably accurate results can be calculated in the submodel.
The program does not restrict submodeling to structural (stress) analyses only. Submodeling is effective
in other disciplines as well. For example, in a magnetic field analysis, you can use submodeling to calculate more accurate magnetic forces in a region of interest.
Aside from the obvious benefit of yielding more accurate results in a region of your model, the submodeling technique has other advantages:
It reduces, or even eliminates, the need for complicated transition regions in solid finite element
models.
It enables you to experiment with different designs for the region of interest (different fillet radii, for
example).
It helps you in demonstrating the adequacy of mesh refinements.
The following restrictions apply to submodeling:
It is valid only for solid elements and shell elements.
The principle behind submodeling assumes that the cut boundaries are far enough away from the
stress concentration region. You must verify that this assumption is adequately satisfied.

8.1.1. Nonlinear Submodeling

For load-history-dependent problems (for example, when plastic materials exist), you must cut boundary
conditions from the coarse model at multiple substeps to simulate the load history dependency in the
fine-mesh model analysis.
In a fine-mesh-model analysis, the cut boundary conditions must be applied as loads at multiple load
steps. The more boundary cutting you do, the more accurate are the results of the fine-mesh-model
analysis.
Nonlinear submodeling does not apply to shell-to-solid submodels.

8.2. Using Submodeling

The process for using submodeling is as follows:
1. Create and analyze the coarse model.
2. Create the submodel.
3. Perform cut-boundary interpolation.
4. Analyze the submodel.

220

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Submodeling
5. Verify that the distance between the cut boundaries and the stress concentration is adequate.
The steps are explained in detail next.

8.2.1. Create and Analyze the Coarse Model

The first step is to model the entire structure and analyze it.
To easily identify this initial model, it is referred to as the coarse model. This does not mean that the
mesh refinement is coarse, only that it is relatively coarse compared to the submodel.
The analysis type may be static (steady-state) or transient and follows the same procedure as described
in the individual analysis guides. Some additional points to keep in mind are listed below.
Jobname - Use different job names for the coarse model and the submodel as a precaution against
overwriting files, and to easily refer to files from the coarse model during cut-boundary interpolation.
To specify a job name, use one of these methods:
Command(s): /FILNAME
GUI: Utility Menu> File> Change Jobname
Element Types -- Only solid and shell elements support the submodeling technique. Your analysis may
include other element types (such as beams added as stiffeners), but the cut boundary should only
pass through the solids or shells.
A special technique called shell-to-solid submodeling allows you to build your coarse model with shell
elements and your submodel with 3-D solid elements. This technique is discussed in Shell-to-Solid
Submodels (p. 233).
Modeling -- In many cases, the coarse model need not include local details such as fillet radii, as shown
in the following figure. However, the finite element mesh must be fine enough to produce a reasonably
accurate degree of freedom solution. This is important because the results of the submodel are almost
entirely based on interpolated degree of freedom results at the cut boundary.
Figure 8.2: Coarse Model

Initial, coarse model may not need to include many details

Files - Both the results file (Jobname.RST, Jobname.RMG, etc.) and the database file (Jobname.DB,
containing the model geometry) are required from the coarse-model analysis. Be sure to save the

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

221

Submodeling
database after solving the coarse model, but before going on to create the submodel. To save the
database, use either of these methods:
Command(s): SAVE
GUI: Utility Menu> File> Save as
Utility Menu> File> Save as Jobname.db

8.2.2. Create the Submodel

The submodel is completely independent of the coarse model; therefore, the first step after the initial
analysis is to clear the database at the Begin level. To clear the database at the Begin level, use one of
these methods:
Command(s): /CLEAR
GUI: Utility Menu> File> Clear & Start New
Also, be sure to use a different job name for the submodel so that the coarse-model files are not overwritten. To specify a job name, use one of these methods:
Command(s): /FILNAME
GUI: Utility Menu> File> Change Jobname
Then enter PREP7 and build the submodel. Some points to remember are:
Use the same element type (solid or shell) that was used in the coarse model. Also, specify the same
element real constants (such as shell thickness) and material properties. (Shell-to-solid submodeling
allows you to switch from shell elements in the coarse model to 3-D solid elements in the submodel.)
The location of the submodel (with respect to the global origin) must be the same as the corresponding
portion of the coarse model, as shown in Figure 8.3: Submodel Superimposed Over Coarse Model (p. 222).
Figure 8.3: Submodel Superimposed Over Coarse Model

Specify appropriate node rotations. Node rotation angles on cut boundary nodes should not be
changed after they have been written to the node file in interpolation step 1 (see Perform CutBoundary Interpolation (p. 223)). To specify node rotations, use one of these methods:
Command(s): NROTAT
GUI: Main Menu> Preprocessor> Modeling> Create> Nodes> Rotate Node CS> To Active CS
Main Menu> Preprocessor> Modeling> Move/Modify> Rotate Node CS> To Active CS

222

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Submodeling
Be aware that node rotation angles might be changed by application of nodal constraints (DSYM), by
transfer of line constraints (SFL), or by transfer of area constraints (SFA), as well as by more obvious
methods (NROTAT and NMODIF).
The presence or absence of node rotation angles in the coarse model has no effect upon the submodel.
Loads and boundary conditions for the submodel are covered in subsequent steps.

8.2.3. Perform Cut-Boundary Interpolation

This is the key step in submodeling. You identify the nodes along the cut boundaries, and the program
calculates the degree-of-freedom values (displacements, potentials, etc.) at those nodes by interpolating
results from the full (coarse) model. For each node of the submodel along the cut boundary, the program
uses the appropriate element from the coarse mesh to determine the degree-of-freedom values. These
values are then interpolated onto the cut-boundary nodes using the element shape functions.
The following tasks are involved in performing the cut-boundary interpolation:
1. Identify and write the cut-boundary nodes of the submodel to a file (Jobname.NODE by default).
You can do this in PREP7 by selecting nodes along the cut boundaries and then using one of these
methods to write the nodes to a file:
Command(s): NWRITE
GUI: Main Menu> Preprocessor> Modeling> Create> Nodes> Write Node File
Here is an example of issuing the NWRITE command:
NSEL,...
NWRITE

! Select nodes along cut boundaries

! Writes all selected nodes to Jobname.NODE

Figure 8.4: Cut Boundaries on the Submodel

At this point, it is worthwhile to discuss temperature interpolation. In an analysis with temperaturedependent material properties, or in a thermal-stress analysis, the temperature distribution
must be the same between the coarse model and the submodel. For such cases, you must also
interpolate the temperatures from the coarse model to all nodes in the submodel. To do so,
select all submodel nodes and write them to a different file using NWRITE,Filename,Ext.
Specify a file name; otherwise, your file of cut-boundary nodes will be overwritten. Step 7 shows
the command to do temperature interpolation.
2. Restore the full set of nodes, write the database to Jobname.DB, and leave PREP7. You must write
the database to Jobname.DB because you need to continue with the submodel later.
To restore the full set of nodes, use one of these methods:
Command(s): ALLSEL
GUI: Utility Menu> Select> Everything
To write the database to Jobname.DB, use one of these methods:
Command(s): SAVE
GUI: Utility Menu> File> Save as Jobname.db
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

223

Submodeling
3. To do the cut-boundary interpolation (and the temperature interpolation), the database must contain
the geometry for the coarse model. Therefore, you must resume the database using one of the
methods shown below, making sure to identify the name of the coarse-model database file:
Command(s): RESUME
GUI: Utility Menu> File> Resume from
For example, if the job name for the coarse-model analysis was COARSE, issue the command
RESUME,COARSE,DB.
4. Enter POST1, which is the general postprocessor (/POST1 or menu path Main Menu> General
Postproc). Interpolation can only be performed in POST1.
5. Point to the coarse results file (FILE or menu path Main Menu> General Postproc> Data & File
Opts).
6. Read in the desired set of data from the results file (SET or menu path Main Menu> General
Postproc> Read Results> option).
7. Initiate cut-boundary interpolation. To do so, use one of these methods:
Command(s): CBDOF
GUI: Main Menu> General Postproc> Submodeling> Interpolate DOF
By default, the CBDOF command assumes that the cut-boundary nodes are on file Jobname.NODE. The program then calculates the cut-boundary degree-of-freedom values and
write them in the form of D commands to the file Jobname.CBDO.
To do temperature interpolation, use one of these methods, being sure to identify the name
of the file containing all submodel nodes:
Command(s): BFINT
GUI: Main Menu> General Postproc> Submodeling> Interp Body Forc
Interpolated temperatures are written in the form of BF commands to the file Jobname.BFIN.
If real and imaginary data are involved, steps 6 and 7 must be performed twice. Issue the SET
command to get the real data, followed by the interpolation step (CBDOF and/or BFINT). Issue
the SET command with the field set to 1 to get the imaginary data, and repeat the interpolation
step, this time writing the interpolated imaginary data to a different file.
For load-history-dependent problems, steps 6 and 7 can be performed multiple times at different
data sets or substeps. The cut-boundary degree-of-freedom values of each data set will reside
in the same file but in different data blocks, separated by a :CBnn label (where nn is the cumulative iteration number of the results set used).
8. All interpolation work is now done, so leave POST1 (FINISH) and restore the submodel database
(RESUME or menu path Utility Menu> File> Resume from). (Be sure to use the submodel database
job name.)

8.2.4. Analyze the Submodel

In this step, you define the analysis type and analysis options, apply the interpolated degree-of-freedom
values (and temperatures), define other loads and boundary conditions, specify load step options, and
obtain the submodel solution.
The first step is to enter the solution processor (/SOLU or menu path Main Menu> Solution).

224

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Submodeling
Then define the appropriate analysis type (usually static) and analysis options.
To apply the cut-boundary degree-of-freedom constraints, simply read in the file of D commands (created
by CBDOF) using one of these methods (for example, /INPUT,,CBDO):
Command(s): /INPUT
GUI: Utility Menu> File> Read Input from
Similarly, to apply the interpolated temperatures, read in the file of BF commands (created by BFINT)
using one of these methods (for example, /INPUT,,BFIN):
Command(s): /INPUT
GUI: Utility Menu> File> Read Input from
If real and imaginary data are involved, first read in the file(s) containing the real data, specify whether
degree-of-freedom constraint values and/or nodal body force loads are to be accumulated, and then
read in the file containing the imaginary data.
Specify that degree-of-freedom constraint values are to be accumulated:
Command(s): DCUM,ADD
GUI: Main Menu> Preprocessor> Loads> Define Loads> Settings> Replace vs Add> Constraints
Main Menu> Solution> Define Loads> Settings> Replace vs Add> Constraints
Specify that nodal body force loads are to be accumulated:
Command(s): BFCUM,,ADD
GUI: Main Menu> Preprocessor> Loads> Define Loads> Settings> Replace vs Add> Nodal Body
Ld
Main Menu> Solution> Define Loads> Settings> Replace vs Add> Nodal Body Ld
Be sure to reset the DCUM and BFCUM commands to their default status before proceeding.
It is important that you duplicate on the submodel any other loads and boundary conditions that existed
on the coarse model. Examples are symmetry boundary conditions, surface loads, inertia forces (such
as gravity), concentrated force loads, etc. (see Figure 8.5: Loads on the Submodel (p. 226)).
Then specify load step options (such as output controls) and initiate solution calculations using one of
these methods:
Command(s): SOLVE
GUI: Main Menu> Solution> Solve> Current LS
Main Menu> Solution> Run FLOTRAN
For most problems, exit the solution processor (FINISH) after the solution has completed.
For load-history-dependent problems, apply the cut-boundary degree-of-freedom constraints in the
next data block in the file created by cut-boundary interpolation (CBDOF), and solve (SOLVE) as a new
load step. Because it is a new load step, for any rate-dependent materials and transient analysis, the
time value must match the time when cut-boundary interpolation occurred. This step can be repeated
until all cut-boundary degree-of-freedom constraints are applied. Then, exit the solution processor.
The overall data flow for submodeling (without temperature interpolation) is shown in Figure 8.6: Data
Flow Diagram for Submodeling (Without Temperature Interpolation) (p. 226).

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

225

Submodeling
Figure 8.5: Loads on the Submodel

Figure 8.6: Data Flow Diagram for Submodeling (Without Temperature Interpolation)

8.2.5.Verify the Distance Between the Cut Boundaries and the Stress Concentration
The final step is to verify that the cut boundaries of the submodel are far enough away from the concentration. You can do this by comparing results (stresses, magnetic flux density, etc.) along the cut
boundaries with those along the corresponding locations of the coarse model. If the results are in good
agreement, it indicates that proper cut boundaries have been chosen; otherwise, you will need to recreate
and reanalyze the submodel with different cut boundaries further away from the region of interest.
An effective way to compare results is to obtain contour displays and path plots, as shown in Figure 8.7: Contour Plots to Compare Results (p. 227) and Figure 8.8: Path Plots to Compare Results (p. 227).

226

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Submodeling Analysis Input

Figure 8.7: Contour Plots to Compare Results

Figure 8.8: Path Plots to Compare Results

8.3. Example Submodeling Analysis Input

The following example submodeling analysis input listings are available:
8.3.1. Submodeling Analysis Input: No Load-History Dependency
8.3.2. Submodeling Analysis Input: Load-History Dependency

8.3.1. Submodeling Analysis Input: No Load-History Dependency

The following example submodeling input does not include load-history dependency:
! Start with coarse model analysis:
/FILNAME,coarse
! Jobname = coarse
/PREP7
! Enter PREP7
...
...
! Generate coarse model
...
FINISH

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

227

Submodeling
/SOLU
ANTYPE,...
...
D,...
DSYMM,...
ACEL,...
...
SOLVE
SAVE
FINISH
! Create submodel:
/CLEAR
/FILNAME,submod
/PREP7
...
...
...

! Enter SOLUTION
! Analysis type and analysis options
! Loads and load step options

! Coarse model solution

! Results are on coarse.rst (or rmg, etc.)
! Coarse model database file coarse.db

! Clear the database (or exit program and re-enter)

! New jobname = submod
! Re-enter PREP7
! Generate submodel

! Perform cut-boundary
NSEL,...
!
NWRITE
!
ALLSEL
!
NWRITE,temps,node
!
!
SAVE
!
FINISH

interpolation:
Select nodes on cut boundaries
Write those nodes to submod.node
Restore full sets of all entities
Write all nodes to temps.node (for
temperature interpolation)
Submodel database file submod.db

RESUME,coarse,db
/POST1
FILE,coarse,rst
SET,...
CBDOF

Resume coarse model database (coarse.db)

Enter POST1
Use coarse model results file
Read in desired results data
Reads cut-boundary nodes from submod.node and
writes D commands to submod.cbdo
Reads all submodel nodes from temps.node and
writes BF commands to submod.bfin (for
temperature interpolation)
End of interpolation

BFINT,temps,node

FINISH

!
!
!
!
!
!
!
!
!
!

RESUME
/SOLU
ANTYPE,...
...
/INPUT,submod,cbdo
/INPUT,submod,bfin
DSYMM,...
ACEL,...
...
SOLVE
FINISH

! Resume submodel database (submod.db)

! Enter SOLUTION
! Analysis type and options

/POST1
...
...
...
FINISH

! Enter POST1

! Cut-boundary DOF specifications

! Interpolated temperature specifications
! Other loads and load step options

! Submodel solution

! Verify submodel results

8.3.2. Submodeling Analysis Input: Load-History Dependency

The following example submodeling input includes load-history dependency:
! Start with coarse model analysis:
/FILNAME,coarse
! Jobname = coarse
/PREP7
! Enter PREP7
...
...
! Generate coarse model
...
FINISH
/SOLU
! Enter SOLUTION
ANTYPE,...
! Analysis type and analysis options
...

228

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Submodeling Analysis Input

D,...
DSYMM,...
ACEL,...
...
SOLVE

! Loads and load step options

! Coarse model solution

! Results are on coarse.rst (or rmg, etc.)
! Coarse model database file coarse.db

SAVE
FINISH
! Create submodel:
/CLEAR
!
/FILNAME,submod
!
/PREP7
!
...
...
!
...
! Perform cut-boundary
NSEL,...
!
NWRITE
!
ALLSEL
!
NWRITE,temps,node
!
!
SAVE
!
FINISH
RESUME,coarse,db
!
/POST1
!
FILE,coarse,rst
!
SET,...
!
CBDOF
!
!
BFINT,temps,node
!
!
!
SET,...
!
CBDOF
!
!
BFINT,temps,node
!
!
!
...
FINISH
!
RESUME
!
/SOLU
!
ANTYPE,...
!
...
/INPUT,submod,cbdo
!
/INPUT,submod,bfin
!
DSYMM,...
!
ACEL,...
...
TIME
SOLVE
!
/INPUT,submod,cbdo
!
/INPUT,submod,bfin
!
TIME
SOLVE
..
FINISH
/POST1
!
...
...
!
...
FINISH

Clear the database (or exit program and re-enter)

New jobname = submod
Re-enter PREP7
Generate submodel
interpolation:
Select nodes on cut boundaries
Write those nodes to submod.node
Restore full sets of all entities
Write all nodes to temps.node (for
temperature interpolation)
Submodel database file submod.db
Resume coarse model database (coarse.db)
Enter POST1
Use coarse model results file
Read in desired results data
Reads cut-boundary nodes from submod.node and
writes D commands to submod.cbdo
Reads all submodel nodes from temps.node and
writes BF commands to submod.bfin (for
temperature interpolation)
Read in another desired results data
Reads cut-boundary nodes from submod.node and
writes D commands to submod.cbdo
Reads all submodel nodes from temps.node and
writes BF commands to submod.bfin (for
temperature interpolation)
This(SET,CBDOF,BFINT)can be repeated many time
End of interpolation
Resume submodel database (submod.db)
Enter SOLUTION
Analysis type and options
Cut-boundary DOF specifications
Interpolated temperature specifications
Other loads and load step options

Submodel solution of first steps

Cut-boundary DOF from another set
Interpolated temperature from another set
SOLVE as another load step
This (/INPUT..SOLVE) can be repeated many time
Enter POST1
Verify submodel results

Following is a simple model of a plate with a hole in its middle, with applied displacements in the x
direction. To simulate load-history dependency, cut boundary conditions are done at different substeps
of the coarse-mesh analysis, then applied as different load steps in the submodeling analysis. The material is assumed to be rate-dependent anisotropic plastic with isotropic hardening.
The problem is solved via the coarse-mesh model, the submodel, and the fine-mesh model, as shown
in this figure:
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

229

Submodeling
Figure 8.9: Coarse-Mesh Model, Submodel, and Fine-Mesh Model

Both the coarse-mesh model and the fine-mesh model are solved in 20 substeps. The fine-mesh model
yields a benchmark solution for comparison purposes.
The submodel analysis is performed in four solutions, as follows:
1. With cut-boundary conditions at only the last substep of the coarse-mesh analysis, solved in one
load step with 20 substeps.
2. With cut-boundary conditions at five substeps (that is, at every four substeps) of the coarse-mesh
analysis, solved in five load steps each with about four substeps.
3. With cut-boundary conditions at 10 substeps (that is, at every two substeps) of the coarse-mesh
analysis, solved in 10 load steps, each with about two substeps.
4. With cut-boundary conditions at all 20 substeps of the coarse-mesh analysis, solved in 20 load steps,
each with about two substeps.
The specified load steps and substeps make the size of the substep in submodel solutions close to
those in whole-model solutions.
The following figure shows the equivalent plastic strains at point P in each solution:

230

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Submodeling Analysis Input

Figure 8.10: Equivalent Plastic Strain Through Various Cut-Boundary Conditions and Load Steps

When boundary conditions are cut at more substeps, the results are closer to the fine mesh (the correct
solution).
The following figure shows the contour plots of equivalent plastic strain at the end of loading for each
of the four solutions:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

231

Submodeling
Figure 8.11: Equivalent Plastic Strain Distributions in a Submodeling Analysis with Load-History
Dependency

232

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Shell-to-Solid Submodels

The results with 20 boundary-condition cuttings most resemble the fine-mesh model.
As the example indicates, boundary conditions should be cut from more substeps of the coarse solution
to reflect load-history dependency more accurately.

8.4. Shell-to-Solid Submodels

In the shell-to-solid submodeling technique, the coarse model is a shell model, and the submodel is a
3-D solid model, as shown in this example:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

233

Submodeling
Figure 8.12: 3-D Solid Submodel Superimposed on Coarse Shell Model

The procedure for shell-to-solid submodeling is essentially the same as that for solid-to-solid submodeling, with these exceptions:
Shell-to-solid submodeling is activated by setting KSHS to 1 on the CBDOF command (Main Menu>
General Postproc> Submodeling> Interpolate DOF) and the BFINT command (Main Menu>
General Postproc> Submodeling> Interp Body Forc). This feature is not applicable to offsets used
with SHELL181 (SECOFFSET), or SHELL281 (SECOFFSET).
Cut boundaries on the submodel are the end planes that are normal to the shell plane (see Figure 8.13: Node Rotations (p. 235)). Nodes on these cut boundaries are written to the node file (NWRITE)
(Main Menu> Preprocessor> Modeling> Create> Nodes> Write Node File).
To determine the degree-of-freedom values at a cut-boundary node (CBDOF), the program first projects
the node onto the nearest element in the shell plane. The degree-of-freedom values of this projected
point are then calculated by interpolation and assigned to the corresponding node. Interpolated
temperatures (BFINT) are calculated based on the average temperature at the midplane of the nearest
shell element.

Note
The nodes on the cut boundary must lie within a distance of 0.75 times the average
thickness of the nearest shell element, as shown in Figure 8.13: Node Rotations (p. 235).
That is, the submodel should be approximately centered on the coarse model.

In a structural analysis, only translational displacements are calculated for the cut-boundary nodes,
but their values are based on both the translations and rotations of the projected point. Also, the
node is rotated such that the nodal UY direction is always perpendicular to the shell plane, as shown
234

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Where to Find Examples

in Figure 8.13: Node Rotations (p. 235). A UY constraint is calculated only for nodes that are within 10
percent of the average shell element thickness from the shell plane, preventing overconstraint of the
submodel in the transverse direction.
Figure 8.13: Node Rotations

Node rotations: (a) before CBDOF command, (b) after CBDOF command
The .CBDO file written by the CBDOF command consists of two blocks:
a block of NMODIF commands (indicating node rotation angles) and DDELE commands (to delete
UY constraints)
a block of D commands (to apply the interpolated degree-of-freedom values).
The two blocks are separated by a /EOF command and a :CBnn label (where nn is the cumulative iteration number of the results set used).
You must read in the .CBDO file in PREP7, because the NMODIF command is only valid in PREP7. To
do so, enter the preprocessor, then use one of these methods:
Command(s): /INPUT
GUI: Utility Menu> File> Read Input from
Because the two blocks of commands are separated by a /EOF command, it is necessary to read in the
.CBDO file twice. The second time you read in the file, use the LINE field on /INPUT ("Optional line
number or label" in the GUI) to instruct the program to read the file starting with the :CBnn label, as
shown below:
/PREP7! The .CBDO file must be read in PREP7
/INPUT,,cbdo
! Reads Jobname.cbdo up to the /EOF command
/INPUT,,cbdo,,:cb1 ! Reads same file from the label :cb1

8.5. Where to Find Examples

The Mechanical APDL Verification Manual consists of test case analyses demonstrating the analysis capabilities of the program. While these test cases demonstrate solutions to realistic analysis problems, the
Mechanical APDL Verification Manual does not present them as step-by-step examples with lengthy data
input instructions and printouts. However, most users who have at least limited finite element experience
should be able to fill in the missing details by reviewing each test case's finite element model and input
data with accompanying comments.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

235

Submodeling
The Mechanical APDL Verification Manual contains the following submodeling test case:
VM142 - Stress Concentration at a Hole in a Plate

236

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 9: Substructuring
Substructuring is a procedure that condenses a group of finite elements into one element represented
as a matrix. The single-matrix element is called a superelement. You can use a superelement in an analysis as you would any other element type. The only difference is that you first create the superelement
by performing a substructure generation analysis.
Substructuring is available in the ANSYS Multiphysics, the ANSYS Mechanical, and the ANSYS Structural
products.
The following substructuring topics are available:
9.1. Benefits of Substructuring
9.2. Using Substructuring
9.3. Sample Analysis Input
9.4.Top-Down Substructuring
9.5. Automatically Generating Superelements
9.6. Nested Superelements
9.7. Prestressed Substructures
9.8. Where to Find Examples

9.1. Benefits of Substructuring

Substructuring reduces computer time and allows solution of very large problems with limited computer
resources. Nonlinear analyses and analyses of structures containing repeated geometrical patterns are
typical candidates for substructuring. In a nonlinear analysis, you can substructure the linear portion of
the model so that the element matrices for that portion need not be recalculated at every equilibrium
iteration. In a structure with repeated patterns (such as the four legs of a table), you can generate one
superelement to represent the pattern and simply make copies of it at different locations, thereby saving
a significant amount of computer time.
You can also use substructuring on models with large rotations. For these models, the program assumes
each substructure to rotate about its mass center. In 3-D cases, the substructures contain three rigid
body rotations and three translational motions. With a large rotation model, you do not constrain the
substructure until the use pass because each substructure is treated as a single finite element that
should allow rigid body motions.
An example is an analysis that is too large for the computer in terms of model size or disk space requirements. In such a situation, you can analyze the model in pieces, where each piece is a superelement
small enough to fit on the computer.

9.2. Using Substructuring

A substructure analysis involves three distinct steps, called passes:
9.2.1. Step 1: Generation Pass (Creating the Superelement)
9.2.2. Step 2: Use Pass (Using the Superelement)
9.2.3. Step 3: Expansion Pass (Expanding Results Within the Superelement)

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

237

Substructuring
The following figure shows the data flow for a complete substructure analysis and some of the files
involved.
Figure 9.1: Applicable Solvers in a Typical Substructuring Analysis

!F
f x 


esrps Xsrevlos ellive

e xyseixiq
gi Dgj
Dgp Desrps Xsrevlos ellive
e i

!hF
f %%
ph #$#"

gp Desrps
XAdeteles otu@ srevlos ellive
dohtem noitutitsuskf
e xysxei

p "

hF
PPvF hvF
F wF
f

9.2.1. Step 1: Generation Pass (Creating the Superelement)

The generation pass is where you condense a group of "regular" finite elements into a single superelement.
The condensation is done by identifying a set of master degrees of freedom, used mainly to define the
interface between the superelement and other elements and to capture dynamic characteristics for
dynamic analyses. Figure 9.2: Example of a Substructuring Application (p. 238) shows a plate-like structure
that is to be analyzed with contact (interface) elements. Since the contact elements require an iterative
solution, substructuring the plate portion can result in a significant savings in computer time. The
master DOF required in this case are the degrees of freedom that connect the plate to the contact
elements.
Figure 9.2: Example of a Substructuring Application

Generating a superelement involves two primary tasks:

9.2.1.1. Building the Model

238

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Substructuring
9.2.1.2. Applying Loads and Creating the Superelement Matrices

9.2.1.1. Building the Model

In this step, you specify the jobname and analysis title and then use /PREP7 to define the element
types, element real constants, material properties, and the model geometry. These tasks are common
to most analyses and are described in the Basic Analysis Guide. For the generation pass, you should
keep in mind the following additional points:
Jobname-This takes on special significance in a substructure analysis. By using jobnames effectively, you
can eliminate much of the file handling inherent in a three-pass analysis.
Use one of these methods to specify the jobname:
Command(s): /FILNAME
GUI: Utility Menu> File> Change Jobname
For example,
/FILNAME,GEN

gives the jobname GEN to all files produced by the generation pass. The default jobname is file or
whatever name was specified while entering the program.
Element Types - Most element types can be used to generate a substructure. In general, the only restriction
is that elements within the superelement are assumed to be linear and cannot use Lagrange multipliers.
If you include bilinear elements, they will be treated as linear elements (in their initial state).

Caution
Coupled-field elements used in a direct method coupled-field analysis with load vector
coupling are not valid in a substructure analysis. Other elements in the same shape family
should be used instead. See the Coupled-Field Analysis Guide for details. In addition, elements
with Lagrange multipliers cannot be used in substructuring. These type of elements include
MPC184, CONTA171, CONTA172, CONTA173, CONTA174, CONTA175, CONTA176, CONTA177,
and CONTA178 with appropriate KEYOPT(2) setting, and elements PLANE182, PLANE183,
SOLID185, SOLID186, SOLID187, SOLID272, SOLID273, and SOLID285 when using KEYOPT(6)
> 0.
Material Properties - Define all necessary material properties. For example, if the mass matrix is to be
generated, density (DENS) (or mass in some form) must be defined; if the specific heat matrix is to be
generated, the specific heat (C) must be defined; and so on. Again, because a superelement is linear,
any nonlinear material properties will be ignored.
Model Generation - In the generation pass, you are mainly concerned with creating the superelement
portion of the model. The nonsuperelement portion, if any, is defined later in the use pass. However,
you should plan the modeling approach for both portions before you start building the model. In particular, decide on how you want to connect the superelement to the other elements. To ensure the
connection, use the same node numbers at the interface. (Other methods requiring less effort on your
part are discussed in the use pass section later in this chapter.)
Edge Outline - Adjust the edge outline used to plot the superelement in the use pass [/EDGE]. A smaller
angle will produce more edges.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

239

Substructuring
One approach might be to develop the entire model, save it on a named database file, and select only
the portion to be substructured for the generation pass. In the use pass then, you can RESUME (Utility
Menu> File> Resume from) from the named database file, unselect the portion that was substructured
and replace it with the superelement matrix.

9.2.1.2. Applying Loads and Creating the Superelement Matrices

The "solution" from a substructure generation pass consists of the superelement matrix (or matrices).
As with any other analysis, you define the analysis type and options, apply loads, specify load step options,
and initiate the solution. Details of how to do these tasks are explained below.
Enter SOLUTION using either of these methods
Command(s): /SOLU
GUI: Main Menu> Solution
Define the analysis type and analysis options

The applicable options are explained below.

Analysis Type - Choose a substructure generation using one of these methods:

Command(s): ANTYPE
GUI: Main Menu> Solution> Analysis Type> New Analysis
New analysis or restart - If you are starting a new analysis, choosing the analysis type (as described
above) is all you need to do. However, if the run is a restart, you must also indicate this by setting
STATUS = REST on the ANTYPE command (Main Menu> Solution> Analysis Type> Restart). A restart
is applicable if you need to generate additional load vectors. (The files Jobname.EMAT, Jobname.ESAV,
and Jobname.DB from the initial run must be available for the restart.)

Note
Restarting a substructuring analysis is valid only if the backsubstitution method is chosen.
You cannot restart a run if the full resolve option is selected using the SEOPT command.
Name of the superelement matrix file - Specify the name (Sename) to be assigned to the superelement
matrix file. The program will automatically assign the extension SUB, so the complete file name will be
Sename.SUB. The default is to use the jobname [/FILNAME]. To specify the name of the superelement
matrix file:
Command(s): SEOPT
GUI: Main Menu> Solution> Analysis Type> Analysis Options
Equation Solver - The SPARSE solver is the only solver available for the generation pass of the substructure
analysis. To specify an equation solver:
Command(s): EQSLV
GUI: Main Menu> Solution> Analysis Type> Analysis Options
Matrices to be generated - You can request generation of just the stiffness matrix (or conductivity matrix,
magnetic coefficient matrix, etc.); stiffness and mass matrices (or specific heat, etc.); or stiffness, mass,
and damping matrices. The mass matrix is required if the use pass is a structural dynamic analysis or if
you need to apply inertia loads in the use pass. For the thermal case, the specific heat matrix is required
only if the use pass is a transient thermal analysis. Similar considerations apply to other disciplines and
to the damping matrix. To make your request, use the SEOPT command as described above.

240

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Substructuring
Matrices to be printed - This option allows you to print out the superelement matrices. You can request
listing of both matrices and load vectors, or just the load vectors. The default is not to print any matrices.
To print out the matrices, use the SEOPT command:
Expansion Pass Method - Allows you to select the expansion pass method you plan to use during subsequent expansion passes with this superelement. The backsubstitution method (default) saves the
factorized matrix files needed to perform a backsubstitution of the master DOF solution during the expansion pass. The full resolve method does not save any factorized matrix files. The factorized matrix
files are named Sename.LNxx for the sparse solver.

Note
Factorized matrix files can become very large as the problem size increases, but are not
needed if the full resolve method is chosen during the expansion pass.
During the expansion pass, the full resolve method reforms the elements used to create the superelement,
reassembles the global stiffness matrix, and applies the master DOF solution as displacement boundary
conditions internally. These displacement boundary conditions are deleted upon finishing the expansion
pass solution.

Note
You cannot restart a substructure analysis with the full resolve expansion pass method
chosen.
Mass matrix formulation - Applicable only if you want the mass matrix to be generated. You can choose
between the default formulation (which depends on the element type) and a lumped mass approximation.
We recommend the default formulation for most applications. However, for dynamic analyses involving
"skinny" structures, such as slender beams or very thin shells, the lumped mass approximation has been
shown to give better results. To specify a lumped mass approximation, use one of these methods:
Command(s): LUMPM
GUI: Main Menu> Solution> Analysis Type> Analysis Options
Modes to be used - For superelements being used in a subsequent dynamic analysis [ANTYPE,MODAL,
HARMONIC, or TRANSIENT], you may include mode shapes as extra degrees of freedom to obtain better
accuracy [CMSOPT]. See the chapter on Component Mode Synthesis for more information.
Define master degrees of freedom using one of these methods
Command(s): M
GUI: Main Menu> Solution> Master DOFs> User Selected> Define
In a substructure, master DOFs serve three purposes:
They serve as the interface between the superelement and other elements. Be sure to define master DOFs
at all nodes that connect to nonsuperelements, as shown in Figure 9.2: Example of a Substructuring Application (p. 238). All degrees of freedom at these nodes should be defined as master DOFs (Lab = ALL on
the M command). Master DOFs must be defined even if you plan to have no elements in the model other
than a superelement.
If the superelement is to be used in a dynamic analysis, master DOFs characterize the dynamic behavior
of the structure if the Component Mode Synthesis method [CMSOPT] is not used. See Modal Analysis in
the Structural Analysis Guide for guidelines.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

241

Substructuring
If constraints [D] or force loads [F] are to be applied in the use pass, master DOFs must be defined at
those locations with the M command.
If this superelement is to be transformed [SETRAN] later in the use pass or used in a large deflection
analysis [NLGEOM,ON], then all nodes that have master DOFs must have all six DOFs (UX, UY, UZ, ROTX,
ROTY, ROTZ) defined and active.
For large deflections, master DOFs are typically defined at the joints of the flexible body and are at the
nodes connected to a joint element (MPC184), another rigid or flexible body node, or ground. At least
two master DOFs must be defined for each substructure, as the average rotation of the superelement
is computed from the average rotation of its master DOF. If only one node is a joint node, then another
must be chosen at the free end. See the Multibody Analysis Guide for more details.
Apply loads on the model You can apply all types of loads in a substructure generation pass (see
Table 9.1: Loads Applicable in a Substructure Analysis (p. 243)), but keep in mind the following points:
The program will generate a load vector that includes the effect of all applied loads. One load vector per
load step is written to the superelement matrix file. This load vector is the equivalent load for the combined
loads in the load step. A maximum of 31 load vectors are allowed.
Nonzero DOF constraints can be used in the generation pass and will become part of the load vector. (In
the expansion pass, if the load step being expanded contains nonzero DOF constraints, the database must
have matching DOF values. If it does not, the DOF constraints must be respecified [D] in the expansion
pass.)
Application of constraints [D] or force loads [F] can be postponed until the use pass, but master DOF must
be defined at those locations with the M command or corresponding GUI path.

Note
If a mass matrix is generated, apply the degree of freedom constraints in the use pass at
the master DOF (defined in the generation pass) to ensure that all mass is accounted for
in the substructure. For analyses with acceleration loadings, the load should be applied
in the generation pass and used in the use pass for greater accuracy, rather than apply
the acceleration load on the reduced mass matrix.

Similarly, application of linear and angular accelerations can be postponed until the use pass, but only if
a mass matrix is generated. This is recommended if you plan to rotate the superelement in the use pass,
because load vector directions are "frozen" and rotated with the superelement.
The Maxwell force flag (MXWF label on the SF family of commands) is normally used in a magnetic analysis to flag element surfaces on which the magnetic force distribution is to be calculated. This flag has
no effect (and therefore should not be used) for a superelement in a magnetic analysis.
If you intend to create an imaginary force vector, you should generate it as a real load vector, then use
it as an imaginary load vector in the use pass (SFE,,,,KVAL = 2) and expansion pass (SEEXP,,,ImagKy =
ON).

242

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Substructuring
For large rotation analyses - Do not apply constraints to the model in this pass. You will apply constraints
for large rotation analyses in the use pass.
Table 9.1: Loads Applicable in a Substructure Analysis
Load Name

Load Category

Commands[1]
Solid Model Loads

Finite Element Loads

Displacement Temperature Constraints

Mag. Potential ...

DK, DKLIST, DKDELE,

DL, DLLIST, DLDELE, DA,
DALIST, DADELE,
DTRAN

D, DSYM, DLIST, DDELE,

DSCALE, DCUM

Force Heat Flow Rate Mag.

Flux ...

Forces

FK, FKLIST, FKDELE,

FTRAN

F, FLIST, FDELE, FSCALE,

FCUM

Pressure Convection Maxwell Surface ...

Surface
Loads

SFL, SFLLIST, SFLDELE,

SFA, SFALIST, SFADELE,
SFGRAD, SFTRAN

SF, SFLIST, SFDELE, SFE,

SFELIST, SFEDELE, SFBEAM, SFGRAD, SFFUN,
SFSCALE, SFCUM

Temperature Heat Generation Rate Current Density

...

Body Loads

BFK, BFKLIST, BFKDELE,

BFL, BFLLIST, BFLDELE,
BFA, BFALIST, BFADELE,
BFV, BFVLIST, BFVDELE,
BFTRAN

BF, BFLIST, BFDELE,

BFE, BFELIST, BFEDELE,
BFSCALE, BFCUM

Gravity, Linear and Angular Inertia Loads

Acceleration

ACEL, DOMEGA

1. The menu path used to access each command in the GUI will vary depending on the engineering discipline
of the analysis (structural, magnetic, etc.). For a list of menu paths, see the description of individual
commands in the Command Reference.
Specify load step options
options (damping).

The only options valid for the substructure generation pass are dynamics

Damping (Dynamics Options) - Applicable only if the damping matrix is to be generated.

To specify damping in the form of alpha (mass) damping:
Command(s): ALPHAD
GUI: Main Menu> Solution> Load Step Opts> Time/Frequenc> Damping
To specify damping in the form of beta (stiffness) damping:
Command(s): BETAD
GUI: Main Menu> Solution> Load Step Opts> Time/Frequenc> Damping
To specify damping in the form of material-dependent alpha damping:
Command(s): MP,ALPD
GUI: Not accessible from the GUI.
To specify damping in the form of material-dependent beta damping:
Command(s): MP,BETD
GUI: Not accessible from the GUI.
Save a backup copy of the database on a named file Doing this is required because you need to
work with the same database in the expansion pass. To save a backup copy, use one of these methods:
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

243

Substructuring
Command(s): SAVE
GUI: Utility Menu> File> Save as Jobname.db
Start solution calculations using one of these methods:
Command(s): SOLVE
GUI: Main Menu> Solution> Solve> Current LS Output from the solution consists of the superelement
matrix file, Sename.SUB, where Sename is the name assigned as an analysis option [SEOPT] or the
jobname [/FILNAME]. The matrix file includes a load vector calculated based on the applied loads. (The
load vector will be zero if no loads are defined.)
Repeat for additional load steps (that is, to generate additional load vectors) The load vectors
are numbered sequentially (starting from 1) and appended to the same superelement matrix file. See
Loading in the Basic Analysis Guide for other methods for multiple load steps.
Leave SOLUTION using one of these methods
Command(s): FINISH
GUI: Main Menu> Finish

9.2.2. Step 2: Use Pass (Using the Superelement)

The use pass is where you use the superelement in an analysis by making it part of the model. The entire
model may be a superelement or, as in the plate example, the superelement may be connected to
other nonsuperelements (see Figure 9.2: Example of a Substructuring Application (p. 238)). The solution
from the use pass consists only of the reduced solution for the superelement (that is, the degree of
freedom solution only at the master DOF) and complete solution for nonsuperelements.
The use pass can involve any analysis type (except a FLOTRAN or explicit dynamics analysis). The only
difference is that one or more of the elements in the model is a superelement that has been previously
generated. The individual analysis guides contain detailed descriptions about performing various analyses.
In this section, we will concentrate on the steps you need to make the superelement a part of your
model.

9.2.2.1. Clear the Database and Specify a New Jobname

The use pass consists of a new model and new loads. Therefore, the first step is to clear the existing
database. This has the same effect as leaving and re-entering the program. To clear the database, use
one of these methods:
Command(s): /CLEAR
GUI: Utility Menu> File> Clear & Start New
By default, clearing the database causes the START.ANS file to be reread. (You can change this setting
if you so desire.)

Caution
If you are using the command input method to clear the database, additional commands
may not be stacked on the same line (using the $ delimiter) as the /CLEAR command.
Be sure to define a jobname that is different from the one used for the generation pass. This way, you
can ensure that no generation pass files will be overwritten. To define a jobname, use one of these
methods:
Command(s): /FILNAME
GUI: Utility Menu >File> Change Jobname
244

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Substructuring

9.2.2.2. Build the Model

This step is performed in PREP7 and consists of the following tasks:
1. Define MATRIX50 (the superelement) as one of the element types. Use one of these methods:
Command(s): ET
GUI: Main Menu> Preprocessor> Element Type> Add/Edit/Delete
2. Define other element types for any nonsuperelements. Nonlinear behavior may or may not be allowed,
depending on the type of analysis to be performed.
3. Define element real constants and material properties for the nonsuperelements. Nonlinear properties
may or may not be allowed, depending on the type of analysis to be performed.
4. Define the geometry of the nonsuperelements. Take special care in defining the interfaces where
the nonsuperelements connect to the superelements. The interface node locations must exactly match
the locations of the corresponding master nodes on the superelements (see Figure 9.3: Node Locations (p. 245)).
There are three ways to ensure connectivity at these interfaces:
Use the same node numbers as the ones in the generation pass.
Use the same node number increment (or offset) between interface nodes in the generation pass
and interface nodes in the use pass. (Use SETRAN, as described below in step 5b.)
Couple the two sets of nodes in all degrees of freedom using the CP family of commands [CP,
CPINTF, etc.]. This method is helpful if you cannot use one of the above two methods. For example,
to define a set of coupled degrees of freedom use one of the following:
Command(s): CP
GUI: Main Menu> Preprocessor> Coupling/Ceqn> Couple DOFs
If the superelement is not connected to any other elements, you do not need to define any nodes
in the use pass.
Figure 9.3: Node Locations

Interface nodes between superelement and nonsuperelement must exactly match the master
node locations.
5. Define the superelement by pointing to the proper element type reference number and reading in
the superelement matrix. To point to the element type:
Command(s): TYPE
GUI: Main Menu> Preprocessor> Modeling> Create> Elements> Elem Attributes
Now read in the superelement matrix using one of these methods (you may first need to use
other commands to modify the matrix, as explained below):
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

245

Substructuring
Command(s): SE
GUI: Main Menu> Preprocessor> Modeling> Create> Elements> Superelements> From .SUB
File
a. If there are no nonsuperelements in the model, or if there are nonsuperelements and the interface
nodes have the exact same node numbers as the master nodes on the superelement, then simply
read in the superelement using the SE command:
TYPE,...! Element type reference number
SE,GEN! Reads in superelement from file GEN.SUB

The Sename field on the SE command shown above identifies the name of the superelement
matrix file. The extension .SUB is assumed, so the complete file name is Sename.SUB
(GEN.SUB in the above example). The superelement is given the next available element
number.
b. If there are nonsuperelements in the model and the interface nodes have a constant node
number offset from the master nodes, you must first create a new superelement matrix with
new node numbers and then read in the new matrix.
To create a new superelement matrix, use one of these methods:
Command(s): SETRAN
GUI: Main Menu> Preprocessor> Modeling> Create> Elements> Superelements> By
CS Transfer
Main Menu> Preprocessor> Modeling> Create> Elements> Superelements> By Geom
Offset
To read in the new matrix, use one of these methods:
Command(s): SE
GUI: Main Menu> Preprocessor> Modeling> Create> Elements> Superelements> From
.SUB File
For example, given an existing superelement matrix file GEN.SUB and a node number offset
of 2000, the commands would be:
SETRAN,GEN,,2000,GEN2,SUB
TYPE,...
SE,GEN2

! Creates new superelement GEN2.SUB with

!
node offset = 2000
! Element type reference number
! Reads in new superelement from file GEN2.SUB

c. If there are nonsuperelements in the model and the interface nodes have no relationship with
the master nodes (as would be the case with automatically meshed models), first observe the
following caution.

Caution
It is quite likely that the node numbers of the master nodes from the generation
pass overlap with node numbers in the use pass model. In such cases, reading
in the superelement [SE] will cause existing use pass nodes to be overwritten by
the superelement's master nodes. To avoid overwriting existing nodes, use a
node number offset [SETRAN] before reading in the superelement. In any case,
save the database [SAVE] before issuing the SE command.
Thus you should first save the database [SAVE], use the SETRAN command to create a new
superelement matrix with a node number offset, and then use the SE command to read in
246

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Substructuring
the new matrix. The CPINTF command (Main Menu> Preprocessor> Coupling/Ceqn>
Coincident Nodes) can then be used to connect the pairs of nodes at the interface. For
example, given a superelement matrix file called GEN.SUB:
*GET,MAXNOD,NODE,,NUM,MAX
SETRAN,GEN,,MAXNOD,GEN2,SUB
SE,GEN2
NSEL,...
CPINTF,ALL

!
!
!
!
!
!
!

MAXNOD = maximum node number

Creates new superelement with
node offset = MAXNOD, name = GEN2.SUB
Reads in new superelement
Select all nodes at the interface
Couples each pair of interface nodes in
all DOF

NSEL,ALL

d. If the superelement is to be transformed - moved or copied to a different position, or symmetrically reflected - you must first use the SETRAN command or SESYMM command (Main Menu>
Preprocessor> Modeling> Create> Elements> Superelements> By Reflection), with the appropriate node number offsets, to create new superelement matrix files and then use SE to read
in the new matrices. Connecting the superelements to the nonsuperelements is done the same
way as above - by using common node numbers, a constant node number offset, or the CPINTF
command.

Note
If you use SETRAN to transfer the superelement to a different coordinate system,
the superelement's master nodes are rotated with it by default. This is typically useful
if the original superelement's master nodes are rotated, into a cylindrical system for
example. (In this case, the transfer does not effect the superelement stiffness matrix.)
If the original superelement has no rotated nodes, it is likely that the transferred
superelement will not need rotated nodes either. You can prevent node rotation in
such cases by setting the NOROT field on SETRAN to 1. (The superelement stiffness
matrix and load vector are modified by the program for this type of transfer.)

6. Verify the location of the superelement using graphics displays and listings. Superelements are
represented by an edge outline display, the data for which are written to the matrix file in the
generation pass. To produce a graphics display:
Command(s): EPLOT
GUI: Utility Menu> Plot> Elements
To produce a listing:
Command(s): SELIST
GUI: Utility Menu> List> Other> Superelem Data
7. Save the complete model database:
Command(s): SAVE
GUI: Utility Menu> File> Save as Jobname.db
Leave PREP7 using one of these methods:
Command(s): FINISH
GUI: Main Menu> Finish

9.2.2.3. Apply Loads and Obtain the Solution

This step is performed during the solution phase of the analysis. The procedure to obtain the use-pass
solution depends on the analysis type. As mentioned earlier, you can subject a superelement to any
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

247

Substructuring
type of analysis. You should, of course, have the appropriate matrices generated during the generation
pass. For example, if you intend to do a structural dynamic analysis, the mass matrix must be available.
The procedure is as follows:
1. Enter SOLUTION using one of these methods:
Command(s): /SOLU
GUI: Main Menu> Solution
2. Define the analysis type and analysis options.
For large rotation analyses - turn large deformation effects on [NLGEOM,ON], and define the
proper number of substeps for the nonlinear analysis.
3. Apply loads on the nonsuperelements. These may consist of DOF constraints and symmetry conditions
[D family of commands], force loads [F family], surface loads [SF family], body loads [BF family], and
inertia loads [ACEL, etc.]. Remember that inertia loads will affect the superelement only if its mass
matrix was generated in the generation pass.

Note
For large rotation analyses, be sure to apply the proper constraints in this step.

4. Apply superelement load vectors (if any) using one of these methods:
Command(s): SFE
GUI: Main Menu> Solution> Define Loads> Apply> Load Vector> For Superelement
One load vector per load step (created during the generation pass) is available on the superelement matrix file, and is identified by its reference number:
SFE,63,1,SELV,0,0.75

applies, on element number 63, load vector number 1, with the load applied as a real load and
with a scale factor of 0.75. Thus the ELEM field represents the element number of the superelement, LKEY represents the load vector number (default = 1), Lab is SELV, KVAL is for a real
or imaginary load vector, and VAL1 represents the scale factor (default = 0.0). (See the SFE
command description for more information.)

Note
The load vector orientation is fixed (frozen) to the superelement, so if the superelement is used in a rotated position, the load vector rotates with it. The same applies
to the degree of freedom directions (UX, UY, ROTY, etc.). They too are fixed to the
superelement and will rotate with the superelement if it is rotated (unless NOROT =
1 on the SETRAN command, in which case the nodal coordinate systems will not
be rotated).

5. Specify load step options that are appropriate for the analysis type. Use the EQSLV command to
select an appropriate equation solver based on the chosen analysis type and the physics of the
problem.
6. Initiate the solution:

248

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Substructuring
Command(s): SOLVE
GUI: Main Menu> Solution> Solve> Current LS
Results from the solution consist of the complete solution for nonsuperelements and the reduced
solution - DOF solution at masters - for the superelements. The complete solution for nonsuperelements is written to the results file (Jobname.RST, RTH, or RMG), which you can postprocess
using normal postprocessing procedures.
The reduced solution is written to the file Jobname.DSUB. You can review this file using one
of these methods:
Command(s): SEDLIST
GUI: Main Menu> General Postproc> List Results> Superelem DOF
Utility Menu> List> Results> Superelem DOF Solu
To expand the reduced solution to all elements within the superelement, you will need to
perform the expansion pass, explained next.
7. Leave SOLUTION:
Command(s): FINISH
GUI: Main Menu> Finish

9.2.3. Step 3: Expansion Pass (Expanding Results Within the Superelement)

The expansion pass is where you start with the reduced solution and calculate the results at all degrees
of freedom in the superelement. If multiple superelements are used in the use pass, a separate expansion
pass will be required for each superelement.
The procedure for the expansion pass assumes that the .EMAT, .ESAV, .SUB, .LN22, .DB, and .SELD
files from the generation pass and the .DSUB file from the use pass are available. For larger substructures,
the files .LN09 and .LN20 will also be required if they were created in the generation pass. The expansion pass logic automatically detects which, if any, factorized matrix files are available and chooses
the appropriate expansion pass method and solver accordingly. If an offset of node numbers was used
in the use pass [SETRAN or SESYMM], it will automatically be taken into account in the expansion pass.
The backsubstitution method uses the reduced solution from the use pass and substitutes it back into
the available factorized matrix file to calculate the complete solution. The full resolve solution reforms
the element stiffness matrices originally used to create the superelement. The global stiffness matrix
for these elements is then assembled. The reduced solution is applied to the model as displacement
boundary conditions, and the complete solution within the superelement is solved.

Note
The displacement boundary conditions are automatically applied internally at the master
degrees of freedom during the expansion pass solution and are automatically deleted when
the solution completes.
The expansion pass logic for substructuring analyses first searches for the superelement .LN22 file
and, if found, chooses the sparse solver to perform a backsubstitution (the EQSLV command is ignored).
Otherwise, the program will stop the expansion pass and give a message suggesting an alternate expansion method.
If the .LN22 file is not detected for the specified superelement, the full resolve method is chosen. The
PCG solver is chosen by default for the full resolve method. You can select the sparse solver using the

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

249

Substructuring
EQSLV command to override the default. Other equation solvers cannot be used with the full resolve
method.
1. Clear the database:
Command(s): /CLEAR
GUI: Utility Menu> File> Clear & Start New
This has the same effect as leaving and re-entering the program.
2. Change the jobname to what it was during the generation pass. This way, the program can easily
identify the files required for the expansion pass:
Command(s): /FILNAME
GUI: Utility Menu> File> Change Jobname
3. Restore the generation pass database:
Command(s): RESUME
GUI: Utility Menu> File> Resume Jobname.db
4. Enter SOLUTION using one of these methods:
Command(s): /SOLU
GUI: Main Menu> Solution
5. Activate the expansion pass and its options:
Command(s): EXPASS
GUI: Main Menu> Solution> Load Step Opts> ExpansionPass
Expansion pass on or off - Choose "on."
Name of superelement to be expanded - Specify the name (Sename):
Command(s): SEEXP
GUI: Main Menu> Solution> Load Step Opts> ExpansionPass> Expand Superelem
(The complete name of the file is assumed to be Sename.SUB.)
Name of the reduced solution file from use pass - Specify the name (Usefil) using the SEEXP command (or the menu path shown above). The complete name of the file is assumed to be Usefil.DSUB.
Real or imaginary component of displacement - Applicable only if the use pass was a harmonic analysis. Use the Imagky key on the SEEXP command (or the menu path shown above). If all solutions
are to be expanded (NUMEXP,ALL), Imagky is ignored and both the real and imaginary solutions
are expanded.
6. Identify the use pass solution(s) to be expanded. You can either expand a single solution [EXPSOL] or a
range of solutions (including all) [NUMEXP]:
Single Solution - use either the load step and substep numbers or the time (or frequency) to identify
a solution:
Command(s): EXPSOL
GUI: Main Menu> Solution> Load Step Opts> ExpansionPass> Single Expand> By Load Step

250

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Substructuring
Main Menu> Solution> Load Step Opts> ExpansionPass> Single Expand> By Time/Freq

Note
If the load step being expanded contains nonzero DOF constraints, the database must
have matching DOF values. If it does not, the DOF constraints must be respecified [D] in
the expansion pass.
Range of Solutions - Identify the number of solution and time or frequency range:
Command(s): NUMEXP
GUI: Main Menu> Solution> Load Step Opts> ExpansionPass> Range of Solu's
7. Specify load step options. The only options valid for a substructure expansion pass are output controls:
Output Controls - These options control printed output, database and results file output, and extrapolation of results.
If you want to include any results data on the printed output file (Jobname.OUT):
Command(s): OUTPR
GUI: Main Menu> Solution> Load Step Opts> Output Ctrls> Solu Printout
If you want to control the data on the results file (Jobname.RST):
Command(s): OUTRES
GUI: Main Menu> Solution> Load Step Opts> Output Ctrls> DB/Results File
If you want to review element integration point results by copying them to the nodes instead of
extrapolating them (default):
Command(s): ERESX
GUI: Main Menu> Solution> Load Step Opts> Output Ctrls> Integration Pt
8. Start expansion pass calculations:
Command(s): SOLVE
GUI: Main Menu> Solution> Solve> Current LS
9. Repeat steps 6 to 8 for additional use pass solutions to be expanded. If you need to expand the solution
for a different superelement, you will need to leave and re-enter SOLUTION.

Note
If the superelement to be expanded contains contact elements and has multiple use
passes, use separate expansions (repeating steps 1 through 7) or issue the NUMEXP
command to expand multiple load steps together.

10. Finally, leave SOLUTION:

Command(s): FINISH
GUI: Main Menu> Finish

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

251

Substructuring
11. Postprocess results in the superelement using standard techniques.

Note
An expansion pass is not valid if the use pass was a PSD analysis.

9.3. Sample Analysis Input

A sample command input listing for a substructuring analysis is shown below. This example assumes
a single superelement which is possibly combined with nonsuperelements.
!
GENERATION PASS
! Build the model (superelement portion)
/FILNAME,GEN
! Jobname = GEN (for example)
/TITLE,...
/PREP7
! Enter PREP7
----! Generate superelement portion of model
FINISH
! Apply loads and create the superelement matrices
/SOLU
! Enter SOLUTION
ANTYPE,SUBST
! Substructure analysis
SEOPT,GEN,...
! Superelement name and other substructure analysis options
M,...
! Master DOF
D,...
! Loads. A load vector will be generated and
--!
written to the superelement matrix file
--! Load step options
SAVE
! Save the database for later expansion pass
SOLVE
! Initiate solution -- creates GEN.SUB file
!
containing superelement matrix and load vector
--! Loads for second load vector (D and M may not changed)
SOLVE
! Add load vector 2
--! Repeat loading and SOLVE sequence for additional load vectors
--!
(Up to 31 total)
FINISH
!
USE PASS
! Build the model
/CLEAR
! Clear the database (or exit and re-enter program)
/FILNAME,USE
! Jobname = USE (for example)
/PREP7
! Enter PREP7
ET,1,MATRIX50
! MATRIX50 is the superelement type
ET,2,...
! Element type for nonsuperelements
--! Generate nonsuperelement model
--TYPE,1
! Point to superelement type reference number
SETRAN,...
! May be required for node number offset
SE,...
! Read in the superelement created by SETRAN
EPLOT
! Verify location of superelement
NSEL,...
! Select nodes at interface
CPINTF,ALL
! Couple node pairs at interface (required if
!
node numbers at interface don't match)
NSEL,ALL
FINISH
! Apply loads and obtain the solution
/SOLU
! Enter SOLUTION
ANTYPE,...
! Analysis type and analysis options
----D,...
! Loads on nonsuperelements
----SFE,...
! Apply superelement load vector
--! Load step options
--SAVE
! Save database before solution
SOLVE
! Initiate solution -- calculates complete solution
!
for nonsuperelements (USE.RST, RTH or RMG) and

252

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Top-Down Substructuring
!

reduced solution for superelements (USE.DSUB)

FINISH
! ... Review results in nonsuperelements
!
EXPANSION PASS
/CLEAR
! Clear the database
/FILNAME,GEN
! Change jobname back to generation pass jobname
RESUME
! Restore generation pass database
/SOLU
! Enter SOLUTION
EXPASS,ON
! Activate expansion pass
SEEXP,GEN,USE...
! Superelement name to be expanded (GEN, unless SETRAN used)
--! Load step options (mainly output controls)
--SOLVE
! Initiate expansion pass solution. Full
!
superelement solution written to GEN.RST (or
!
RTH or RMG).
FINISH
! ... Review results in superelements

For more information, see the ANTYPE, SEOPT, M, ET, SETRAN, SE, CPINTF, EXPASS, and SEEXP
command descriptions.

9.4. Top-Down Substructuring

The substructuring procedure described in the previous section is called bottom-up substructuring,
meaning that each superelement is separately generated in an individual generation pass, and all superelements are assembled together in the use pass. This method is suitable for very large models
which are divided into smaller superelements so that they can "fit" on the computer.
For substructuring of smaller models or of systems with global project geometry controls, and for isolated
component analysis, you can use a slightly different technique known as top-down substructuring. This
method is suitable, for instance, for substructuring of the linear portion of nonlinear models that are
small enough to fit on the computer. An advantage of this method is that the results for multiple superelements can be assembled in postprocessing. The procedure for top-down substructuring is briefly
explained below, and is followed by a sample input.
1. First build the entire model, including both the superelement and nonsuperelement portions. Save
this model on a named database file (for example, FULL.DB). The full model database is later required
for the expansion pass. It will also be required for the use pass if the model consists of nonsuperelements.
2. Perform the generation pass on a selected subset of the entire model. Because the full model has
already been built, all you need to do is to select the elements for the superelement portion, apply
the desired loads (for the load vector), and create the superelement with the SOLVE command
(Main Menu> Solution> Solve> Current LS).
The use of components may be helpful for this. To group items into a component, use the CM
command (Utility Menu> Select> Comp/Assembly> Create Component).
If multiple superelements are to be generated, you will need to exit and re-enter SOLUTION
each time and repeat the select-load-solve steps. Be sure to use a different jobname for each
superelement.
3. Perform the use pass. Enter PREP7 and start by restoring the full model database and then selecting
only the nonsuperelement portion of the model. Next, define the superelement type [ET, TYPE] and
read in the appropriate superelement matrices. In most cases, you don't need to worry about the

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

253

Substructuring
connecting nodes between the superelements, because they were all generated from a single
model.
Enter SOLUTION and define the analysis type and analysis options. Apply loads on the nonsuperelements, read in load vectors (if any), specify load step options, and initiate the use pass
solution.
4. Perform the expansion pass. Start by restoring the full model database, with all elements and nodes
active. Then expand each superelement separately, using the appropriate jobnames and exiting and
re-entering SOLUTION each time. You can then review the results in each superelement using normal
postprocessing procedures. Use of the full database, FULL.DB, allows the reading in of multiple
superelement results:
RESUME,FULL,DB
/POST1
FILE,GEN1
SET,...
FILE,GEN2
SET,...!Will not clear previous superelement results

A sample input for top-down substructuring follows. This example assumes a model with one superelement and other nonsuperelements.
!
Sample
!
!
BUILD THE FULL
!
/FILNAME,FULL
/TITLE,...
/PREP7
--------SAVE

input for top-down substructuring

MODEL
! Jobname = FULL (for example)
! Enter PREP7
! Generate entire model, including both the
!
superelement and nonsuperelement portions
! Save the full model database. It is required for
!
the (use pass and) expansion pass.

FINISH
!

GENERATION PASS

!
/FILNAME,GEN
/SOLU
ANTYPE,SUBSTR
SEOPT,GEN,...
ESEL,...
NSLE
M,...
D,...
------SOLVE
--SOLVE
----FINISH
!
USE PASS
!
/CLEAR
/FILNAME,USE
RESUME,FULL,DB
ESEL,...
NSLE

254

!
!
!
!
!
!
!
!
!
!

Jobname = GEN (for example)

Enter SOLUTION
Substructure analysis type
Analysis options
Select elements and
nodes in the superelement portion
Master DOF
Loads. A load vector will be generated and written to the
superelement matrix file
Load step options

!
!
!
!
!
!

Initiate solution -- creates superelement

matrix file GEN.SUB.
Loads for second load vector (D and M may not changed)
Add load vector 2
Repeat loading and SOLVE sequence for additional load vectors
(Up to 31 total)

!
!
!
!
!

Clear database for use pass

Jobname = USE (for example)
Restore full model database (for nonsuperelements)
Select elements and
nodes in the nonsuperelement portion

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Automatically Generating Superelements

/PREP7
ET,...,MATRIX50
TYPE,...
SE,GEN
EPLOT
FINISH
/SOLU
ANTYPE,...
--D,...
----SFE,...
------SOLVE

! Superelement type (type number not used by nonsuperelements)

! Point to superelement type reference number
! Read in superelement matrix (GEN.SUB created above)

! Analysis type and analysis options

! Loads on nonsuperelements

! Superelement load vector

! Load step options
! Initiates solution -- calculates complete
!
solution for nonsuperelements (USE.RST, etc.)
!
and reduced solution for superelement (USE.DSUB)

FINISH
!
EXPANSION PASS
!
/CLEAR
! Clear database for expansion pass
/FILNAME,GEN
! Change jobname back to generation pass jobname
RESUME,FULL,DB
! Restore full model database
/SOLU
ANTYPE,SUBSTR
EXPASS,ON
EXPSOL,...
SEEXP,GEN,USE,...
----SOLVE

! Enter SOLUTION
!
!
!
!

Activate expansion pass

Specifies the solution to be expanded
Superelement name to be expanded
Load step options (mainly output controls)

! Initiate expansion pass solution. Full

!
superelement solution written to GEN.RST (or
!
RTH or RMG).

FINISH
! ... Review results in superelement

Please see the ANTYPE, SEOPT, M, ET, SE, EXPASS, and SEEXP command descriptions for more information.

9.5. Automatically Generating Superelements

When creating multiple superelements, the two methods described in the previous sections (bottomup substructuring and top-down substructuring) both require repeating a set of /SOLU commands for
each superelement you want to create. These methods also require a master DOF to be defined for
each superelement. If any superelements connect to each other, then the master DOF must be chosen
carefully on the interface(s) between each connecting superelement.
When creating multiple superelements, use the following automatic superelement generation process
to quickly create superelements (.SUB files), as well as the master DOF necessary on the interfaces
between each superelement. This simplifies the creation of the superelements and it efficiently breaks
a larger model into smaller models, for example, to be used in a nonlinear analysis.
To automatically generate superelements:
1.

If using the bottom-up substructuring method first create the part of the model that will become superelements. If using the top-down substructuring method, first create whole model, then select the part
of the model that will become superelements.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

255

Substructuring
2.

Perform the generation pass using SEOPT and any other /SOLU commands to define any necessary
options for the substructuring analysis.

3.

Use SEGEN to define the options for the automatic superelement generation process. If stopStage
= PREVIEW is selected, then the model is only broken into domains (superelements). No reduced matrices
are created and the superelements (.SUB files) are not actually created. You can then graphically
(visually) preview each domain by using /PNUM,DOMAIN. By default, master DOFs are automatically
defined at each of the following locations: all DOFs on the interfaces between each superelement, all
DOFs associated with contact elements (TARGE169 to CONTA177), and at all DOFs associated with nodes
having a point load defined. The option to manually define the master DOF only makes sense AFTER a
'preview pass' has been made, as the exact number of superelements and the superelement boundaries
for each superelement cannot be known until the process is completed at least once.

Note
Due to the heuristics in the automatic domain decomposer, which is used to create the
domains that will become superelements, the number of defined superelements may
exceed the number of requested superelements.
After completing a preview pass, you can then add master DOFs or remove master DOFs that were
automatically defined during the preview pass. At least one master DOF must be defined for each
superelement. Then set stopStage = GEN, and if any master DOFs were added or removed, set
mDof = YES, and solve the model.
4.

Use SOLVE to either preview or generate the automatically created superelements. Note that multiple
load steps are not supported with automatic superelement generation.

9.6. Nested Superelements

A powerful substructuring feature is the ability to use nested superelements (one superelement containing
another). When you generate a superelement, one of the elements in the generation pass may be a
previously generated superelement.
For example, suppose that you have a superelement named PISTON. You can generate another superelement named CYLINDER which contains the superelement PISTON. Now, for a complete analysis of
the cylinder and its piston, you will need to perform one use pass and two expansion passes. The use
pass calculates the reduced solution for the master DOF in the superelement CYLINDER. The first expansion pass calculates the complete solution for CYLINDER and the reduced solution for PISTON. The
second expansion pass then gives you the complete solution for PISTON.
You can also generate transformed nested superelements (using SETRAN or SESYMM commands). It
is important in this case to perform the expansion pass in the exact reverse order used for the generation
pass.

9.7. Prestressed Substructures

In modeling a system's behavior properly, it may be important to consider its stress state. That stress
state will influence the values of the stiffness matrix terms. The stress state from a previous structural
solution may be included when the stiffness matrix is formed in a superelement generation pass. Stress
stiffening can provide strength in a structure which would normally not have any resistance to certain
loadings. For example, a stretched cable can resist a normal loading while a slack cable cannot. Stress

256

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Where to Find Examples

stiffening can also change the response frequencies of a system which impacts both modal and transient
dynamic problems.
Two different approaches can be used to generate prestressed substructures. These approaches are a
static analysis prestress and a substructuring analysis prestress.

9.7.1. Static Analysis Prestress

1. Build the model, define a static analysis (ANTYPE command or menu path Main Menu> Solution>
Analysis Type> New Analysis) and apply the stiffening load.
2. Specify that prestress effects be calculated (PSTRES command or menu path Main Menu> Solution>
Analysis Type> Analysis Options or Main Menu> Solution> Unabridged Menu> Analysis Type>
Analysis Options).
3. Specify that the File.EMAT be written (EMATWRITE command; not available via the GUI)
4. Initiate the static analysis (SOLVE command or menu path Main Menu> Solution> Solve> Current LS).
5. Perform the generation pass. Include the prestress effects from the static analysis by issuing the PSTRES
command. (It is important to have the prestress effects active during the static analysis and the generation
pass.)
6. Perform the use and expansion passes.
7. Review the results.

9.7.2. Substructuring Analysis Prestress

Note
This method does not require a static analysis on the full DOF model.
1. Build the model and perform the generation pass. Make sure to reserve space for the stress stiffening
matrix by setting SESST = 1 on the SEOPT command (Main Menu> Solution> Analysis Type> Analysis
Options).
2. Apply loads and perform the static use pass.
3. Perform the expansion pass using the PSTRES command (Main Menu> Solution> Analysis Type>
Analysis Options) to calculate prestress effects.
4. Repeat the generation pass for a new substructure generation, while continuing the prestress effects
using PSTRES.
5. Solve the second generation pass and perform the use pass.
6. Perform the expansion pass and review the results.

9.8. Where to Find Examples

The Mechanical APDL Verification Manual presents test-case analyses demonstrating the analysis capabilities of the program. While these test cases demonstrate solutions to realistic analysis problems, the
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

257

Substructuring
Mechanical APDL Verification Manual does not present them as step-by-step examples with lengthy
data-input instructions and printouts; however, most users with at least limited finite-element experience
should be able to fill in the missing details by reviewing each test case's finite element model and input
data with accompanying comments.
The Mechanical APDL Verification Manual contains the following substructuring cases:
VM125
VM141
VM147
VM153

258

Radiation Heat Transfer Between Concentric Cylinders

Diametrical Compression of a Disk
Gray-Body Radiation within a Frustum of a Cone
Nonaxisymmetric Vibration of a Stretched Circular Membrane (Using Cyclic Symmetry)

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 10: Component Mode Synthesis

Component mode synthesis (CMS) is a form of substructure coupling analysis frequently used in structural dynamics.
CMS allows you to derive the behavior of the entire assembly from its constituent components. First,
the dynamic behavior of each of the components is formulated. Then, by enforcing equilibrium and
compatibility along component interfaces, the program forms the dynamic characteristics of the full
system model.
CMS is available in the ANSYS Mechanical and ANSYS Structural products.
The following CMS topics are available:
10.1. Understanding Component Mode Synthesis
10.2. Using Component Mode Synthesis
10.3. Example Component Mode Synthesis Analysis

10.1. Understanding Component Mode Synthesis

Although breaking up a single large problem into several reduced-order problems via substructuring
saves time and processing resources, component mode synthesis (CMS) offers the following additional
advantages:
More accurate than a Guyan reduction for modal, harmonic and transient analyses. CMS includes
truncated sets of normal mode generalized coordinates defined for components of the structural
model.
The ability to include experimental results, as the substructure model need not be purely mathematical.
A typical use of CMS involves a modal analysis of a large, complicated structure (such as an aircraft or
nuclear reactor) where various teams each design an individual component of the structure. With CMS,
design changes to a single component affect only that component; therefore, additional computations
are necessary only for the modified substructure.
Finally, CMS supports these substructuring features:
Top-down substructuring
Nested superelements
Prestressed substructures (static analysis prestress approach only).

10.1.1. CMS Methods Supported

The following component mode synthesis methods are available:
Fixed-interface (CMSOPT,FIX)
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

259

Component Mode Synthesis

Free-interface (CMSOPT,FREE)
Residual-flexible free-interface (CMSOPT,RFFB)
For most analyses, the fixed-interface CMS method is preferable. The free-interface method and the
residual-flexible free-interface method are useful when your analysis requires more accurate eigenvalues
computed at the mid- to high-end of the spectrum. The following table describes the primary characteristics of each interface method:
CMS Methods Supported
Fixed (CMSOPT,FIX)

Residual-Flexible Free (CMSOPT,RFFB)

Free (CMSOPT,FREE)

Interface nodes are constrained

during the CMS superelement
generation pass.

Interface nodes remain free durInterface nodes remain free during the CMS superelement gener- ing the CMS superelement generation pass.
ation pass.

No requirement to specify rigid

body modes.

You must specify the number of

rigid body modes (CMSOPT).

If rigid body motion exists, you

must specify pseudo-constraints
(D).

Generally recommended when

accuracy on only the lower
modes of the assembled structure (use pass) is necessary.

Generally recommended when

accuracy on both lower and
higher modes of the assembled
structure (use pass) is required.

Generally recommended when

accuracy on both lower and
higher modes of the assembled
structure (use pass) is required.

For more information, see the discussion of component mode synthesis theory and methods in the
Mechanical APDL Theory Reference.

10.1.2. Solvers Used in Component Mode Synthesis

Following are the solvers and files used in a typical component mode synthesis analysis:
Figure 10.1: Applicable CMS Solvers and Files

260

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Component Mode Synthesis

10.2. Using Component Mode Synthesis

As in substructuring, a component mode synthesis (CMS) analysis involves three distinct steps, called
passes:
1. Generation pass ( EOPT, CMSOPT)
S
2. Use pass ( ETRAN, SESYMM, CPINTF)
S
E
3. Expansion pass ( XPASS, SEEXP, EXPSOL, NUMEXP)
The CMS generation pass condenses a group of finite elements into a single CMS superelement, which
includes a set of master degrees of freedom (DOFs) and truncated sets of normal mode generalized
coordinates. The master DOFs serve to define the interface between the superelements or other elements.
The following CMS usage topics are available:
10.2.1.The CMS Generation Pass: Creating the Superelement
10.2.2.The CMS Use and Expansion Passes
10.2.3. Superelement Expansion in Transformed Locations
10.2.4. Plotting or Printing Mode Shapes
CMS Wizard
A user-friendly wizard is available to help you better understand the CMS process as well as to guide
you through the generation, use, and expansion passes for the fixed-interface (CMSOPT,FIX) and freeinterface (CMSOPT,FREE) methods for modal analyses. The wizard also provides file organization and
management support as files are generated by a CMS analysis. Use the Solution (/SOLU) processor to
access the CMS Wizard.

10.2.1. The CMS Generation Pass: Creating the Superelement

The process for generating a CMS superelement consists of two primary tasks:
1. Building the model
This step is identical to building the model for a substructuring analysis. Define density (DENS)--or
mass in some form--because CMS must generate both stiffness and mass matrices.
2. Creating the superelement matrices
The "solution" from a CMS generation pass consists of the superelement matrices (generalized stiffness
and mass matrix). This flowchart illustrates the process necessary for creating the superelement
matrix file:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

261

Component Mode Synthesis

Figure 10.2: Process Flow for Creating a CMS Superelement Matrix

CMS Superelement
Matrix

Enter the solution processor

/SOLU

Define a substructure analysis

type (ANTYPE,SUBSTR)

Define master DOFs

M

Specify component mode

synthesis method (CMSOPT)

Save the database

SAVE

Name superelement matrix

file (SEOPT)

Obtain the CMS generation

pass solution (SOLVE)

Specify mass matrix formulation

LUMPM

Exit the solution processor

FINISH

Specifying the CMS method When specifying the CMS method, also specify the number of modes
and, optionally, the frequency range used to generate the superelement. The program supports the
fixed-interface (CMSOPT,FIX), free-interface (CMSOPT,FREE), and residual-flexible free-interface
(CMSOPT,RFFB) CMS methods. If using the free-interface method, also specify the rigid body modes
(CMSOPT,,,,FBDDEF). If using the residual-flexible free-interface method, specify pseudo-constraints
(D,,,SUPPORT).
Naming the superelement matrix file The program assigns the .sub extension to the superelement matrix file name that you specify (SEOPT,Sename); therefore, the complete file name is Sename.SUB. The default file name is the Jobname (/FILNAME).
Specifying the lumped mass matrix formulation Specify the lumped mass matrix formulation
(LUMPM) if necessary. For most applications, ANSYS, Inc. recommends the default formulation (depending upon the element type); however, for dynamic analyses involving "skinny" structures such
as slender beams or very thin shells, the lumped mass approximation typically yields better results.
Defining master DOFs In a substructure, master degrees of freedom (DOFs) serve as the interface
between the superelements or other elements. Define master DOFs (M) at all nodes that connect
to non-superelements (Lab1 = ALL), as shown in Example of a Substructuring Application. You must
define master DOFs even if you intend to have no elements in the model other than a superelement.
If this superelement is to be transformed (SETRAN) later in the use pass or used in a large deflection
analysis (NLGEOM,ON), all nodes that have master DOFs must have all six DOFs (UX, UY, UZ, ROTX,
ROTY, ROTZ) defined and active.
For large deflections, master DOFs are typically defined at the joints of the flexible body and are at
the nodes connected to a joint element (MPC184), another rigid or flexible body node, or ground.
At least two master DOFs must be defined for each substructure, as the average rotation of the su-

262

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Component Mode Synthesis

perelement is computed from the average rotation of its master DOF. If only one node is a joint
node, then another must be chosen at the free end. See the Multibody Analysis Guide for more details.
Specifying pseudo-constraints
Required for the residual-flexible free-interface method (CMSOPT,RFFB). For each superelement where rigid-body modes exist, specify pseudo-constraints. Apply
only the minimum number of displacement constraints (D,,,SUPPORT) necessary to prevent rigid
body motion: three constraints (or fewer, depending on the element type) for 2-D models and six
(or fewer) for 3-D models.
Saving a copy of the database Saving a copy of the database (SAVE) is necessary because you
must work with the same data in the expansion pass.
Obtaining the CMS generation pass solution Output from the solution (SOLVE) consists of the
superelement matrix file (Sename.SUB), where Sename is the file name you assigned (via the SEOPT
command).
After obtaining the CMS superelement matrices, proceed to the use pass and then the expansion pass,
as you would in a substructuring analysis.
For a detailed example of how to use CMS, see Example Component Mode Synthesis Analysis (p. 264).

10.2.2. The CMS Use and Expansion Passes

The CMS use pass and expansion pass are identical to those in a substructuring analysis. The CMS use
pass supports the following analysis types:
Modal (ANTYPE,MODAL)
Static (ANTYPE,STATIC)
Transient (ANTYPE,TRANS) -- full or mode superposition method
Harmonic (ANTYPE,HARMIC) -- full or mode superposition method
Spectrum (ANTYPE,SPECT)
The use pass also supports substructuring analysis prestress.
As in substructuring, the generation and expansion passes occur for each part (CMS superelement) of
the entire structure, and the use pass occurs only once because it uses all superelements together to
build the full model. The use pass extracts the eigenvalues of the full model (but not the eigenvectors,
because the expansion pass recovers them).
In a modal analysis using the free-interface CMS method, the use pass may not always extract all of the
modes requested via the MODOPT command. In such cases, increase or decrease the number of modes
to extract and run the use pass eigensolution again.

10.2.3. Superelement Expansion in Transformed Locations

When creating a new CMS superelement from an existing superelement (via the SESYMM or SETRAN
command) during the use pass, you can specify an offset value to the node or element IDs in the FE
geometry record saved in the .rst results file. The command to do so is RSTOFF.
With appropriate offsets, you can write additional results files with unique node or element IDs and
thus display the entire model, even if the original superelements have overlapping node or element
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

263

Component Mode Synthesis

sets. Results files containing the offset node or element IDs are incompatible with the .db database
files saved at the generation pass.
After performing the use pass and expansion pass for all CMS superelements, the mode shape display
of the entire assembled structure shows the offset superelements in their transformed locations.
For more information, see Example: Superelement Expansion in a Transformed Location (p. 279).

10.2.4. Plotting or Printing Mode Shapes

Plotting or printing the mode shapes of the assembled structure occurs during postprocessing. The
postprocessor (/POST1) uses the results files generated by the CMS superelements to display shape
results. Issue the CMSFILE command to import the CMS superelement results files into the postprocessor
where you can view the assembled structure. (You can issue the command as often as needed to include
all or some of the component results files.)
If you created new CMS superelements from existing superelements (via SESYMM or SETRAN commands)
during the use pass, you can issue the SEEXP command to expand the results with the offset superelements appearing in their transformed locations.
Set the desired mode shape of the assembled structure via the SET command, then plot (PLNSOL) or
print (PRNSOL) the mode shapes.

10.3. Example Component Mode Synthesis Analysis

This section introduces you to the program's component mode synthesis (CMS) analysis capabilities by
way of example. The example CMS presents a modal analysis of a 2-D tuning fork.

10.3.1. Problem Description

The model is an unconstrained stainless steel tuning fork. After dividing the fork into three CMS superelements, you must determine the vibration characteristics (natural frequencies and mode shapes) of
the entire model.

10.3.2. Problem Specifications

The geometric properties for this analysis follow.

264

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Component Mode Synthesis Analysis

m IFH
mm S
htdiw
m SPHFH
suidr
m SQHFH

The fork is divided into three CMS superelements:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

265

Component Mode Synthesis

Q tr

P tr
I tr

The three interfaces are as follows:

efens
efens

efens

266

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Component Mode Synthesis Analysis

The material properties for this analysis are as follows:
Young's modulus (E) = 190e9
Poisson's ratio () = 0.3
Density = 7.7e3
The first 10 eigenfrequencies are extracted, and the fourth mode shape (the first nonrigid body mode)
is expanded.

10.3.3. Input for the Analysis: Fixed-Interface Method

Use this input file (named cms_sample.inp) to perform the example CMS analysis via the fixed-interface method. The file contains the complete geometry, material properties, and components (nodes
and elements).
/batch,list
/title, 2D Tuning Fork
! Component Mode Synthesis - 2-D example
! The Structure is divided into 3 CMS Superelements
! STEP #1
! Start an interactive session
! STEP #2
! Read in this input file: cms_sample.inp
finish
/clear
/filnam,full
/units,si
blen=0.035
radi=0.025
tlen=0.1
tthk=0.005
/plopts,minm,0
/plopts,date,0
/pnum,real,1
/number,1
/prep7
k,1,-tthk/2
k,2,tthk/2
k,3,-tthk/2,blen
k,4,tthk/2,blen
local,11,1,,blen+tthk+radi
k,5,radi+tthk,-180
k,6,radi,-180
kgen,2,3,4,1,-tthk
k,9,radi
k,10,radi+tthk
a,5,6,7,3
a,3,7,8,4
a,4,8,9,10
csys,0
a,1,2,4,3
k,11,-radi-tthk,blen+tthk+radi+tlen
k,12,-radi,blen+tthk+radi+tlen
k,13,radi,blen+tthk+radi+tlen
k,14,radi+tthk,blen+tthk+radi+tlen
a,5,6,12,11
a,9,10,14,13
eshape,2
esize,tthk/3.5
et,1,plane182,,,3
r,1,tthk
amesh,all

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

267

Component Mode Synthesis

mp,ex,1,190e9
mp,dens,1,7.7e3
mp,nuxy,1,0.3
nsel,s,,,38
nsel,a,,,174,176
nsel,a,,,170
cm,interface1,node
nsel,s,,,175
nsel,a,,,168
nsel,a,,,180,182
nsel,a,,,38,176,138
cm,interface2,node
nsel,s,,,175
nsel,a,,,168
nsel,a,,,180,182
nsel,a,,,170,174,4
cm,interface3,node
esel,s,,,273,372
cm,part1,elem
esel,s,,,373,652
esel,a,,,1,129
esel,a,,,130
esel,a,,,133,134
esel,a,,,137,138
esel,a,,,141,142
cm,part2,elem
cmsel,s,part1
cmsel,a,part2
esel,inve
cm,part3,elem
allsel,all
save
finish
/eof
! STEP #3 (a. through j.)
! Generation pass
! Generation pass 1
! a.
! Change the active jobname which will become the superelement name
/filnam,part1
! b.
! Specify the analysis type as substructuring
/solu
antype,substr
! c.
! Specifies the name to be assigned to superelement matrix file
! Strongly suggested to be the same as the active jobname
seopt,part1,2
! d.
! Specifies CMS options
cmsopt,fix,10
! e.
! Selects element component named "part1"
cmsel,s,part1
! f.
! Selects node component named "interface1"
cmsel,s,interface1

268

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Component Mode Synthesis Analysis

! g.
! All the active DOFs (that is, on the nodes which belong to "interface1")
! are set as masters
m,all,all
! h.
! Selects all the nodes attached to the selected elements
! (that is, elements which belong to "part1")
nsle
! i.
! solve the first CMS generation pass
solve
finish
! j.
! Save the generation pass 1 database
save
! Repeat the generation pass for "part2"
! Generation pass 2
/filnam,part2
/solu
antype,substr
seopt,part2,2
cmsopt,fix,10
cmsel,s,part2
cmsel,s,interface2
m,all,all
nsle
solve
finish
save

! Repeat the generation pass for "part3"

! Generation pass 3
/filnam,part3
/solu
antype,substr
seopt,part3,2
cmsopt,fix,10
cmsel,s,part3
cmsel,s,interface3
m,all,all
nsle
solve
finish
save

! STEP #4 (a. through i.)

! Use pass
! a.
! Clears the database
/clear,nostart
! b.
! Change the active jobname which will become the use pass name
/filnam,use

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

269

Component Mode Synthesis

! c.
! A superelement element type is created
/prep7
et,1,matrix50
! d.
! Element type attribute pointer set to 1
type,1
! e.
! Brings in the three superelements created above
se,part1
se,part2
se,part3
finish
! f.
! A modal analysis is performed
/solu
antype,modal
! g.
! Specifies modal analysis options
modopt,lanb,10
! h.
! Expands 10 modes
mxpand,10
! i.
! Solve the modal analysis
solve
finish
! STEP #5 (a. through g.)
! Expansion pass
! Expansion pass 1
! a.
! Clears the database
/clear,nostart
! b.
! Changes the jobname to superelement 1 name
/filnam,part1
! c.
! resume the database
resume
! d.
! Specifies the expansion pass
/solu
expass,on
! e.
! Specifies superelement name and use pass name
seexp,part1,use

270

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Component Mode Synthesis Analysis

! f.
! Specifies the loadstep and substep to be expanded
expsol,1,4
! g.
! Solve the first expansion pass
solve
finish
! Repeat the expansion pass for "part2"
! Expansion pass 2
/clear,nostart
/filnam,part2
resume
/solu
expass,on
seexp,part2,use
expsol,1,4
solve
finish
! Repeat the expansion pass for "part3"
! Expansion pass 3
/clear,nostart
/filnam,part3
resume
/solu
expass,on
seexp,part3,use
expsol,1,4
solve
finish

! STEP #6 (a. through d.)

! Reads results for "load step 1 - substep 4"
! a.
! Clears the database
/clear,nostart
! b.
! Specifies the data file where results are to be found

/post1
cmsfile,add,part1,rst
cmsfile,add,part2,rst
cmsfile,add,part3,rst
! c.
! Reads the first data set
set,first
! d.
! Plots the displacement contour in the x direction

10.3.4. Analysis Steps: Fixed-Interface Method

The following table describes the input listing and the steps involved in the example fixed-interface
CMS analysis in more detail.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

271

Component Mode Synthesis

Step

Description

Command(s)

1.

Start an interactive session.

2.

Read the input file: cms_sample.inp

3.

Perform the generation pass.

/INPUT,CMS_SAMPLE.INP

a. Change the Jobname to PART1.

/FILNAME,PART1

b. Specify the analysis type as substructuring.

/SOLU
ANTYPE,SUBSTR

c. Assign a name to the superelement matrix file.

SEOPT,PART1,2

d. Specify CMS options.

CMSOPT,FIX,10

e. Select element component PART1.

CMSEL

f. Select node component INTERFACE1.

CMSEL,S,INTERFACE1

g. Set all active DOFs as masters.

M,ALL,ALL

h. Select all nodes attached to the selected elements.

NSLE

i. Solve the current analysis.

SOLVE

j. Save the database.

SAVE

---As coded in the input file, generation passes for the remaining
parts PART2 and PART3 occur here. Steps a through j are repeated for PART2 and again for PART3. (The Jobname and
superelement matrix file name change accordingly.) Also, for
passes 2 and 3, the node component is INTERFACE2 and INTERFACE3, respectively.
4.

Perform the use pass.

a. Clear the database.

/CLEAR,NOSTART

b. Change the Jobname to USE.

/FILNAME,USE

c. Define the element type.

/PREP7
ET,1,MATRIX50

d. Define the element type attribute pointer.

TYPE,1

e. Define the three superelements to use in the model

(PART1, PART2 and PART3).

SE,PART1
SE,PART2
SE,PART3
FINISH

f. Specify the analysis type as modal.

/SOLU
ANTYPE,MODAL

g. Specify modal analysis options.

h. Expand 10 modes.

272

MODOPT,LANB,10
MXPAND,10

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Component Mode Synthesis Analysis

Step

Description

Command(s)

i. Solve the current analysis.

SOLVE
FINISH

5.

Perform the expansion pass.

a. Clear the database.

/CLEAR,NOSTART

b. Change the Jobname to PART1.

/FILNAME,PART1

c. Resume the database.

RESUME

d. Perform the expansion.

/SOLU
EXPASS,ON

e. Name the superelement and use pass.

SEEXP,PART1,USE

f. Specify the loadstep and substep to expand.

EXPSOL,1,4

g. Solve the current analysis.

SOLVE

----

FINISH

As coded in the input file, expansion passes for the remaining

parts PART2 and PART3 occur here. Steps a through g are
repeated for PART2 and again for PART3. (The Jobname and
superlement name change accordingly.)
6.

Read the results.

a. Specify the superelement matrix file containing the
results.

/CLEAR,NOSTART
/POST1
CMSFILE,ADD,PART1,RST
CMSFILE,ADD,PART2,RST
CMSFILE,ADD,PART3,RST

b. Read the first data set.

SET,FIRST

c. Plot the displacement contour in the X direction.

PLNSOL,U,X

----

FINISH

This step completes the sample fixed-interface CMS analysis.

Your results should match those shown in Figure 10.3: Example CMS Analysis Results: Fixed-Interface Method (p. 274).
The results of your fixed-interface CMS analysis should match those shown here:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

273

Component Mode Synthesis

Figure 10.3: Example CMS Analysis Results: Fixed-Interface Method

10.3.5. Input for the Analysis: Free-Interface Method

This input file fragment shows how to set up the generation pass to perform the example CMS analysis
via the free-interface method. (All other input remains the same, as shown in Input for the Analysis:
Fixed-Interface Method (p. 267).)
.
.
.
! STEP #3 (a. through j.)
! Generation pass
! Generation pass 1
! a.
! Change the active jobname which will become the superelement name
/filnam,part1
! b.
! Specify the analysis type as substructuring
/solu
antype,substr
! c.
! Specifies the name to be assigned to superelement matrix file
! Strongly suggested to be the same as the active jobname
seopt,part1,2
! d.
! Specifies CMS options
cmsopt,FREE,10,,,FNUM,3
! If not otherwise specified, the CMSOPT command default behavior
! is to automatically determine rigid body modes in the calculation
! e.
! Selects element component named "part1"
cmsel,s,part1

274

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Component Mode Synthesis Analysis

! f.
! Selects node component named "interface1"
cmsel,s,interface1
! g.
! All the active DOFs (that is, on the nodes which belong to "interface1")
! are set as masters
m,all,all
! h.
! Selects all the nodes attached to the selected elements
! (that is, elements which belong to "part1")
nsle
! i.
! solve the first CMS generation pass
solve
finish
! j.
! Save the generation pass 1 database
save
! Repeat the generation pass for "part2"
! Generation pass 2
/filnam,part2
/solu
antype,substr
seopt,part2,2
cmsopt,free,10,,,FNUM,3
cmsel,s,part2
cmsel,s,interface2
m,all,all
nsle
solve
finish
save
! Repeat the generation pass for "part3"
! Generation pass 3
/filnam,part3
/solu
antype,substr
seopt,part3,2
cmsopt,free,10,,,FNUM,3
cmsel,s,part3
cmsel,s,interface3
m,all,all
nsle
solve
finish
save
.
.
.

10.3.6. Analysis Steps: Free-Interface Method

The following table describes the input code fragment for the generation pass used in a free-interface
CMS analysis. (All other analysis steps remain the same, as shown in Analysis Steps: Fixed-Interface
Method (p. 271).)
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

275

Component Mode Synthesis

Step

Description

Command(s)

1.

...

2.

...

3.

Perform the generation pass.

a. Change the Jobname to PART1.

/FILNAME,PART1

b. Specify the analysis type as substructuring.

/SOLU
ANTYPE,SUBSTR

c. Assign a name to the superelement matrix file.

SEOPT,PART1,2

d. Specify CMS options.

CMSOPT,FREE,10,,,FNUM,3

e. Select element component PART1.

CMSEL

f. Select node component INTERFACE1.

CMSEL,S,INTERFACE1

g. Set all active DOFs as masters.

M,ALL,ALL

h. Select all nodes attached to the selected elements.

NSLE

i. Solve the current analysis.

SOLVE

j. Save the database.

SAVE

---As coded in the input file, generation passes for the remaining
parts PART2 and PART3 occur here. Steps a through j are repeated for PART2 and again for PART3. (The Jobname and
superelement matrix file name change accordingly.) Also, for
passes 2 and 3, the node component is INTERFACE2 and INTERFACE3, respectively.
4.

...

5.

...

6.

...

10.3.7. Input for the Analysis: Residual-Flexible Free-Interface Method

This input file fragment shows how to set up the generation pass to perform the example CMS analysis
via the residual-flexible free-interface method. (All other input remains the same, as shown in Input for
the Analysis: Fixed-Interface Method (p. 267).)
.
.
.
! STEP #3 (a. through j.)
! Generation pass
! Generation pass 1
! a.
! Change the active jobname which will become the superelement name
/filnam,part1
! b.
! Specify the analysis type as substructuring

276

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Component Mode Synthesis Analysis

/solu
antype,substr
! c.
! Specifies the name to be assigned to superelement matrix file
! Strongly suggested to be the same as the active jobname
seopt,part1,2
! d.
! Specifies CMS options
cmsopt,rffb,10
! e.
! Selects element component named "part1"
cmsel,s,part1
! f.
! Selects node component named "interface1"
cmsel,s,interface1
! g.
! All the active DOFs (that is, on the nodes which belong to "interface1")
! are set as masters
m,all,all
! h.
! Selects all the nodes attached to the selected elements
! (that is, elements which belong to "part1")
nsle
! i.
! Specify only the minimum number of displacement constraints necessary
! to prevent rigid body motion: three constraints (or fewer, depending
! on the element type) for 2-D models and six (or fewer) for 3-D models.
d,430,all,support
d,440,ux,support
! j.
! solve the first CMS generation pass
solve
finish
! k.
! Save the generation pass 1 database
save
! Repeat the generation pass for "part2"
! Generation pass 2
/filnam,part2
/solu
antype,substr
seopt,part2,2
cmsopt,rffb,10
cmsel,s,part2
cmsel,s,interface2
m,all,all
nsle
d,705,all,support
d,715,ux,support
solve

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

277

Component Mode Synthesis

finish
save
! Repeat the generation pass for "part3"
! Generation pass 3
/filnam,part3
/solu
antype,substr
seopt,part3,2
cmsopt,rffb,10
cmsel,s,part3
cmsel,s,interface3
m,all,all
nsle
d,1050,all,support
d,1060,ux,support
solve
finish
save
.
.
.

10.3.8. Analysis Steps: Residual-Flexible Free-Interface Method

The following table describes the input code fragment for the generation pass used in a residual-flexible
free-interface CMS analysis. (All other analysis steps remain the same, as shown in Analysis Steps: FixedInterface Method (p. 271).)
Step

Description

Command(s)

1.

...

2.

...

3.

Perform the generation pass.

a. Change the Jobname to PART1.

/FILNAME,PART1

b. Specify the analysis type as substructuring.

/SOLU
ANTYPE,SUBSTR

c. Assign a name to the superelement matrix file.

SEOPT,PART1,2

d. Specify CMS options.

CMSOPT,RFFB,10

e. Select element component PART1.

CMSEL

f. Select node component INTERFACE1.

CMSEL,S,INTERFACE1

g. Set all active DOFs as masters.

M,ALL,ALL

h. Select all nodes attached to the selected elements.

NSLE

i. Specify pseudo-constraints.

D,,,SUPPORT

---Specify only the minimum number of displacement constraints

necessary to prevent rigid body motion: three constraints (or
fewer, depending on the element type) for 2-D models and
six (or fewer) for 3-D models.
j. Solve the current analysis.

278

SOLVE

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Component Mode Synthesis Analysis

Step

Description

Command(s)

k. Save the database.

SAVE

---As coded in the input file, generation passes for the remaining
parts PART2 and PART3 occur here. Steps a through j are repeated for PART2 and again for PART3. (The Jobname and
superelement matrix file name change accordingly.) Also, for
passes 2 and 3, the node component is INTERFACE2 and INTERFACE3, respectively.
4.

...

5.

...

6.

...

10.3.9. Example: Superelement Expansion in a Transformed Location

This input file fragment shows how to create a superelement from an existing superelement, apply
offsets to the node and element IDs, and then expand it in the transformed location. (All other input
remains the same, as shown in Input for the Analysis: Fixed-Interface Method (p. 267).)
.
.
.
! STEP #3
! Generation pass
! Generation pass 1
/filnam,part1
/solu
antype,substr
seopt,part1,2
cmsopt,fix,10
cmsel,s,part1
cmsel,s,interface1
m,all,all
nsle
solve
finish
save
! Generation pass 2
/filnam,part2
/solu
antype,substr
seopt,part2,2
cmsopt,fix,10
cmsel,s,part2
cmsel,s,interface2
m,all,all
nsle
solve
finish
save
! No generation pass is necessary for PART3. We will
! create the third component of the model from the existing
! superelement PART2 and name it PART2SYM
! STEP #4
! Use pass
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

279

Component Mode Synthesis

/clear,nostart
/filnam,use
/prep7
et,1,matrix50
type,1
! Define the three superelements to use in the model
se,part1
se,part2
*get,nmax,node,,num,max
sesymm,part2,x,nmax,part2sym,sub
se,part2sym
cpintf,all,0.001
finish
/solu
antype,modal
modopt,lanb,10
mxpand,10
solve
finish
! STEP #5
! Expansion pass
! Expansion pass 1
/clear,nostart
/filnam,part1
resume
/solu
expass,on
seexp,part1,use
expsol,1,4
solve
finish
! Expansion pass 2
/clear,nostart
/filnam,part2
resume
/solu
expass,on
seexp,part2,use
expsol,1,4
solve
finish
! Obtain the third part of the model from PART2
! Expand the solution in the transformed location
/assign,rst,part2sym,rst
/solu
expass,on
seexp,part2sym,use,,on
!
Offset node and element IDs in the new superelement
rstoff,node,10000
rstoff,elem,10000
expsol,1,4
solve
finish

! STEP #6
! Reads results for "load step 1 - substep 4"

280

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Component Mode Synthesis Analysis

! Specify the data file where results are to be found
/clear,nostart
/post1
cmsfile,add,part1,rst
cmsfile,add,part2,rst
cmsfile,add,part2sym,rst

10.3.9.1. Analysis Steps: Superelement Expansion in a Transformed Location

The following table describes the input code fragments used to create a superelement from an existing
superelement, apply offsets to the node and element IDs, and then expand it in the transformed location.
All other analysis steps remain the same, as shown in Analysis Steps: Fixed-Interface Method (p. 271).)
Step

Description

Command(s)

1.

...

2.

...

3.

Perform the generation pass.

As coded in the input file, generation passes for PART1 and PART2 occur here. (The Jobname and superelement matrix file names change accordingly.) Also, for passes 1 and 2,
the node component is INTERFACE1 and INTERFACE2, respectively.
A third generation pass to generate PART3 is unnecessary. Instead, the third component
of the model (named PART2SYM) will be created from the existing superelement PART2.

4.

Perform the use pass.

...
Define the three superelements to
use in the model (PART1, PART2 and
PART2SYM).

SE,PART1

----

*GET,NMAX,NODE,,NUM,MAX

The PART2SYM superelement is created

from the existing PART2 superelement.

SESYMM,PART2,X,NMAX,PART2SYM,SUB

SE,PART2

SE,PART2SYM
CPINTF,ALL,0.001
FINISH

...
5.

Perform the expansion pass.

As coded in the input file, expansion passes for parts PART1 and PART2 occur here. (The
Jobname and superelement name change accordingly.)
Obtain the third component of the
model (named PART2SYM) from
PART2.

/ASSIGN,RST,PART2SYM,RST
/SOLU
EXPASS,ON
SEEXP,PART2SYM,USE,,ON

Offset node and element IDs.

RSTOFF,NODE,10000

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

281

Component Mode Synthesis

Step

Description

Command(s)
RSTOFF,ELEM,10000

Expand and solve.

EXPSOL,1,4
SOLVE
FINISH

6.

Read the results.

Specify the superelement matrix file
containing the results.

/CLEAR,NOSTART
/POST1
CMSFILE,ADD,PART1,RST
CMSFILE,ADD,PART2,RST
CMSFILE,ADD,PART2SYM,RST

...
...

282

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 11: Rigid-Body Dynamics and the ANSYS-ADAMS Interface

The ADAMS software marketed by MSC Software is one of several special-purpose programs used to
simulate the dynamics of multibody systems.
One drawback of the ADAMS program is that all components are assumed to be rigid. In the ADAMS
program, tools to model component flexibility exist only for geometrically simple structures. To account
for the flexibility of a geometrically complex component, ADAMS relies on data transferred from finiteelement programs. The ANSYS-ADAMS Interface is a tool provided by ANSYS, Inc. to transfer data from
Mechanical APDL to the ADAMS program.
The following ANSYS-ADAMS interface topics are available:
11.1. Understanding the ANSYS-ADAMS Interface
11.2. Building the Model
11.3. Modeling Interface Points
11.4. Exporting to ADAMS
11.5. Running the ADAMS Simulation
11.6.Transferring Loads from ADAMS
11.7. Methodology Behind the ANSYS-ADAMS Interface
11.8. Example Rigid-Body Dynamic Analysis

11.1. Understanding the ANSYS-ADAMS Interface

Use the ANSYS-ADAMS Interface whenever you want to include flexibility of a body in an ADAMS simulation. Flexibility can be an important aspect in a multibody system, for example, to recognize resonances
or to accurately simulate forces and movements of the components. Often, the flexibility of a system
is not negligible. A typical example is the model of a piston moving in an engine. The movement of
the piston significantly depends on the flexibility of the crankshaft and/or the connecting rod. Because
the geometry of a connecting rod can be complex, the ANSYS-ADAMS Interface can be used to account
for the connecting rod flexibility.
To use the ANSYS-ADAMS Interface, you first model a flexible component using standard commands.
While building the model, you must give special attention to modeling interface points where joints
will be defined in ADAMS. The next step is to use the ANSYS-ADAMS Interface to write a modal neutral
file (Jobname.MNF) that contains the flexibility information for the component. This file is written in
the format required by ADAMS/Flex, an add-on module available for ADAMS. See Exporting to
ADAMS (p. 286) for details on how to use the ANSYS-ADAMS Interface to create the .MNF file. For a
complete description of the method used to create the modal neutral file and the information it contains,
see The Modal Neutral File (p. 293).
After performing the dynamic simulation in ADAMS, you can use the export capabilities of ADAMS to
create an input file containing accelerations and rotational velocities of the rigid part and forces acting
in the joints of the component. You can then import the to perform a stress analysis. See Transferring
Loads from ADAMS (p. 289) for details on how to import the loads and perform a subsequent static
structural analysis.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

283

Rigid-Body Dynamics and the ANSYS-ADAMS Interface

The process for transferring flexible components to ADAMS and forces back to Mechanical APDL consists
of these general steps:
1. Build the model.
2. Model interface points.
3. Export to ADAMS (and create the modal neutral file).
4. Run the ADAMS simulation using the modal neutral file.
5. Transfer resulting loads from ADAMS to Mechanical APDL and perform a static analysis.
For more information and an example analysis, see Methodology Behind the ANSYS-ADAMS Interface (p. 293) and Example Rigid-Body Dynamic Analysis (p. 294).

11.2. Building the Model

To use the ANSYS-ADAMS Interface, you must first create a complete finite element model in Mechanical APDL.
When building your model, consider that:
The interface is designed to support most element types that have displacement degrees of freedom.
Exceptions are axisymmetric elements (for example, PLANE25) and explicit dynamic elements (for example,
SOLID164).
Only linear behavior is allowed in the model. If you specify nonlinear elements, they are treated as linear.
For example, if you include nonlinear springs (like COMBIN39), their stiffnesses are calculated based on
their initial status and never change.
Material properties can be linear, isotropic or orthotropic, constant or temperature-dependent. You must
define both Young's modulus (EX, or stiffness in some form) and density (DENS, or mass in some form)
for the analysis. Nonlinear properties are ignored.
Damping is ignored when the interface computes the modal neutral file (Jobname.MNF). Damping of
the flexible component can be added later in the ADAMS program.
The ADAMS program requires a lumped mass approach (LUMPM,ON). This requirement results in the
following special considerations.
For most structures that have a reasonably fine mesh, this approximation is acceptable. If a model has
a coarse mesh, the inertia properties may have errors. To determine what the effect will be, start a
modal analysis with and without LUMPM,ON and compare the frequencies.
When using SHELL181, set KEYOPT(3) = 2 to activate a more realistic in-plane rotational stiffness.
SHELL181 KEYOPT(3) = 2 is also a good choice if the elements are warped.
When using two dimensional elements, the corresponding ADAMS model must lie in the X-Y-plane.
Remember that ADAMS models are always three dimensional. The 2-D flexible component transferred
will not have any component in the Z-direction.
Nodes of a plane element only have two degrees of freedom: translations in the X- and Y-direction.
Thus, no moment loads (forces, joints) can be applied in the ADAMS analysis. Likewise, nodes of a solid
element only have translational degrees of freedom.

284

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Modeling Interface Points

You cannot apply constraints (D command) to the model. Also, make sure that no master degrees of
freedom (M command) were defined in an earlier analysis.

11.3. Modeling Interface Points

When building a model that will be used in an ADAMS simulation, an important consideration is how
to represent interface points within the structure. An interface point is a node that will have an applied
joint or force in the ADAMS program. Keep in mind that, in ADAMS, the forces can only be applied to
interface points.
The number of interface points used will determine the number of constraint modes for the model.
Constraint modes are the static shapes assumed by the component when one degree of freedom of
an interface point is given a unit deflection while holding all other interface degrees of freedom fixed.
The number of constraint modes is equal to the number of degrees of freedom of all interface points.
(For 3-D models, the interface points have 6 DOF; therefore, each interface point has 6 constraint modes.)
You must pay special attention to modeling interface points for these reasons:
An interface point must have six degrees of freedom (except for 2-D elements).
Force (applied directly or via a joint) should be applied to the structure by distributing it over an area
rather than applying it at a single node.
If there is no node in the structure where you can apply the force or joint in ADAMS (for example, a pin
center), you need to create a geometric location for that point.
Use the following guidelines to determine the best way to model the interface points for your structure:
To ensure that all your loads will be projected on the deformation modes in the ADAMS simulation, you
must define all nodes where you are going to apply a joint or a force as interface points.
Interface points in Mechanical APDL must always have six degrees of freedom, except for 2-D elements.
If your model consists of solid elements, use constraint equations or a spider web of beam elements (as
shown in Figure 11.1: Connecting a Structure to an Interface Point (p. 286)) to ensure that the interface
node has 6 degrees of freedom.
A good practice for modeling interface points is to reinforce the area using beam elements or constraint
equations. Using one of these techniques will distribute the force over an area rather than applying it to
a single node, which would be unrealistic.
If you use a spider web of beam elements, use a high stiffness and a small mass for the beams. Otherwise,
you will alter the stiffness and mass of your model, which could result in eigenmodes and frequencies
that do not represent the original model.
If you use constraint equations, we recommend using contact elements and the internal multipoint
constraint (MPC) algorithm (see Surface-based Constraints in the Contact Technology Guide) to attach
the interface node. As an alternative to the MPC method, you may use constraint equation commands
such as CE and CERIG (for example, CERIG,MASTE,SLAVE,UXYZ, where MASTE is the interface node).
(Avoid the RBE3 command since problems can occur with the master degrees of freedom.) If you use
constraint equations, mesh the interface point with a MASS21 element (use KEYOPT(3) = 0) that has
small (negligible) inertias.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

285

Rigid-Body Dynamics and the ANSYS-ADAMS Interface

Do not define interface points that lie next to each other and are connected by constraint equations or
short beams. This type of connection would require too many eigenmodes and result in a model that is
not well conditioned.
Figure 11.1: Connecting a Structure to an Interface Point (p. 286) shows three different ways that you
may attempt to attach an interface point to a structure. The first two examples (a and b) demonstrate
valid methods of attachment. The third example (c) demonstrates a poor method of attachment that
should not be used.
Figure 11.1: Connecting a Structure to an Interface Point

Each method depicted in the figure is described below.

a) Constraint equations are connecting the interface point to the structure. This method is recommended
because:
Force is distributed over an area.
A MASS21 element is used to define the six degrees of freedom of the interface point.
Moment loads are transmitted.
b) A spider web of beams is connecting the interface point to the structure. This method is recommended
(and preferred) because:
Force is distributed over an area.
No MASS21 element is necessary (because the beams supply the six degrees of freedom).
Moment loads are transmitted.
c) One beam is used to connect the interface point to the structure. This is not recommended because:
The force is applied to the structure at a single node.
Solid elements do not have rotational degrees of freedom. Therefore, moments will not be properly
transmitted from the interface point to the structure (a spider web scheme should be used).

11.4. Exporting to ADAMS

After building the model in Mechanical APDL (including all interface points), the next step is to invoke
the ANSYS-ADAMS Interface to create the modal neutral file, Jobname.MNF. Creation of this file is
driven by a command macro called ADAMS.MAC.
To start the interface, select the following GUI path.
Main Menu> Solution> ADAMS Connection> Export to ADAMS

286

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Exporting to ADAMS
The Select Interface Points dialog box appears first. From this dialog box, you must select two or more
interface points.
Do not select too many interface points, as one point gives rise to six degrees of freedom in ADAMS.
Too many interface points may lead to huge files and models.
After you confirm your selection, the Export to ADAMS dialog box appears.
Figure 11.2: Export to ADAMS Dialog Box

Complete the following steps using this dialog box.

1. System of Model Units: The units used for the model is important to the ADAMS program, whereas
Mechanical APDL only requires that you use a consistent set of units. The units chosen will be written
to the .MNF file and can be recalled with the ADAMS/Flex module. If no units are specified, ADAMS assumes that the same units were used in Mechanical APDL as the ones chosen in the ADAMS model. See
the /UNITS command for details. If you specify user defined units, a Define User Units dialog box will
appear for you to input the conversion factors (for length, mass, force, and time) between SI units and
your chosen units. Below is an example of user defined units in which the component has been modeled
using millimeter, tonne (metric ton), newton, and second.
Length Factor
=

1 meter/millimeter

=
1000

Mass Factor =

1 kilogram/tonne

=
0.001

Force Factor
=

1 newton/newton

=1

Time Factor =

1 second/second

=1

2. Number of Modes to Extract: Input the number of normal modes to compute. Normal modes are the
eigenmodes of the component with all degrees of freedom of all interface points fixed. The number of
normal modes depends on the frequency range of the excitation you will apply in your ADAMS model.
You must select a sufficient number of modes to represent your structure in that frequency range. In
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

287

Rigid-Body Dynamics and the ANSYS-ADAMS Interface

ADAMS, if you have chosen too many normal modes, you are able to deactivate eigenmodes based on
the frequency or an energy criterion.
3. Element Results: Specify whether or not the program should write stress and/or strain results. This option
has no effect on the output for beam elements. If you want to output stress and strain for only a subset
of nodes, you should create a node component named "STRESS" before running the ADAMS command
macro.
4. Shell Element Result Output Control: Specify the shell element output location (top, middle, bottom).
This option has no effect on the output for solid elements and beam elements.
5. Filename: Specify a filename for the modal neutral file. The default name is Jobname.MNF. If a file with
the chosen name exists, it will be moved to a file named filename.MNFBAK.
6. Export to ADAMS: Select Solve and create export file to ADAMS to initiate the solution sequence.
Static and normal modes are computed and all information required by ADAMS is written to the .MNF
file specified above. Only the selected elements are considered. The current model is written to the
database file Jobname.DBMNF.

Note
Note that the algorithm used to compute the .MNF file adds constraints to the interface
points. If you create the .MNF file a second time using the same model in the same run, be
sure to delete all constraints on the interface points (or resume the database file Jobname.DBMNF) before you run it again.

11.4.1. Exporting to ADAMS via Batch Mode

If you prefer to work in batch mode, you may choose to run the ADAMS command macro by command
input. After building the model and defining interface points, use the following commands to compute
the .MNF file.
/UNITS,Label
NSEL,...
SAVE
ADAMS,NMODES,...

!
!
!
!
!

Specify the units chosen for modeling

Select two or more interface points
Save the model for a possible resume from
this point
Activate ADAMS.MAC to compute the .MNF file

See the ADAMS command description for more information. When you use command input to compute
the .MNF file, there is no option to change the file name. The default name of Jobname.MNF will be
used.

11.4.2. Verifying the Results

It is a good practice to verify the correctness of the results after the .MNF file is created. Below are
guidelines you can use to complete this task.
Check the number of orthonormalized eigenmodes in the Mechanical APDL output window. These
eigenmodes are the result of an orthonormalization of the normal modes and the constraint modes.
You should observe the following:
The number of modes equals the number of normal modes plus the number of constraint modes.

288

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Transferring Loads from ADAMS

The first six modes are rigid-body modes. These are marked with (probable rigid-body mode). If
there is a mode close to a rigid-body mode but not marked, you may deactivate it later in the
ADAMS program.
If a mode is marked with Infinity. Possible mass singularity. Ignored, check your model carefully.
There might be a problem with the Interface points.
The first few modes are equal to the free-free eigenmodes of the component. You might want to
verify this by doing a modal analysis: Set analysis option to ANTYPE,MODAL with MODOPT,LANB
(Block Lanczos); activate the lumped mass approach with LUMPM,ON.
Review the normal modes (load step 1) and the constraint modes (load step 2) in the General Postprocessor.
Verify the transfer by doing a modal analysis of the component in ADAMS with all interface degrees
of freedom fixed. Compare the results with the normal modes calculated in Mechanical APDL (load
step 1).

11.5. Running the ADAMS Simulation

After you have verified that the .MNF file contains accurate information, you are ready to run an ADAMS
simulation with a flexible component. Import the .MNF file into your ADAMS model and attach it to
the rigid bodies using joints. To keep any numerical imbalance between inertia and external loads small,
make sure you simulate your ADAMS model with high accuracy.
For general information about the ADAMS program and how to import flexible bodies, refer to the
ADAMS manuals (provided by Mechanical Dynamics, Inc.), especially the documentation provided for
the ADAMS/Flex product.

11.6. Transferring Loads from ADAMS

There are two ways to transfer loads and/or deformations from ADAMS to Mechanical APDL:
If the component is assumed to be rigid in ADAMS, you can transfer joint and external forces, accelerations,
and rotational velocity acting on the component as described in Transferring Loads on a Rigid Body (p. 289).
If the component is flexible, you can transfer the deformed shape of the component using the MSR toolkit
from ADAMS. This type of transfer is not supported by the ANSYS-ADAMS Interface. See Transferring the
Loads of a Flexible Body (p. 292) for more information on this transfer method.

11.6.1. Transferring Loads on a Rigid Body

If you model your component as a rigid body in ADAMS, you can use the Export FEA Loads feature in
ADAMS to export the loads to a file. This file can then be imported into Mechanical APDL for a subsequent
stress analysis.
If you model your component as a flexible body, ADAMS allows you to use the Export FEA Loads feature
to transfer the loads, but the loads will be incomplete. Therefore, this load transfer procedure should
generally not be used for flexible bodies. The transfer of loads may work, however, if the flexible bodies
experience only small dynamic effects. If this is not the case, you may want to change the component
temporarily to a rigid body (using the Modify utility in ADAMS) and run another simulation before you
transfer the loads.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

289

Rigid-Body Dynamics and the ANSYS-ADAMS Interface

11.6.1.1. Exporting Loads in ADAMS

After performing an ADAMS simulation, you can export loads on a specific component at specific times.
In ADAMS, select File> Export> FEA Loads to access the ADAMS Export FEA Loads dialog box.
Figure 11.3: ADAMS Export FEA Loads Dialog Box

Complete the following steps in this dialog box.

1. File Type: Select FEA Loads
2. File Format: Select ANSYS
3. File Name: Specify a name for the load file. The default extension is .LOD.
4. Specify whether you want to export loads on a rigid body or a flexible body.
Rigid body: You must define a marker on the body that has the same position and orientation relative
to the body as the global origin does in the Mechanical APDL model.
Flexible body: The marker is set automatically since this information is known from the .MNF file.
5. Click on Add Load Points to Nodes Table.
If you chose a rigid body, you can input the node IDs of the nodes where the loads have to be applied
in Mechanical APDL.
If you chose a flexible body, ADAMS automatically inputs the correct node IDs.
6. Output at times: Specify at what time steps you want to export the loads.
7. Finally, ADAMS will ask you about the units. The units for export must be the same as those chosen for
building the Mechanical APDL model. If they are not the same, change them temporarily to the Mechanical APDL units or scale the loads in the load file later.

290

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Transferring Loads from ADAMS

Every time step in ADAMS is treated as a load step in Mechanical APDL. In ADAMS versions up to 11.0.0,
ADAMS writes the LSWRITE command before the load commands. Therefore, if you are using ADAMS
version 11.0.0 or earlier, you must use a text editor to move the LSWRITE command to the end of each
time step in the .LOD file.
The following loads will be included in the load file:
Joint forces (F command)
External forces (F command)
Accelerations and rotational velocities (ACEL, OMEGA, DOMEGA commands)

11.6.1.2. Importing Loads

After exporting the load file from ADAMS, you can use the ANSYS-ADAMS Interface to import the load
file and initiate a static structural analysis. To access the Import from ADAMS dialog box, pick:
Main Menu> Solution> ADAMS Connection> Import fr ADAMS
Figure 11.4: Import from ADAMS Dialog Box

Complete the following steps in the dialog box:

1. Import file from ADAMS: Enter the name of the load file that was exported from ADAMS.
2. Import option: Theoretically, external forces and inertia forces are in equilibrium. Due to numerical errors
or due to mass discrepancies between ADAMS and Mechanical APDL, this is insufficient to prevent a rigidbody motion of the component. Hence, you must constrain the component against rigid-body motion
in order to do a static structural analysis. The ANSYS-ADAMS Interface offers two import options to
achieve this.
Import loads only. The program applies inertia loads and external forces to the structure according to
the load file. For this option, you must manually add constraints to the Mechanical APDL model that
are compatible with the constraints used in the ADAMS model (if possible), or use common engineering
sense to prevent rigid-body motion.
Add weak springs: The program adds weak springs (COMBIN14 elements) to the corners of the
bounding box of the component. (For more information, see the WSPRINGS command documentation).
The weak springs prevent rigid-body motion without influencing the stress results. (See Adding Weak
Springs (p. 294) for more information on how the program adds weak springs to the model.)
3. Import button: When you pick the Import button, one load step file is written per time step exported
from ADAMS; existing load step files are deleted. If you chose the Import loads only option, you will
have to start the static solution manually by issuing the SOLVE command for each load step. If you chose

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

291

Rigid-Body Dynamics and the ANSYS-ADAMS Interface

the Add weak springs option, inertia relief is activated (IRLF,1) to compute accurate acceleration loads,
and the static analysis is started automatically.

Note
If you use the import procedure a second time with the Add weak springs option, additional
weak springs will be added to the model. This will have only a small influence on the results.

11.6.1.3. Importing Loads via Commands

If you prefer to work in batch mode, you may choose to import the load file and initiate the solution
by command input. After exporting the load file from ADAMS, use the following commands to read
the load file and initiate the static solution.
/INPUT,...
WSPRINGS (or D)
*DO,par,ival,fval,inc
IRLF,1
LSREAD,par
SOLVE
*ENDDO

!
!
!
!
!
!
!
!

Read the load file from ADAMS

Apply weak springs (or use D commands to
apply rigid-body constraints)
Specify load steps to solve
Activate inertial relief to achieve higher accuracy
(this step is optional)
Read load step
Solve model

Every subsequent call of the WSPRINGS command will apply weak springs. Therefore, this command
may be omitted when importing new loads.

11.6.1.4. Reviewing the Results

When the structural analysis is complete, you review the results as you would for any linear structural
analysis.
When using the weak springs option with inertia relief check that:
The accelerations that Mechanical APDL calculated computed for inertia relief are small compared to the
applied acceleration loads from ADAMS (ACEL, OMEGA, DOMEGA). Issue the command IRLIST to view
the inertia relief accelerations (translational and rotational).
The forces in the springs are small compared to the external forces. The forces in the springs can be
viewed by listing the reaction forces. Use the PRRSOL command to list reaction forces.
The external forces have to be balanced by the applied inertia forces only. If one of the above is not
true, there is an imbalance in your model that must be removed. Check your Mechanical APDL and
ADAMS models, respectively.

11.6.2. Transferring the Loads of a Flexible Body

If you want to model a flexible component in the ADAMS program and perform a subsequent stress
analysis, you may want to use the Modal Stress Recovery (MSR) toolkit provided by Mechanical Dynamics,
Inc. Using the features of this toolkit, it is possible to transfer the loads of a flexible component from
ADAMS to Mechanical APDL for stress analysis. This toolkit provides several strategies for interfacing
with Mechanical APDL:
Export Time-Domain Displacements: If you have only a few time steps to analyze, this is a fast option. The
displacement of every node of the component is written directly into a Mechanical APDL input file. This

292

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Methodology Behind the ANSYS-ADAMS Interface

file can then be imported using the /INPUT command. A simple static analysis can be started in Mechanical APDL after the import of this file.
Export Mode Shapes: The toolkit writes a Mechanical APDL input file that can be used to compute the
orthonormalized or unorthonormalized eigenmodes of the component. By using the Export of Modal
Coordinates option, these eigenmodes can be scaled in Mechanical APDL, and the stresses in the component
can be computed for every time step.
Export Nodal Loads: Using this feature, you can write a Mechanical APDL input file to perform stress recovery as a superposition of unit force load steps. This method ignores the inertia load contribution to
the flexible body deformation, so it may be inaccurate when interpreting dynamic effects.

Note
The MSR toolkit features described here are not supported by the ANSYS-ADAMS Interface.

11.7. Methodology Behind the ANSYS-ADAMS Interface

Some tasks performed by the ANSYS-ADAMS Interface involve substantial behind-the-scenes work.
Two tasks in particular fall in this category: the creation of the modal neutral file (Jobname.MNF) and
the addition of weak springs via the WSPRINGS command.
The following sections provide details concerning how Mechanical APDL performs those tasks.
11.7.1.The Modal Neutral File
11.7.2. Adding Weak Springs

11.7.1. The Modal Neutral File

The algorithm used to create the modal neutral file (.MNF) is based on a formulation called component
mode synthesis (also known as dynamic substructuring). ADAMS uses the approach of Craig Bampton
with some slight modifications. According to this theory, the motion of a flexible component with interface points is spanned by the interface constraint modes and the interface normal modes. Constraint
modes and interface normal modes together are referred to as component modes.
Because the algorithm relies on the component mode synthesis method, which is based on the modal
analysis, only linear properties are considered during the formation of the modal neutral file. All geometric and physical nonlinearities are ignored. If significant geometric nonlinear effects are present in
your component, you must subdivide the component into several smaller components and transfer
each one separately. You can then assemble the subdivided components in ADAMS to form a flexible
component with geometric nonlinearity.
The modal neutral file contains the following information:
Header information: date, ANSYS version, title, .MNF version, units
Body properties: mass, moments of inertia, center of mass
Reduced stiffness and mass matrices in terms of the interface points
Interface normal modes (the user requests the number of modes generated)
Interface constraint modes

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

293

Rigid-Body Dynamics and the ANSYS-ADAMS Interface

To supply the above information, ANSYS does a sequence of analyses through a macro file called
ADAMS.MAC (see the ADAMS command) in order to generate the required interface constraint modes
and interface normal modes.

11.7.2. Adding Weak Springs

During the import of loads from ADAMS to Mechanical APDL, you can instruct Mechanical APDL to add
weak springs to the model via the WSPRINGS command. The weak springs are added to the corners
of the bounding box of the component. The stiffnesses of the springs are many orders of magnitude
less than the stiffness of the structure, and hence prevent rigid-body motion without influencing the
stress results. The program takes the following steps when adding weak springs:
To define the bounding box, the algorithm finds the nodes with the maximum and minimum coordinates.
Six nodes are created by this approach. These nodes define the bounding box of the component. Because
a three dimensional model is required for this approach, simple beam models that only have an extension
in one dimension cannot be handled by the weak springs options.
COMBIN14 elements are used to link the six nodes of the bounding box to the ground in all three translational directions. The stiffness of the spring element is computed as k = (Emean)(106), where Emean is the
mean value of all moduli of elasticity defined. This is a very rough approach, but one which has proven
to be effective in practical applications. If the stress results are influenced by the springs, you can change
the stiffness by changing the corresponding COMBIN14 real constant.

11.8. Example Rigid-Body Dynamic Analysis

This example analysis demonstrates how to model a flexible component in Mechanical APDL and then
export the flexible body information to a file for use in ADAMS. The example also provides brief instructions on how to perform the rigid-body dynamic analysis in ADAMS, and details on how to transfer the
loads from ADAMS to Mechanical APDL in order to perform a stress analysis.

11.8.1. Problem Description

In the linkage assembly shown below, Link3 is a flexible component. Link3 is modeled as a rectangular
rod in Mechanical APDL using SOLID185 elements. The joints in ADAMS will be attached to interface
points (nodes) at the middle of the holes at either end of Link3. These middle points are connected to
the cylindrical joint surfaces by a spider web of BEAM188 elements.

294

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Rigid-Body Dynamic Analysis

Figure 11.5: Linkage Assembly

Link 3
U1
Input motion

Crank

U4
Link 2

U3

Link 1: Output
motion

U2

11.8.2. Problem Specifications

The figure below shows the Link3 component as it is modeled in Mechanical APDL.
Figure 11.6: Link3 Component

The following are dimensions and properties for the Link3 component.
Radius of holes (radh) = 6mm
Width of rectangular rod (width) = 25mm
Thickness of rectangular rod (thick) = 10mm
Length of rectangular rod (length) = 300mm + 4*Radius of holes = 324mm
Young's modulus for rod = 7.22 x 104 MPa
Poisson's ratio for rod = 0.34
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

295

Rigid-Body Dynamics and the ANSYS-ADAMS Interface

Density of rod = 2.4 x 10-9 tons/mm3
Young's modulus for beams = 2.1 x 105 MPa
Poisson's ratio for beams = 0.3
Density of beams = 0.1 x 10-9 tons/mm3

11.8.3. Command Input

/BATCH,list
/FILNAME,adamsout ! Define jobname
/TITLE,Export flexible component to ADAMS
!
/PREP7
! Enter preprocessor
!
! Define Parameters of rectangular rod
radh=6
! Radius of the holes in the rod
thick=10
! Rod thickness
width=25
! Rod width
length=300+4*radh ! Rod length
! Build geometry
RECTNG,0,length,0,width
CYL4,2*radh,width/2,radh
CYL4,length-2*radh,width/2,radh
ASBA,1,2
ASBA,4,3
VEXT,1, , ,0,0,thick
!
ET,1,SOLID185,,3
! Define SOLID185 as element type 1
ET,2,BEAM188,,,3
! Define BEAM188 as element type 2
!
MP,EX,1,7.22e4
! Material of the rectangular rod
MP,PRXY,1,0.34
MP,DENS,1,2.4e-9
!
MP,EX,2,2.1e5
! Material of the beams used for the spider web
MP,PRXY,2,0.3
MP,DENS,2,0.1e-9
!
SECTYPE,1,BEAM,ASEC
SECDATA,78.528,490.67,,490.67,,10,,,0.85716,0.85716
!
TYPE,1
! Set element type attribute pointer to 1
MAT,1
! Set material attribute pointer to 1
ESIZE,thick/3,0,
! Define global element size
VSWEEP,1
! Mesh rod
!
! Define interface points: numbers must be higher than highest
! node number already defined
N,100000,2*radh,width/2,thick/2
! Define interface point 1
N,100001,length-2*radh,width/2,thick/2
! Define interface point 2
!
NWPAVE,100000
! Set working plane to interface point 1
WPSTYL,,,,,,1
! Set working plane type to cylindrical
CSYS,4
! Activate working plane
NSEL,S,LOC,X,radh ! Select nodes on cylindrical hole
NSEL,A,,,100000
! Also select interface node
!
! Generate spider web of beams
*GET,nmin,node,,num,min
*GET,nnum,node,,count
*SET,jj,0
TYPE,2
MAT,2
REAL,1
*DO,jj,1,nnum-2
E,100000,nmin
NSEL,u,,,nmin
*GET,nmin,node,,num,min
*ENDDO
!

296

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Rigid-Body Dynamic Analysis

ALLS
!
NWPAVE,100001
! Set working plane to interface point 2
WPSTYL,,,,,,1
! Set working plane type to cylindrical
CSYS,4
! Activate working plane
NSEL,S,LOC,X,radh ! Select nodes on cylindrical hole
NSEL,A,,,100001
! Also select interface node
!
! Generate spider web of beams
*GET,nmin,node,,num,min
*GET,nnum,node,,count
*SET,jj,0
TYPE,2
MAT,2
REAL,1
*DO,jj,1,nnum-2
E,100001,nmin
NSEL,u,,,nmin
*GET,nmin,node,,num,min
*ENDDO
!
ALLS
!
/UNITS,MPA
! Define units used: millimeter
! megagram, second, newton
SAVE
! Save database
NSEL,s,,,100000,100001
! Select interface points
ADAMS,20,1
! Start ADAMS macro,
! adamsout.mnf is written
FINISH
/EXIT,nosave

At this point you may import the adamsout.mnf file into your ADAMS model and perform a rigidbody dynamics simulation. The ADAMS model should consist of the components shown in Figure 11.5: Linkage Assembly (p. 295). After the simulation is done, export the loads acting on the Link3
component at five arbitrary `time steps. Name the load file loads.lod.
Once you have exported the load file, you can perform a stress analysis for Link3 in Mechanical APDL
using the command input shown below.
RESUME,adamsout,db
! Resume model
/FILNAM,adamsin
! Change jobname
/TITLE,Import loads from ADAMS
! Change title
!
WSPRINGS
! Create weak springs
!
! Enter Solution and solve all load steps
/SOLU
/INPUT,loads,lod
! Read in 5 load steps written by ADAMS
*DO,i,1,5
! Use a do loop to solve each load step
LSREAD,i
! Read in load step
IRLF,1
! Activate inertia relief
SOLVE
! Solve current load step
*ENDDO
!
/POST1
! Enter the general postprocesser
! Write deformation and equivalent stress to graphics file
/VIEW,1,1,1,1
/AUTO,1
EPLOT
/TYPE,1,4
/SHOW,
EPLOT
*DO,i,1,5
SET,i
PLNSOL,u,sum
PLNSOL,s,eqv
*ENDDO
/SHOW,term

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

297

Rigid-Body Dynamics and the ANSYS-ADAMS Interface

FINISH
/EXIT,nosave

298

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 12: Element Birth and Death

If material is added to or removed from a system, certain elements in your model may become come
into existence or cease to exist. In such cases, you can use element birth and death options to activate
or deactivate selected elements, respectively.
The element birth and death feature is useful for analyzing excavation (as in mining and tunneling),
staged construction (as in shored bridge erection), sequential assembly (as in fabrication of layered
computer chips), and many other applications in which you can easily identify activated or deactivated
elements by their known locations.
The birth and death feature is available in the ANSYS Multiphysics, ANSYS Mechanical, and ANSYS
Structural products.
The following additional element birth and death topics are available:
12.1. Elements Supporting Birth and Death
12.2. Understanding Element Birth and Death
12.3. Element Birth and Death Usage Hints
12.4. Using Birth and Death
12.5. Where to Find Examples

12.1. Elements Supporting Birth and Death

If an element supports birth and death, it is indicated in the Special Features section of the documentation for that element.
User-defined elements can also be given the birth and death capability.
In some circumstances, an element's birth and death status depend upon a program-calculated quantity,
such as temperature, stress, and strain. You can issue commands such as ETABLE and ESEL to determine
the value of such quantities in selected elements, and to change the status (melted, solidified, ruptured,
etc.) of those elements accordingly. This capability is useful for modeling effects due to phase changes
(as in welding processes, when structurally inactive molten material solidifies and becomes structurally
active), failure-surface propagation, and other analysis-dependent element changes.

12.2. Understanding Element Birth and Death

To achieve the "element death" effect, the program does not actually remove "killed" elements. Instead,
it deactivates them by multiplying their stiffness (or conductivity, or other analogous quantity) by a
severe reduction factor (ESTIF). This factor is set to 1.0E-6 by default, but can be given other values.
(For more information, see Apply Loads and Obtain the Solution (p. 301).)
Element loads associated with deactivated elements are zeroed out of the load vector, however, they
still appear in element-load lists. Similarly, mass, damping, specific heat, and other such effects are set
to zero for deactivated elements. The mass and energy of deactivated elements are not included in the
summations over the model. An element's strain is also set to zero as soon as that element is killed.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

299

Element Birth and Death

In like manner, when elements are "born," they are not actually added to the model; they are simply
reactivated. You must create all elements, including those to be born in later stages of your analysis,
while in PREP7. You cannot create new elements in SOLUTION. To "add" an element, you first deactivate
it, then reactivate it at the proper load step.
When an element is reactivated, its stiffness, mass, element loads, etc. return to their full original values.
Elements are reactivated with no record of strain history (or heat storage, etc.); that is, a reactivated
element is generally strain-free. Initial strain defined as a real constant, however, is not be affected by
birth and death operations.
Unless large-deformation effects are activated (NLGEOM,ON), some element types will be reactivated
in their originally specified geometric configuration. (Large-deformation effects should be included to
obtain meaningful results.)
Thermal strains are computed for newly-activated elements based on the current load step temperature
and the reference temperature. Thus, newborn elements with thermal loads may not be stress-free as
intended. The material property REFT can be used instead of the global TREF to specify material-dependent reference temperatures, allowing you to specify the activation temperature as a stress-free temperature.

12.3. Element Birth and Death Usage Hints

The following guidelines apply to analyses using the element birth and death capability:
Constraint equations (CE, CEINTF, etc.) cannot be applied to inactive DOFs. Inactive DOFs occur when a
node has no active ("alive") elements attached to it.
You can model stress-relieving operations (such as annealing) by deactivating and then reactivating elements.
In nonlinear analyses, be careful not to deactivate or reactivate elements in such a way as to create singularities (such as sharp re-entrant corners in a structural analysis) or sudden large changes in stiffness. Such
situations are likely to cause convergence difficulties.
If the model is entirely linear--that is, if the model has no contact or other nonlinear element present and
the material is linear--except for birth and death elements, the program does not activate optimized defaults
(SOLCONTROL,ON).
Killing contact/target elements or their underlying elements will cause the status of the contact pair to
change to far field contact (open and not near contact), even for bonded contact. You may need to kill
both the contact/target elements and their underlying elements to reestablish the pre-death contact
status when the elements are later reactivated.
The full Newton-Raphson option with adaptive descent activated (NROPT,FULL,,ON) often yields good
results in analyses employing element birth and death.
You can retrieve a parameter whose value will indicate the status (active or inactive) of an element
(*GET,Par,ELEM, n, ATTR, LIVE) This parameter could be used in APDL logical branching (*IF, etc.) or in
other applications for which you need to monitor the birth-and-death status of an element.
Since a Multiframe restart will recreate the database using the *.rdb file, the elements selected in /POST1
can not be killed in a multiframe restart.

300

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Using Birth and Death

The load-step file method (LSWRITE) for solving multiple load steps cannot be used with the birth-death
option, because it will not write the status of deactivated or reactivated elements to the load step file.
Birth and death analyses having multiple load steps must therefore be performed using a series of explicit
SOLVE commands.

12.3.1. Changing Material Properties

You might be tempted to deactivate or reactivate elements by changing their material properties via
the MPCHG command.
You must proceed cautiously if you attempt such a procedure. The safeguards and restrictions applying
to "killed" elements do not apply to elements that have had their material properties changed in the
solution phase of the analysis. (Element forces will not be automatically zeroed out; nor will strains,
mass, specific heat, etc.) Many problems can result from careless use of the MPCHG command. For example, if you reduce an element's stiffness to almost zero, but retain its mass, it could result in a singularity if subjected to acceleration or inertial effects.
One application of the MPCHG command would be in modeling construction sequences in which the
strain history of a "born" element is maintained. Using MPCHG in such cases will enable you to capture
the initial strain experienced by elements as they are fitted into the displaced nodal configuration.

12.4. Using Birth and Death

You can apply element birth and death behavior to most static and nonlinear transient analyses using
the same basic procedures described in the various analysis guides.
Modify your basic analysis procedure as follows to incorporate the element birth and death feature:

12.4.1. Build the Model

While in the PREP7 preprocessor (/PREP7), create all elements, even those that will not be activated
until later load steps. You cannot create new elements after preprocessing.

12.4.2. Apply Loads and Obtain the Solution

For all analyses employing element birth and death, perform the following actions in the solution (SOLU)
phase:

12.4.2.1. Define the First Load Step

In the first load step, choose the analysis type and all appropriate analysis options (ANTYPE).
For a structural analysis, activate large-deflection effects (NLGEOM,ON).
For all birth and death applications, set the Newton-Raphson option to full explicitly in the first load
step (NROPT). (The program cannot predict the presence of an EKILL command in a subsequent load
step.) Deactivate (EKILL) all of the initially inactive elements that you intend to add (reactivate) in later
load steps.
Elements are deactivated (or activated) in the first substep of the load step, and maintain that status
through the rest of the load step. The default reduction factor used as a stiffness multiplier might not
suffice for some problems; sometimes, you may need to use a more severe reduction factor. To provide
a new value for the reduction factor, issue the ESTIF command.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

301

Element Birth and Death

Nodes not connected to any active elements may "float," or pick up stray degree-of-freedom (DOF) responses. You may want to constrain inactive DOFs (D, CP, etc.) in some cases to reduce the number of
equations to be solved and to avoid ill-conditioning. Constraining inactive DOFs can become more
important for cases in which you want to reactivate elements with a specific shape (or temperature,
etc.). If so, remove the artificial constraints when you reactivate elements, and remove nodal loads from
inactive DOFs (that is, at nodes not connected to any active elements). Similarly, you must specifically
add nodal loads (if any) to reactivated DOFs.

12.4.2.1.1. Sample Input for First Load Step

Part of your input listing could look like this for your first load step:
! First load step
TIME,...
NLGEOM,ON
NROPT,FULL
ESTIF,...
ESEL,...
EKILL,...
ESEL,S,LIVE
NSLE,S
NSEL,INVE
D,ALL,ALL,0
NSEL,ALL
ESEL,ALL
D,...
F,...
SF,...
BF,...
SAVE
SOLVE

!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!

Sets TIME value (optional for static analyses)

Turns large-deflection effects on
You must explicitly set the Newton-Raphson option
Sets non-default reduction factor (optional)
Selects elements to be deactivated in this load step
Deactivates selected elements
Selects all active elements
Selects all active nodes
Selects all inactive nodes (those not attached to any
active elements)
Constrains all inactive DOFs (optional)
Selects ALL nodes
Selects ALL elements
Adds constraints as appropriate
Adds nodal loads to active DOFs as appropriate
Adds element loads as appropriate
Adds body loads as appropriate

12.4.2.2. Define Subsequent Load Steps

In the remaining load steps, you can deactivate and reactivate elements as desired. As before, be sure
to apply and delete constraints and nodal loads as appropriate.
To deactivate and reactivate elements, issue the EKILL and EALIVE commands, respectively.

12.4.2.2.1. Sample Input for Subsequent Load Steps

The following simplified input listing demonstrates how you might deactivate and reactivate elements:
! Second (or subsequent) load step:
TIME,...
ESEL,...
EKILL,...
! Deactivates selected elements
ESEL,...
EALIVE,...
! Reactivates selected elements
...
FDELE,...
! Deletes nodal loads at inactive DOFs
D,...
! Constrains inactive DOFs
...
F,...
! Adds nodal loads as appropriate to active DOFs
DDELE,...
! Deletes constraints from reactivated DOFs
SAVE
SOLVE

12.4.3. Review the Results

Typically, you will follow standard procedures when postprocessing an analysis containing deactivated
or reactivated elements.

302

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Where to Find Examples

Be aware that "killed" elements are still present in your model, even though they make an insignificant
contribution to the stiffness (conductivity, etc.) matrix; therefore, they are included in element displays,
output listings, etc. For example, deactivated elements are included in nodal results averaging (via the
PLNSOL command) and will "smear" the results. Ignore the entire element printout for deactivated
elements because many items computed make little physical sense.
To remove deactivated elements for element displays and other postprocessing operations, issue the
ESEL command.

12.4.4. Use Analysis Results to Control Birth and Death

At times, you will not explicitly know the location of elements that you need to deactivate or reactivate.
For example, if you want to "kill" melted elements in a thermal analysis (that is, to model the removal
of melted material), you will not know the location of those elements beforehand and will need to
identify them on the basis of their program-calculated temperatures. When the decision to deactivate
or reactivate an element depends on the value of a result item (such as temperature, stress, strain, etc.),
you can use commands to identify and select the critical elements.
To identify the critical elements, issue the ETABLE command. To select the critical elements, issue the
ESEL command.
You could then deactivate or reactivate the selected elements. To deactivate the selected elements, issue
the EKILL,ALL command. To reactivate the selected elements, issue the EALIVE,ALL command.

Note
You can also use the ANSYS Parametric Design Language to write a macro to perform such
an operation. See the ANSYS Parametric Design Language Guide for more information.

12.4.4.1. Sample Input for Deactivating Elements

The following simplified input listing demonstrates how you might deactivate elements that rupture
when their total strain has exceeded some critical value:
/SOLU
! Enter SOLUTION
..
! Typical solution procedure
SOLVE
FINISH
!
/POST1
! Enter POST1
SET,...
ETABLE,STRAIN,EPTO,EQV ! Store total equivalent strain in ETABLE
ESEL,S,ETAB,STRAIN,0.20 ! Select all elements with total equivalent strain
! greater than or equal to 0.20
FINISH
!
/SOLU
! Re-enter SOLUTION
ANTYPE,,REST
EKILL,ALL
! Deactivate selected (overstrained) elements
ESEL,ALL
! Restore full element set
...
! Continue with solution

12.5. Where to Find Examples

The Mechanical APDL Verification Manual consists of test case analyses demonstrating various analysis
capabilities, including element birth and death.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

303

Element Birth and Death

While the test cases demonstrate solutions to realistic analysis problems, the Mechanical APDL Verification
Manual does not present them as step-by-step examples with lengthy data input instructions and
printouts; however, if you have some finite-element experience, you should have no trouble understanding the problems by reviewing each test case's finite-element model, input data and accompanying
comments.

304

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 13: User-Programmable Features and Nonstandard Uses

The ANSYS program's open architecture allows you to link it to your own Fortran routines and subroutines. In fact, some standard ANSYS features began as user-programmed features.
Typically, you can obtain good results with the ANSYS program when you exercise documented features
using standard, recommended procedures. In some cases, however, you may need to employ nonstandard
procedures that ANSYS, Inc. Quality Assurance has not or cannot fully test.
The following topics concerning the ANSYS program's open architecture are available:
13.1. User-Programmable Features (UPFs)
13.2. Nonstandard Uses of the ANSYS Program

13.1. User-Programmable Features (UPFs)

User-programmable features (UPFs) are ANSYS capabilities for which you can write your own Fortran
routines. UPFs allow you to customize the ANSYS program to your needs, which may be a user-defined
material-behavior option, element, failure criterion (for composites), and so on. You can even write your
own design-optimization algorithm that calls the entire ANSYS program as a subroutine. UPFs are
available in the ANSYS Multiphysics, ANSYS Mechanical, ANSYS Structural, ANSYS Emag - Low Frequency,
ANSYS Emag - High Frequency, and ANSYS PrepPost products. For detailed information, see the Guide
to User-Programmable Features.

Caution
By linking in your own Fortran routines, you are creating a custom, site-specific version of
the ANSYS program. When you use UPFs, you are using ANSYS in a nonstandard way, one
that ANSYS, Inc. verification testing does not cover. You are responsible for verifying that the
results produced are accurate and that the routines you link to ANSYS do not adversely affect
other, standard areas of the program.
Exercise care when using UPFs on parallel systems. Do not use the /CONFIG command or a
config145.ans file to activate parallelism on a system with UPFs.
The following topics concerning UPFs are available:
13.1.1. Understanding UPFs
13.1.2.Types of UPFs Available

13.1.1. Understanding UPFs

UPFs can range from a simple element output routine for custom output to a much more complex user
element or user-optimization algorithm; therefore, it is difficult to present the process without describing
specific programming details. This section presents a general sequence of steps to follow. The Guide to
User-Programmable Features contains more detail on UPFs.
A typical UPF involves the following steps:

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

305

User-Programmable Features and Nonstandard Uses

1. Design and program the desired user routine in Fortran. For more information on Fortran compilers
please refer to either the ANSYS, Inc. Windows Installation Guide or the ANSYS, Inc. Linux Installation
Guide for details specific to your platform or operating system. The source codes for all user routines
are available on your ANSYS, Inc. distribution medium. Most of them demonstrate at least simple
functionality.
2. Compile and link your user routine into the ANSYS program. The Guide to User-Programmable Features
describes how to do this on your system.
3. Verify that the changes you have made do not affect other, standard ANSYS features. (One way to
do so is by running a set of Mechanical APDL Verification Manual problems.)
4. Verify the user routine using whatever procedures you feel are adequate.
The ANSYS program activates some UPFs (such as user elements) automatically when you use them.
For example, to activate a user element, all you need to do is specify it as one of the element types in
the model (via the ET command), set the element type attribute pointer (via the TYPE command) ),
and define elements using the solid modeling (AMESH, VMESH, etc.) or direct generation (ET, etc.)
method.
For other UPFs, you must issue the USRCAL command to activate them. If you fail to issue the command,
standard ANSYS logic applies by default.
For example, when you apply a convection load, the default is to use standard ANSYS logic even if you
have linked a user convection routine. You must activate the appropriate user routine with the USRCAL
command if you want the user convection routine to be used. Refer to the USRCAL command description
for a list of user routines affected by the command. Use the NSVR command to define the number of
extra variables that need to be saved for such user-programmable element options as user plasticity.
(The NSVR command has no equivalent GUI path.)
Another useful command is /UCMD, which allows you to create your own command from a user routine.
Suppose you link in a user routine for a parabolic distribution of pressure. If you name the routine
USERnn (where nn = 01 to 10), you can create your own command to call the routine:
/UCMD,PARAB,1

PARAB now becomes a valid ANSYS command that simply calls the user routine USER01. You can call
up to ten such user routines as commands. By including /UCMD commands in your start-up file
(start145.ans), you can make the user routines available in all of your ANSYS sessions.

13.1.2. Types of UPFs Available

Many UPFs are available in the ANSYS program. Following is a brief description of each:
User-Programmable Features Available in ANSYS
User-defined elements

Allows you to define your own element type. You can add it to the element
library and use it as you would any other element. See User-Defined Elements
in the Element Reference, and Creating a New Element in the Guide to UserProgrammable Features.

User-defined materials

Allows you to define your own material model. See User-Defined Material
Model, and Subroutines for Customizing Material Behavior in the Guide to UserProgrammable Features.

306

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Nonstandard Uses of the ANSYS Program

User-Programmable Features Available in ANSYS
User-defined contact
interfacial behaviors

The contact elements (CONTA171 to CONTA178) allow you to define your own
interfacial behaviors between surfaces. See Subroutines for Customizing Contact
Interfacial Behavior in the Guide to User-Programmable Features.

User real constants

Elements COMBIN37, CONTA171, CONTA172, CONTA173, CONTA174, CONTA175,

CONTA176, CONTA177, and CONTA178 allow the input real constants to be
modified based upon your own nonlinear function.

User thickness

Available for SHELL181, SHELL208, SHELL209, and SHELL281.

User stresses

Available for LINK180, SHELL181, PLANE182, PLANE183, SOLID185, SOLID186,

SOLID187, SOLSH190, BEAM188, BEAM189, SHELL208, SHELL209, REINF264,
SHELL281, and SOLID285.

User plasticity law

Allows you to calculate plastic strains and form the tangent stress-strain matrix
at an integration point based on your own plasticity law.

User creep equation

Allows you to specify your own creep equation.

User swelling law

If you need to account for swelling in an analysis (due to neutron bombardment, for example), you must write the appropriate swelling law as a user
routine. No built-in swelling laws are available.

User failure criteria

Available for the layered elements (such as SOLID185 Layered Structural Solid,
SOLID186 Layered Structural Solid, and SHELL281). Up to nine user-defined
failure criteria can be supplied.

User viscosity

You can define viscosity as a function of pressure, temperature, position, time,

velocity, and velocity gradients for FLUID141 and FLUID142.

User loads

Body loads such as temperatures, heat generations, and fluences (such as

neutron flux), as well as surface loads such as pressures, convections, heat
fluxes and charge density may be defined by way of user-written logic.

ANSYS as a subroutine

You can call the entire ANSYS program as a subroutine in your own program,
such as a user-written design optimization algorithm.

User access at the beginning and end of

each ANSYS run solution, load step, substep, and equilibrium
iteration

Allows you to evaluate results and perform any desired calculations during
solution.

USRSURF116

Allows you to modify SURF151 and SURF152 film coefficients and bulk temperatures based on information from FLUID116.

13.2. Nonstandard Uses of the ANSYS Program

The ANSYS program endures a rigorous verification testing plan before its release. You can be reasonably
assured of obtaining good results when you exercise documented features using standard, recommended
procedures. In some situations, however, you may need to employ nonstandard procedures or techniques
that have not been or cannot be fully tested by ANSYS, Inc. because of their very nature. (An example
is employing user-programmable features.) Be aware that verifying the results in such cases is your responsibility.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

307

User-Programmable Features and Nonstandard Uses

13.2.1. What Are Nonstandard Uses?

The results of nonstandard uses of the ANSYS program cannot be predicted; therefore, ANSYS, Inc.'s
testing cannot fully cover such uses. Although ANSYS, Inc. does not discourage nonstandard uses, you
must exercise caution and use your engineering judgment when doing so. For example, if you program
your own element and use it in an ANSYS analysis, the results depend primarily on how well you programmed the element. In such cases, you must verify the results and make sure that other, standard
areas of the program are not adversely affected.
Following is a partial list of nonstandard ANSYS features and uses:
User programmable features (UPFs) - writing your own user routines, linking them into the ANSYS
executable, and using them in an analysis. UPFs are described earlier in this chapter.
Reading into the ANSYS program an ANSYS file which was created or modified external to the ANSYS
program, for example, a results file or a superelement file created by you or by another program.
High-risk capabilities such as the following:
Changing element real constants during the solution phase in between load steps. Depending on
the element type being used, the element may not properly use the updated real constant value.
Deactivating the cross-reference checking of the solid model (via the MODMSH,NOCHECK command).
Turning off element shape-checking (via the SHPP,OFF command).
Using undocumented features, such as an element option not documented in the Element Reference
or a command argument not mentioned in the Command Reference. Undocumented features are, by
definition, unsupported and unverified; use them with caution.
If the ANSYS program can detect the use of a nonstandard feature, it will often issue a warning message
to that effect.

13.2.2. Hints for Nonstandard Use of ANSYS

Follow these guidelines when you need to use the ANSYS program in a nonstandard manner:
Use your engineering judgment and carefully review the analysis results.
Do not assume that other, standard areas of the program are not affected. Run a few test problems
to verify.
If you need to contact ANSYS, Inc. Technical Support concerning an analysis involving nonstandard
use of the ANSYS program, be sure to mention the nature and extent of the nonstandard feature
that you employed.
For detailed information on UPFs, see the Guide to User-Programmable Features.

308

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 14: State-Space Matrices Export

The goal of reduced-order modeling based on state-space matrices is to generate a model using fewer
equations while still showing the same dynamics characteristics. The state-space matrices formulation
is an easy way to define this reduced model as a function of the inputs and the outputs. It is used in
multi-domain simulation software like Simplorer.

14.1. State-Space Matrices Based on Modal Analysis

The SPMWRITE command can be used to export state-space matrices based on a modal analysis when:
The damping is proportional (e.g., no element damping).
The mass and stiffness matrices are symmetric.
Prior to calculating the state-space matrices, you need to perform a modal analysis using LANB, LANPCG,
or the SNODE eigensolver (MODOPT command). When obtaining the modal solution, consider the following:
Modes do not need to be expanded.
If you are applying nodal loads (forces or torques), you may specify them directly through table parameter
input (inputs on the SPMWRITE command). For other types of loads (e.g. pressure) you must define the
loads in the modal analysis and use the MODCONT command to generate the corresponding load vectors.
If you need an initial condition for your reduced model, you must define the load in the modal analysis
to generate the corresponding load vector (NIC on the SPMWRITE command).
Damping must be specified (see DMPRAT, MDAMP, BETAD, or ALPHAD command).
State-space matrices are calculated in postprocessing using the file Jobname.MODE. They are written
to the file Jobname.SPM using the requested format (FileFormat on the SPMWRITE command).

14.1.1. Examples of SPMWRITE Command Usage

This section describes how to use SPMWRITE to export state-space matrices based on two sets of
model criteria.
1. The first set of model criteria for export to Simplorer are as follows:
10 modes are taken into account.
There are 2 inputs specified in the array parameter inputTab.
The inputs labels are defined in the character array parameter inputLab.
There is 1 output specified in the array parameter outputTab.
The output label is defined in the character array parameter outputLab.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

309

State-Space Matrices Export

The initial condition is calculated from the first load vector.
Velocity and acceleration results are not included.
The SML format for Simplorer is used to printout the state-space matrices.
If your model can be described by this first set of criteria, issue the following to export the statespace matrices:
! ** inputs definition
*dim,inputTab,,2,2
*dim,inputLab,CHAR,2
inputTab(1,1) = 413
! node 413
inputTab(1,2) = 1
! ux
inputLab(1) = n413_ux
inputTab(2,1) = node(50e-2,0,45e-2)
! node @(.5,0,.45)
inputTab(2,2) = 4
! rotx
inputLab(2) = n01_rotx
! ** outputs definition
*dim,outputTab,,1,2
*dim,outputLab,CHAR,1
outputTab(1,1) = 1022
outputTab(1,2) = 1
outputLab(1) = n1022_ux

! node 1022
! ux

spmwrite,modal,10,inputTab,inputLab,outputTab,outputLab,1

2. The second set of model criteria are as follows:

All modes are taken into account.
The inputs are based on all existing load vectors from the modal analysis.
The outputs are equal to the inputs.
There is no initial condition.
Velocity and acceleration results are not included.
The dense format is used to printout the state-space matrices.
If your model can be described by this second set of criteria, issue the following to export the statespace matrices:
spmwrite,modal, ,, ,, ,, ,0

14.1.2. Example of Reduced Model Generation in ANSYS and Usage in Simplorer

14.1.2.1. Problem Description
The model in this example is a basic representation of a disk drive suspension as described in Hatch[1].
It is a clamped-free beam with a mass and spring at tip. The mass represents the recording head. The
reduced model is generated in ANSYS and imported into Simplorer where global half-sine acceleration
is applied in a transient analysis.

310

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

State-Space Matrices Based on Modal Analysis

14.1.2.2. Problem Specifications

Material
Young Modulus: 190e+6 mN/mm2
Density: 7.83e-6 kg/mm3
Poisson ratio: 0.293
Beam
Length: 20 mm
Area: 0.15 mm2
Cross section inertia IYY: 0.05 mm4
Cross Section inertia IZZ: 0.00007031 mm4
Spring
Stiffness: 1.e+7 N/mm
Mass at tip
Mass: 1.e-6 kg

14.1.2.3. Input File for the Analysis

Use the input below to generate the state-space matrices file (file.spm) based on the first 3 modes
plus the residual vector.
/title, Beam on Shaker (ACEL) - SPMWRITE 3 modes + resvec
/prep7
! ** Parameters
num_elem = 16
nummodes = 2*num_elem
! ** Material
ex,1,190e6
dens,1,7.83e-6
nuxy,1,0.293
! ** Properties
r,1,0.15,0.05,0.00007031,0.075,0.2
r,2,1000000
r,3,0.00002349,0.00002349,0.00002349
! Nodes
n,1,0,0,0
n,num_elem+1,20,0,0
fill,1,num_elem+1
n,num_elem+2,20,0,-3
nall
! ** Beam
et,1,4
type,1
mat,1
real,1
e,1,2
egen,num_elem,1,-1
! ** Spring
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

311

State-Space Matrices Export

et,2,14
type,2
real,2
e,num_elem+1,num_elem+2
! ** Mass
et,3,21
type,3
real,3
e,num_elem+1
! ** Constraints
d,1,all,0
d,num_elem+2,all,0
nsel,s,node,,2,num_elem+1
d,all,ux
d,all,uy
d,all,rotx
d,all,rotz
nsel,all
fini
! ** Modal analysis
/solu
antype,modal
modopt,lanb,3
! << 3 modes
mxpand,3
resvec,on
! << residual vector
acel,,,1.0
! << load vector definition
dmprat,0.02
! << modal damping
solve
fini
/post1
/eshape,1
! ** Definition of the output
*dim,output,,1,2
output(1,1) = num_elem+1 ! << node at tip
output(1,2) = 3
! << UZ
*dim,outputLab,CHAR,1
outputLab(1) = 'UZtip'
! ** Write SPM file: input = load vector
spmwrite, modal, ,,, output, outputLab
*list,,spm
fini

The state-space matrices file can be imported in Simplorer using: Simplorer Circuit > SubCircuit >
Add Mechanical Component
A transient analysis is performed in Simplorer for the 3 following reduced models:
All 32 modes are used.
First 3 modes are used.
First 3 modes plus residual vector are used (ANSYS input is listed above).
The schematic of the simulation is shown on the figure below:

312

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

State-Space Matrices Based on Modal Analysis

Figure 14.1: Simulation Schematic

A zoom on the graph showing the evolution of the spring force is given below:
Figure 14.2: Evolution of Spring Force

The reduced model based on the first 3 modes (red curve) is not accurate. When the residual vector is
included (purple curve), the spring force is similar to the reference given by the 32 modes reduced
model (dotted curve).

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

313

314

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 15: Soil-Pile-Structure Analysis

You can analyze the interaction of a structure, supported on one or more piles, with an elastic or inelastic
soil. The analysis can take into account the lateral force-displacement, and the end-bearing and skinfriction responses, of the soil layers occurring at the pile location. It is not necessary for all piles in the
analysis to be situated in identical geological strata.
Data input is defined primarily via the PILExxxx family of commands, with soil data defined in additional
text files. Piles are connected to MATRIX27 elements which must be defined at the pile caps.
A number of codes of practice can be used to assess the strength of the piles.
The following soil-pile-structure analysis topics are available:
15.1. Soil-Pile-Structure Interaction Analysis
15.2. Soil Data Definition and Examples
15.3. Performing a Soil-Pile Interaction Analysis
15.4. Soil-Pile-Structure Results
15.5. References

15.1. Soil-Pile-Structure Interaction Analysis

MATRIX27 elements are used to locate the pile and to connect it to an above ground structure, or to
apply loads directly for single pile only modeling. For pile stiffness modelling, use MATRIX27 element
with KEYOPT(3) = 4. If pile mass is also required, another MATRIX27 element is required with KEYOPT(3)
= 2. Each pile stiffness and mass element should be given a separate real constant number. The stiffness
and/or mass values will be updated by issuing the relevant commands (PILEMASS, PILESTIF). The two
nodes on the MATRIX27 stiffness element define the pile cap and pile tip position, respectively. The
pile cap node is the pile node that is connected to the supported structure.
The soil material properties must be specified on a layer by layer basis. For each stratum, the lateral
force-displacement curve (P-Y curve) and the skin friction force-displacement curve (T-Z curve) may be
input. For the lowest stratum the end-bearing force-displacement curve may also be defined. All of
these three force-displacement relationships are piecewise linear curves with no restriction on the
number of line segments allowed.
For the initial conditions the soil stiffness, as defined by the curves for zero displacement, is added to
the pile stiffness and the displacements are calculated, assuming no change in the soil properties. These
displacements are compared with the values at which the soil force-displacement curves become nonlinear. If the soil has become non-linear, that is the force-displacement relationship is no longer on the
initial part of the curve, the soil stiffness is recalculated for the current displacement. The modified soil
stiffness is added to the pile stiffness and a new set of displacements calculated. Iteration continues in
this way until the displacements calculated in an iteration differ from those of the preceding iteration
by less than a small tolerance.
These converged values of displacement are taken to be an adequate estimate of the true behavior.
The program outputs the pile cap displacements and forces, and the piles displacements.
Unity checks may be requested against the following codes of practice:
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

315

Soil-Pile-Structure Analysis
American Petroleum Institute (API) Recommended Practice for Planning, Designing and Constructing Fixed
Offshore Platforms - Working Stress Design, RP2A-WSD, Twenty-first Edition, December 2000 (the Seventeenth
and Twentieth editions are also supported).
American Petroleum Institute (API) Recommended Practice for Planning, Designing and Constructing Fixed
Offshore Platforms - Load and Resistance Factor Design, RP2A-LRFD, First Edition, July 1, 1993.
International Standards Organization (ISO) 19902, Petroleum and Natural Gas Industries - Fixed Steel Offshore
Structures, First Edition, December 2007.
In order to calculate these unity checks, yield data must be supplied for the piles. For WSD checks, extreme environmental loads may be specified.
For a transient analysis, the pile calculation should be calculated at each time step, in order for the
updated loading (in the form of prescribed displacements) to be applied to the pile and a new stiffness
and loading to be calculated and applied to the pile cap. In this case FDELE should be used to ensure
that previously calculated nodal forces are not appended to.

Note
Multiple loadsteps and small time increments are necessary to allow for the nonlinear iterations
required for the soil-pile interaction equations, and to enable these calculations to converge.

15.1.1. Automatic Pile Subdivision

Each pile is defined by two nodes on the MATRIX27 stiffness element. In order to calculate the structural
stiffness of the pile and the soil to a sufficient degree of accuracy, the individual piles are automatically
subdivided into a number of finite elements. Pile subdivisions will occur based on the following criteria:
1. The subdivision length will be related to the nominal diameter of the pile multiplied by a basic subdivision
value (default 1.0) plus a percentage of the depth below the local mudline given by the subdivision
modifier (default 0.1 or 10%). The depth used is taken at the mid-point of the existing subdivision. Both
the basic subdivision value and the subdivision modifier may be specified in the PILEGEN command.
2. Each subdivision will be greater than 10% of the nominal pile diameter.
3. Whether there are any changes in the soil properties due to crossing from one stratum to another.
It is important to note that while the automatic subdivision process will provide a satisfactory result for
general soil conditions, there are instances when user defined divisions are preferable. In particular, if
only a few soil layers have been defined then the resultant subdivisions may prove to be too coarse
and provide poor lateral displacements. It is suggested that the resultant subdivision information be
inspected for large sub-elements in the critical top one third of the pile. As a rule of thumb, the subdivisions should be less than two times the pile diameter in the critical zone.
Item 2, above, prevents ill conditioning due to two closely spaced points occurring in the subdivision
process. Where violation of this criterion occurs, the points are adjusted to be coincident. The program
defined default cannot be amended.

15.1.2. Convergence Criteria

The analysis process is based upon the solution of the simultaneous equations relating to the pile caps
and associated structural elements. Since, in general, the soil is of a nonlinear nature the solution

316

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil-Pile-Structure Interaction Analysis

method utilizes an iterative technique whereby the soil stiffness and associated reactive loads are updated
from the results of the previous iteration. Successive iterations modify the soil properties with progressively smaller resultant changes to the behavior of the pile model. In order that the analysis terminates
within a finite number of iterations, a convergence tolerance is applied as follows:
On the first iteration the pile cap displacements and reactions are calculated and stored.
On the second and subsequent iterations, revised pile cap displacements and reactions are computed
and compared with the previous iteration. If the percentage difference is greater than that defined (either
by program default or user input) the revised displacements are stored and the iteration process continued.
The default convergence criteria is taken as 5%. This can be modified using the option in the PILEGEN
command.
It is suggested that a lower limit of 0.1% for a single pile is observed in order to prevent an excessive
number of iterations for convergence.
In addition to the convergence criteria described above, a simple divergence check is included in order
to trap numerical instabilities introduced due to multiple pile failure or highly nonlinear systems. In
some instances it may be desirable to switch this facility off using the ITER option on the PILEGEN
command. It is suggested that this option is included only after reference to the iteration report confirms
that no inherent instability exists.

15.1.3. Soil Representation

The nonlinear foundation analysis method utilizes a finite element representation of the pile and soil
system. Figure 15.1: Pile/Spring Foundation Model (p. 317) illustrates a typical element representing a
pile segment and associated springs employed to model the soil.
Figure 15.1: Pile/Spring Foundation Model

The springs utilized for the soil model are characterized by a nonlinear force-deflection relationship of
the type shown in Figure 15.2: Soil/load Deflection Characteristics (p. 318). These are commonly known
as P-Y curves for lateral behavior and T-Z curves for axial behavior.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

317

Soil-Pile-Structure Analysis
Figure 15.2: Soil/load Deflection Characteristics

The nonlinear nature of the soil properties means an iterative solution technique is required. Each iteration takes an assumed, or calculated, value for the soil spring stiffness based upon the previous iteration.
Two solution methods exist for determining the representative soil stiffness from the defined P-Y and
T-Z curves, namely Tangent and Secant stiffness. These are shown diagramatically below:

For both types of soil representation, the curves are stored as a series of points as defined by the user
and depicted by crosses in the diagrams above. In order that the soil stiffness may be computed, the
soil properties are assumed to vary linearly between these points as shown above. For points beyond
the last y datum defined, linear extrapolation is utilized from the last two points supplied.
In general, the tangent stiffness method will converge faster than the secant stiffness approach, especially
as the pile approaches its limiting capacity. The tangent stiffness method is adopted by default. The
default may be overridden by the STIFF option in the PILEGEN command.
There is no limit to the number of points on any curve. For displacements that occur outside of the
range specified for a given curve, linear extrapolation of the last segment of the curve is assumed. If a
constant value is required, two points with similar P or T and different displacement must be input.

318

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil-Pile-Structure Interaction Analysis

Both symmetric and non-symmetric curves may be defined for soil layer data. If symmetric curves are
required, only that part of the curve for positive displacement values need be input. The program will
automatically generate the remaining information.
If any values are defined for a negative displacement, then the curve is assumed to be non-symmetric.
Care should be taken in providing non-symmetric P-Y data since the soil stiffnesses derived from the
curves are based upon local axes displacements, and these may vary from pile to pile and from iteration
to iteration. Shifted P-Y curves are not permitted, since this can result in undesired lateral deflections.
Use the SLID definition if shifted curves are required.
For T-Z and ENDB data, a positive local displacement for the purposes of soil stiffness formulation is
taken as being defined by the vector going from the pile cap to the pile tip. Thus, zero tensile stiffness
for end bearing forces may be modelled by supplying an ENDB definition with zero stiffness for negative
displacements.
The soil properties must be defined down to the full depth of the pile, or to a greater depth.
If both top and bottom layer depths are supplied on a P-Y or T-Z header, the data is taken as a constant
between these depths. If only one depth is supplied, the data is defining the properties at one depth
in the soil medium. P-Y and T-Z curves, either explicitly defined or generated from the soil properties,
are assumed to vary linearly between these depths.
If a sudden change in soil properties is required at a given depth, one of two options are available:
1. If one or both of the soil definitions represent a constant stratum, the given depth may be supplied for
both layers.
2. If both the soil definitions are single point, they must be separated by a finite distance so that the program
can identify which layer is uppermost. Provided the separation is less than the coordinate tolerance (0.1
x pile diameter) the program will utilize the data as though they represented coincident layers. The
higher of the two levels specified will be adopted as the point of the discontinuity. See also Automatic
Pile Subdivision (p. 316).

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

319

Soil-Pile-Structure Analysis
The procedures used for developing P-Y and T-Z curves from user defined soil properties are as indicated
in the list below. Detailed descriptions are not given but a typical curve for each procedure is shown.

Note
Overburden pressures used in the soil curve generation are computed from the soil densities
provided down the soil profile. If explicit P-Y and/or T-Z data is provided for any soil layer
within a soil profile, then overburden pressures will be computed based upon the soil
density local to the point of calculation.
The factor used for shaft friction is automatically calculated as specified in Clause 6.4.2 of
the API 21st edition. A limiting value of is set to 3.0. However no allowance is made for
pile length.
The resulting soil curves depend upon the pile to which it is to be applied since, in the
general form, pressures are generated. A different soil curve set will thus be produced for
each pile in an analysis. Where stepped piles are utilized, it is important that two soil definitions are provided at the step position(s) in order that the correct geometric data is utilized
for the curve generation.
Procedures for Developing P-Y and T-Z Curves
For sands, the method suggested by Reese, et. al. [1] is used.
Figure 15.3: P-Y Curve for Sand

For clays, the method suggested by Matlock [2], a modified Matlock procedure, and the basic recommendations as listed in API 15 [3] are available.

320

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil-Pile-Structure Interaction Analysis

Figure 15.4: P-Y Curve for Clay: Static Loading

Figure 15.5: P-Y Curve for Clay: Cyclic Loading

For T-Z and ENDB curves, the procedures recommended by Vijayvergiya [4] have been adopted.
For the purposes of the implementation in the ANSYS soil-pile analysis, the ultimate skin friction and
end bearing pressure that can be developed in cohesionless soils (sand) is limited to the values given
in the API RP2A code of practice. The user should also note that for plugged pile conditions it is assumed that any internal soil skin friction is sufficient to sustain the plug in position; there is no internal
check undertaken to check this requirement.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

321

Soil-Pile-Structure Analysis
Figure 15.6: T-Z Curve for Clay and Sand

Figure 15.7: ENDB Curve for Clay and Sand

Note
For P-Y, T-Z and ENDB curves, the curved section is divided into four divisions by default.
For T-Z curves, Zc is taken to be 0.01 x pile diameter.
For ENDB curves, Zc is taken to be 0.05 x pile diameter.
Whenever the soil is defined as banded, P-Y and T-Z curves are evaluated at the mid-point
of the band.

15.1.4. Mudslides
The soil is assumed to be initially in an undisturbed state and all pile displacements are referred to as
absolute soil origin. There are instances, however, where displacement of the soil itself may occur due

322

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil-Pile-Structure Interaction Analysis

to the effects of external influences such as general soil movement (mudslides) or the proximity of adjacent structures.
The soil movement induces an additional loading on the pile, in a similar manner to that experienced
by piles within a group. The loading is a function of the relative displacement between the pile and
the surrounding soil. The soil displacement thus has the effect of shifting the P-Y (for horizontal slides)
and T-Z (for vertical slides) curves by an amount equal to the local soil movement.
Where such soil shifts are required, the SLID input provides a facility to define soil displacements with
respect to the global axes thus permitting directional shifting of the curves which would not be otherwise
possible for P-Y data.

15.1.5. Soil-Pile Interaction Results

In addition to displacements, reactions, stresses, or other results, the PILECALC command produces a
list of data and useful intermediate information. This data can be found in the Jobname_CAPNODE.POUT
file in the solver output folder.

15.1.5.1. Displacements and Reactions

The displacement listing is in three parts:
During the iteration stage of the program, pile cap displacements and reactions, along with the convergence
parameters will be output in a section titled Pile Cap Displacements And Reactions Iteration Report.
Displacements and reactions for all the nodes on the structure (excluding the nodes defining the bottom
of the piles) are output. The displacements and reactions will be in the global coordinate system.
A detailed report of the global displacements down the piles is also output in the global coordinate system.

15.1.5.2. Forces and Stresses

For each pile element, a detailed report providing both forces and axial, bending, and combined stresses
is printed for all the subdivisions down the pile. The combined stress is calculated as the sum of the
absolute values of axial and biaxial bending stresses, and adopts the sign of the axial stress.

15.1.5.3. UNITY Check Data

Optional utilization check results for either API LRFD, API WSD or ISO codes of practice may also be reported by selecting to include them in the PILEGEN commands CHECK option.
The API LRFD standard utilizes limit state checks with resistance coefficients to achieve the desired level
of safety. In keeping with this principle, applied loads must be multiplied by appropriate factors, as
defined in the code of practice (Section C, Loads), to develop the design load case combinations necessary
for processing. Since the pile-soil interaction is nonlinear, any loading must be factored before being
applied.
For the WSD method the desired safety limits are achieved using safety factors. In order to account for
extreme environmental conditions a one-third increase in the allowable stress is permitted, and this
can be requested using the LOADTYPE option in the PILEGEN command.
The ISO 19902 checks adopt a similar philosophy to the API LRFD code. Thus, applied loads must be
multiplied by appropriate factors, as defined in the code of practice (Section 9.10.3). The LOADTYPE

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

323

Soil-Pile-Structure Analysis
option in the PILEGEN command can be used to define extreme loadcases, which will affect the design
axial pile capacity computed because a different resistance factor is used.
The following utilization checks are undertaken when requested:

API WSD (Edition 21)

where

 

fa = axial stress
Fxc = inelastic local buckling stress
fby, fbz = bending stresses (including second order effects using the PDELTA option in the PILEGEN
command)
Fb = allowable bending stress
A one-third increase in the allowable stresses is included if an extreme loadcase is being processed.

API LRFD (Edition 1)



where


 

!
 
" "

fc = axial compressive stress

Fxc = inelastic local buckling strength
fby, fbz = bending stresses (including second order effects using the PDELTA option in the PILEGEN
command)
Fbn = nominal bending strength
c = resistance factor for axial compressive strength
b = resistance factor for bending
Note: for tensile piles the axial term of the above check is changed to

'%&
&$ #

where

324

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil Data Definition and Examples

ft = axial tensile stress
Fy = yield stress
t = resistance factor for axial tensile strength

ISO 19902 (Edition 1)

For axial tension and bending:

 
 
=

For axial compression and bending:

#!

"!  !  ! $   $  
% %
=

where

ft = axial tensile stress

Ft = representative axial tensile strength ( = yield stress)
R,t = partial resistance factor for axial tensile strength
fc = axial compressive stress
Fyc = representative local buckling strength
R,c = partial resistance factor for axial compressive strength
fby, fbz = bending stresses (including second order effects using the PDELTA option in the PILEGEN
command)
Fbn = nominal bending strength
R,b = partial resistance factor for bending strength

15.2. Soil Data Definition and Examples

The following topics related to defining soil data are available:
15.2.1. Soil Profile Data Definition
15.2.2. Soil Data File Examples

15.2.1. Soil Profile Data Definition

The soil data is provided in a text file which must be named Jobname_SOIL.TXT.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

325

Soil-Pile-Structure Analysis
It is assumed that the soil data file is available in the working directory. If this is not the case, the /COPY
command can be used to copy the data file into the working directory and rename it if necessary. For
example, the following command will copy the soil data file soil_1.txt from folder C:\soildata
to the working directory and rename it to be file_soil.txt.
/COPY,'C:\soildata\soil_1',txt,,file_soil

This data defines the inelastic and elastic properties and layering details for each soil profile required.
The data is divided into various categories, their use depending on the type of analysis being carried
out and the input requirements of the user. A complete soil profile is shown below followed by detailed
descriptions of each of the categories in turn.

326

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil Data Definition and Examples

15.2.1.1. Mudline Position Definition

This data type is used for defining a global mudline which is different from the global origin. Only one
statement is permissible and can appear anywhere within the soil information.

Parameters
MUDD
Keyword to identify this as mudline data.
muddat
Z coordinate of mudline datum relative to the global axes system. If omitted, the mudline is assumed
to pass through the global origin.
The MUDD command is used for defining a global mudline which does not coincide with the global
origin. This can be particularly useful if the structure has been defined with its origin at somewhere
other than the mudline and the user wishes to retain the same coordinate system.
If a particular soil profile requires a local scour to be modelled, both the P-Y and T-Z data should start
at a depth corresponding to the scour level. Dummy data does not have to be provided over this depth.

15.2.1.2. Common Factors for P-Y, T-Z Curves

The following data types enable the user to factor P-Y and T-Z data to suit their input requirements.
They do not relate specifically to one soil profile and will factor all appropriate soil layers following their
inclusion. The factors may be modified within the soil data by inclusion of two or more such lines.
The factor capability enables users to automatically multiply all their input data by a constant (or constants). This has several uses:
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

327

Soil-Pile-Structure Analysis
1. If the data supplied to the user are not in the correct form, (for example, pressure curves instead of force
per unit length or the wrong units such as N and mm instead of KN and m) then factors may be applied
to convert from one to the other.
2. If the data supplied is for a given diameter pile and a different diameter pile is to be analyzed, a factor
may be applied to P and T data equal to the ratio of the two diameters.
3. If a simplified group analysis is to be carried out utilizing Y and Z modifiers, the data can be factored
appropriately.

Parameters
PFAC
Keyword to signify the factor relates to P data.
YFAC
Keyword to signify the factor relates to Y data.
TFAC
Keyword to signify the factor relates to T data.
ZFAC
Keyword to signify the factor relates to Z data.
factor
Factor by which the data is to be multiplied (Real).
Note that both T-Z and ENDB data will be modified by the TFAC and ZFAC statements.

15.2.1.3. Horizontal Soil Properties (P-Y)

These properties may be defined in one of two ways, either explicitly or using parametrically generated
curves. These data sets are repeated until all P-Y soil layers for the given soil integer have been defined.

328

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil Data Definition and Examples

15.2.1.3.1. P-Y curves defined explicitly

Parameters
isoil
Identifying number of the soil profile (Integer). Only required on the first P-Y line of a given soil property.
P-Y
Keyword for P-Y data.
toplay
Depth to top of layer (Real).
botlay
Depth to bottom of layer (Real). If omitted, curve is defined at one level only. Depths specified are relative
to global mudline datum as defined using the MUDD data statement. A positive depth corresponds to
a point below the mudline datum.
P
Keyword defining P data ordinates from the P-Y curve.
funitl
Force per unit depth. Value taken from P-Y curve (Real). Only 80 characters are permitted per line. If
more data is required it can be entered on subsequent lines preceded with a colon.
Y
Keyword defining Y data ordinates from the P-Y curve.
disp
Displacement value taken from P-Y curve (Real). The order of the data must correspond with that used
for the P ordinates. Only 80 characters are permitted per line. If more data is required it can be entered
on subsequent lines preceded with a colon.

15.2.1.3.2. P-Y curves generated from given soil properties

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

329

Soil-Pile-Structure Analysis
Parameters
isoil
Identifying number of the soil profile (Integer). Only required on the first P-Y line of a given soil profile.
P-Y
Keyword for P-Y data.
toplay
Depth to top of layer (Real).
botlay
Depth to bottom of layer (Real). If omitted curve is defined at top layer level only. See Soil Representation (p. 317). Depths specified are relative to global mudline datum as defined using the MUDD data
statement. A positive depth corresponds to a point below the mudline datum.
REES, , MATL, , MMAT, , PY15
One of four keywords specifying the procedure to be used to generate the P-Y curve. The possible values
are shown below. See Soil Representation (p. 317) for more information.
REES use the Reese method for Sand soil types.
MATL the Matlock method can be used for Clay soil types.
MMAT the modified Matlock method can be used for Clay soil types.
PY15 the API method can be used for Clay soil types.
SAND, , CLAY
Keywords specifying the type of soil under consideration.
Soils may be of two general types: cohesionless (SAND) or cohesive (CLAY).
The following procedures and soil types are related as follows:
Soil type

Solution procedure

SAND

REES

CLAY

MATL, MMAT or PY15

variables
Soil properties required to generate the P-Y curves. (Real)
For soil type SAND these are:
unit weight of soil
angle of internal friction
initial soil modulus
For soil type CLAY these are:
unit weight of soil
soil shear strength

330

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil Data Definition and Examples

strain at half the failure strength
STAT, , CYCL
Keywords specifying the type of loading under consideration. This may be either static (STAT) or cyclic
(CYCL). If omitted, cyclic loading is assumed.

15.2.1.4. Vertical Soil Properties (T-Z)

These properties may be defined in one of two ways, either explicitly or using parametrically generated
curves.
These data sets are repeated until all T-Z soil layers for the given soil integer have been defined.
For unplugged piles, the internal skin friction will be ignored. Approximate changes to the T-Z data
must be applied to take account of internal skin friction.

15.2.1.4.1. T-Z curves defined explicitly

Parameters
isoil
Identifying number of the soil profile (Integer). Only required on the first T-Z line for a given soil profile.
T-Z
Keyword for T-Z data.
toplay
Depth to top of layer. (Real)
botlay
Depth to bottom of layer (Real). If omitted curve is defined at top layer level only. See Soil Representation (p. 317). Depths specified are relative to a global mudline datum as defined using the MUDD data
statement. A positive depth corresponds to a point below the mudline datum.
T
Keyword defining T data ordinates from the T-Z curve.
funitl
Force per unit depth. Value taken from T-Z curve (Real). Only 80 characters are permitted per line. If
more data is required it can be entered on subsequent lines preceded with a colon.
Z
Keyword defining Z data ordinates from the T-Z curve.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

331

Soil-Pile-Structure Analysis
disp
Displacement value taken from T-Z curve (Real). The order of the data must correspond with that used
for the T ordinates. Only 80 characters are permitted per line. If more data is required it can be entered
on subsequent lines preceded with a colon.

15.2.1.4.2. T-Z curves generated from given soil properties

Parameters
isoil
Identifying number of the soil profile (Integer). Only required on the first P-Y data line of a given soil
profile.
T-Z
Keyword for T-Z data.
toplay
Depth to top of layer. (Real)
botlay
Depth to bottom of layer (Real). If omitted curve is defined at top layer level only. See Soil Representation (p. 317). Depths specified are relative to global mudline datum as defined using the MUDD data
statement. A positive depth corresponds to a point below the mudline datum.
V15O, , V15C
Keyword specifying the procedure to be used to generate the T-Z curve. See Soil Representation (p. 317).
The pile may be considered to be either plugged (V15C) or unplugged (V15O).
SAND, , CLAY
Keywords specifying the type of soil under consideration.
Soils may be of two general types: cohesionless (SAND) or cohesive (CLAY).
Either soil type may be used with each solution procedure.
variables
Soil properties required to generate the T-Z curves. (Real)
For soil type SAND these are:
unit weight of soil
angle of internal friction
For soil type CLAY these are:

332

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil Data Definition and Examples

unit weight of soil
soil shear strength
It is important that the properties are given in the order listed above.
SYMM, , ASYM
Keywords specifying the shape of the T-Z curves. These may be either symmetric (SYMM) or axisymmetric
(ASYM). If omitted, a symmetric curve is assumed.
factor
Multiplication factor required for the tensile (negative) portion of the T-Z curve (Real). If omitted, a value
of 0.5 is assumed.

15.2.1.5. End Bearing Properties (ENDB)

This data is optional and may be omitted if friction piles are being analyzed. Only one curve is required
since it is assumed to act upon the pile tip. These properties may be specified in one of two ways, either
explicitly or using generated curves.

15.2.1.5.1. ENDB curve defined explicitly

Parameters
isoil
Identifying number of the soil profile. (Integer)
ENDB
Keyword for end bearing data.
T
Keyword defining T data ordinates from ENDB curve.
force
Force value taken from ENDB curve (Real). Only 80 characters are permitted per line. If more data is required it can be entered on subsequent lines preceded with a colon.
Z
Keyword defining Z data ordinates from ENDB curve.
disp
Displacement value taken from ENDB curve (Real). The order of the data must correspond with that used
for the T ordinates. Only 80 characters are permitted per line. If more data is required it can be entered
on subsequent lines preceded with a colon.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

333

Soil-Pile-Structure Analysis

15.2.1.5.2. ENDB curves generated from given soil properties

Parameters
isoil
Identifying number of the soil profile. (Integer)
ENDB
Keyword for ENDB data.
V15O, , V15C
Keyword specifying the solution procedure to be used to generate the ENDB curve. See Soil Representation (p. 317). The pile may be considered to be either plugged (V15C) or unplugged (V15O).
SAND, , CLAY
Keywords specifying the type of soil under consideration. Soils may be of two general types: cohesionless
(SAND) or cohesive (CLAY). Either soil type may be used with each solution procedure.
variables
Soil properties required to generate the ENDB curves (Real).
For soil type SAND these are:
unit weight of soil
angle of internal friction
For soil type CLAY these are:
unit weight of soil
soil shear strength
It is important that the properties are given in the order listed above.
SYMM, , ASYM
Keywords specifying the shape of the ENDB curves. These may be either symmetric (SYMM) or axisymmetric (ASYM). If omitted, a symmetric curve is assumed.
factor
Multiplication factor required for the tensile (negative) portion of the ENDB curve (Real). If omitted, a
value of 0.0 is assumed.

334

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil Data Definition and Examples

15.2.1.6. Mudslide Definition

This command enables a soil displacement profile to be defined, allowing computation of relative pilesoil displacements. In addition to traditional mudslide conditions, this command may be used in any
situation where the soil is displaced; for example, conditions produced by the vicinity of an adjacent
structure such as the spud can of a Jack-Up drilling platform. The command has the effect of internally
generating shifted P-Y and/or T-Z curves which correspond to the soil displacement profile.

Parameters
isoil
Identifying number of soil profile as used on P-Y and T-Z data. Only required on the first SLID command
line of a given soil profile.
SLID
Compulsory keyword to signify that soil displacement data is being defined.
depth
Depth from mudline to point where soil displacement data is being defined, positive downwards.
xdisp
Soil displacement in the global X axis.
ydisp
Soil displacement in the global Y axis.
zdisp
Soil displacement in the global Z axis.

Note
There are no limits to the number of soil movement definitions for a given soil.
The depths specified do not have to coincide with other soil data, such as P-Y and T-Z information. The program will interpolate values for unspecified depths. For pile depths outside the
values specified on the SLID definition, a constant value is adopted equal to the value defined
at an adjacent depth.
The vertical soil displacement (zdisp) may be omitted if zero. Similarly if both ydisp and zdisp
are zero then they may be omitted.

15.2.2. Soil Data File Examples

The following examples show how to set up soil data files for different soil conditions.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

335

Soil-Pile-Structure Analysis

15.2.2.1. Example 1: Constant Linear Soil

Simple constant linear soil defined between mudline and 100m depth.
T-Z data factored to account for diameter and unit inconsistencies.
Y data factored to change units.
ENDB data does not require factoring so TFAC and ZFAC set to unity.
YFAC 0.01
10 P-Y 0.0 100.0
P
0.0 10.0
Y
0.0 0.1
TFAC 2.0
ZFAC 0.1
10 T-Z 0.0 100.0
T
0.0 5.75
Z
0.0 0.1
TFAC 1.0
ZFAC 1.0
10 ENDB
T
0.0 75.0
Z
0.0 1.0

15.2.2.2. Example 2: Non-Linear Soil

Non-linear soil with global mudline set at -10.0
3 layers defined:
1 banded layer between mudline and 10m.
Discontinuity defined at 10.0m.
Lower layer defined at 30.0m.
Non-symmetric ENDB data supplied with zero tensile capability.
MUDD -10.0
5 P-Y 0.0 10.0
P 0.0 105.0 175.0 71.0 71.0
Y 0.0 0.014 0.06 1.0 10.0
P-Y 10.0
P 0.0 125.0 200.0 185.0 185.0
Y 0.0 0.014 0.06 1.0 10.0
P-Y 30.0
P 0.0 720.0 1000.0 750.0 750.0
Y 0.0 0.014 0.06 1.0 10.0
5 T-Z 0.0 10.0
T 0.0 50.0 50.0
Z 0.0 0.012 1.0
T-Z 10.0
T 0.0 170.0 170.0
Z 0.0 0.012 1.0
T-Z 30.0
T 0.0 320.0 320.0
Z 0.0 0.012 1.0
5 ENDB
T 0.0 0.0 9000.0 9000.0
Z -1.0 0.0 0.085 1.0

336

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil Data Definition and Examples

15.2.2.3. Example 3: Soil Properties Defined in 5 Layers

Generated soil properties defined in 5 layers.
The soil is clay.
The P-Y data is defined using the PY15 procedure.
T-Z and End bearing data is defined using the V15O procedure.
1 P-Y 0.0 2.0
PY15 CLAY 15. 30.
1 P-Y 2.0 5.0
PY15 CLAY 15. 30.
1 P-Y 5.0 10.0
PY15 CLAY 15. 30.
1 P-Y 10.0 27.5
PY15 CLAY 15. 30.
1 P-Y 27.5 100.0
PY15 CLAY 15. 30.
1 T-Z 0.0
V15O CLAY 15. 30.
1 T-Z 2.0 5.0
V15O CLAY 15. 30.
1 T-Z 5.0 10.0
V15O CLAY 15. 30.
1 T-Z 10.0 27.5
V15O CLAY 15. 30.
1 T-Z 27.5 100.0
V15O CLAY 15. 30.
1 ENDB
V15O CLAY 15. 30.

0.01 STAT
0.01 STAT
0.01 STAT
0.01 STAT
0.01 STAT
SYMM
SYMM
SYMM
SYMM
SYMM
SYMM

15.2.2.4. Example 4: Soil Properties Defined in 5 Layers with Mudslide

As example 3 but including a horizontal mudslide in the global x direction which varies from 200 mm
at the mud line to 50 mm at 30 m depth and beyond.
1 P-Y 0.0 2.0
PY15 CLAY 15. 30. 0.01
1 P-Y 2.0 5.0
PY15 CLAY 15. 30. 0.01
1 P-Y 5.0 10.0
PY15 CLAY 15. 30. 0.01
1 P-Y 10.0 27.5
PY15 CLAY 15. 30. 0.01
1 P-Y 27.5 100.0
PY15 CLAY 15. 30. 0.01
1 T-Z 0.0
V15O CLAY 15. 30. SYMM
1 T-Z 2.0 5.0
V15O CLAY 15. 30. SYMM
1 T-Z 5.0 10.0
V15O CLAY 15. 30. SYMM
1 T-Z 10.0 27.5
V15O CLAY 15. 30. SYMM
1 T-Z 27.5 100.0
V15O CLAY 15. 30. SYMM
1 ENDB
V15O CLAY 15. 30. SYMM
1 SLID 0.0 0.200
SLID 2.5 0.190
SLID 7.5 0.180
SLID 10.0 0.170
SLID 20.0 0.120
SLID 30.0 0.050

STAT
STAT
STAT
STAT
STAT

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

337

Soil-Pile-Structure Analysis

15.3. Performing a Soil-Pile Interaction Analysis

You can use Mechanical APDL commands to perform a soil-pile interaction analysis from an ANSYS
Mechanical analysis. The commands can be placed in a Mechanical APDL input file, or they can be
entered as a command snippet in a Command tree object under a Structural analysis system in Workbench.

15.3.1. Overview of Mechanical APDL Commands Used for Soil-Pile Interaction

Analysis
There are two sets of commands. The first set invokes an automatic pile analysis procedure while the
other is a set of fundamental commands that allow more user control for the pile analysis settings.
For automatic pile analysis, the commands available are:
PILEGEN

To automatically generate pile data.

PILECALC

To carry out soil-pile analysis and import results to Mechanical APDL.

For general pile analysis, the commands available are:

PILEDISPSET

To set up pile cap displacement.

PILELOAD

To import pile load.

PILEMASS

To import pile mass.

PILERUN

To execute a pile run.

PILESEL

To select a set of pile elements.

PILESTIF

To import pile stiffness.

15.3.2. Mechanical APDL Component System Example

This example shows the Mechanical APDL input file used for a single pile analysis.
/PREP7
/units,bft
! KEYOPT(3)=4, INPUT DATA AS 12 X 12 STIFFNESS MATRIX
ET,1,MATRIX27,,,4,1
! TABLE 1 REAL CONSTANTS
R,1
N,1
!
N,2,0,0,-123.3
!
E,1,2
!
OUTPR,ALL,1
!
! Pile material data
MP,EX,1,4.32E6
MP,DENS,1,1.52E-2
MP,PRXY,,0.3

FOR MATRIX27 STIFFNESS MATRIX

Pile cap
Pile tip
STIFFNESS MATRIX ELEMENT
PRINT ALL ITEMS

! Pile section data

SECTYPE,11,pipe,,pipe11
SECDATA,3.0,0.17
ACEL,0.0,0.0,32.2
! Body force required
PILEGEN,,1,11,1
! Generate pile

338

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Performing a Soil-Pile Interaction Analysis

FINISH
/SOLU
! AXIAL 0 to 600. LATERAL 25. SELF WEIGHT IN 6 STEPS
tinc=0.5
t1=tinc
t2=3.0
*DO,tim,t1,t2,tinc
time,tim
fxt=25
fzt=-200*tim
FDELE,ALL
F,1,FX,fxt
F,1,FZ,fzt
PILECALC
SOLVE
*ENDDO

Command PILECALC above can be replaced by the following data:

PILEDISPSET,1
PILERUN,1
PILESTIF,1
PILELOAD,1

!
!
!
!

Set pile cap displacements

Run pile analysis
Import pile stiffness to element 1
Import pile load to cap enode 1

15.3.3. Static Structural Component System Example

To use soil-pile interaction with a Static Structural system in Workbench, a command snippet object
must be added to the tree under the Static Structural branch. The command usage follows the same
method as in the Mechanical APDL component system (in this example, it only shows it being solved
for the first iteration; multiple iterations should be used to enable convergence of the soil-pile interaction
equations).
/PREP7
*get,ar10,etyp,,num,max
! KEYOPT(3)=4, INPUT DATA AS 12 X 12 STIFFNESS MATRIX
ET,ar10+1,MATRIX27,,,4,1
! Set the type that the generated elements are going to use
type,ar10+1
! Set the material that the generated elements are going to use
mat,ar10+1
! Pile material data (steel, in Metric Units)
MP,EX,ar10+1,2.1e11
MP,DENS,ar10+1,7850
MP,PRXY,ar10+1,0.3

! Pile section data

SECTYPE,ar10+1,pipe,,PileSec1
SECDATA,arg2,arg3 !Arg2 is the diameter, Arg3 is the thickness
*DO,PileNo,1,arg1,1 !Arg1 is the number of piles
R,ar10+PileNo
real,ar10+PileNo
nsel,s,node,,Pile%PileNo%
ar2%PileNo%=ndnext(0)
!Arg4 is the pile length
N,,NX(ar2%PileNo%),NY(ar2%PileNo%),NZ(ar2%PileNo%)-arg4
*get,ar3%PileNo%,node,,num,maxd
E,ar2%PileNo%,ar3%PileNo%
*ENDDO

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

339

Soil-Pile-Structure Analysis

!IF RAKED PILES OR MORE CONTROL IS REQUIRED, REPLACE THE ABOVE DO

! LOOP WITH THE FOLLOWING FOR EACH PILE
! Create table of real constant for MATRIX27 stiffness matrix
!R,ar10+1
! Set the real constant number for the element
!real,ar10+1
! This is the named selection that contains the pile node.
!nsel,s,node,,Pile1
! Get the next node number
!ar20=ndnext(0)
! Add pile tip node, here is where you'd change your bottom node
! position,
! relative to the node above.
!N,,NX(ar20),NY(ar20),NZ(ar20)
! Get the next element number
!*get,ar50,node,,num,maxd
! Add stiffness matrix element
!E,ar20,ar50
alls
!OUTPR,ALL,1

! Print all items, uncomment if required

PILEGEN,,ar10+1,ar10+1,1
FINISH

! Generate piles

/SOLU
PILECALC

Using the appropriate geometry, create a named selection that contains a single vertex locating the
pile cap; name the selection PileX, where X is the pile number (starting at 1).
Apply Forces and Standard Earth Gravity as required.

In the commands object, set input arguments as indicated below: the diameter, thickness, and length.
Length is the distance to the pile tip and will be applied to all piles; the length is vertically downwards
(Z direction) from the pile cap.
ARG1 Number of Piles (max 9 without Command Snippet customization)

340

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Soil-Pile-Structure Results
ARG2 Pile Diameter
ARG3 Pile Thickness
ARG4 Pile Length
In the analysis settings change the Solver Controls, Weak Springs to Off.

15.4. Soil-Pile-Structure Results

Soil-Pile-Structure interaction analysis results can also be obtained using a plugin to Excel. See Accessing
Results using Excel in the Mechanical BEAMST User Manual for installation requirements and how to
obtain element and equation based results using the plugin.

Equation Results
Available equation results types are:
DISPLACEMENT
REACTION
Valid associated result components are numbers 1 to 6, being X, Y, Z, RX, RY, and RZ respectively.

Element results
The available element results are listed below. Each surface is the subsection, and when the results
originate from a Splinter analysis initialized using the PILEGEN command, the element number is always
1 and the Structure used in forming the model name is the pile name entered via the StruPile argument. Components marked * are stored as character strings:
Type:

PILE DISPLACEMENT
PILE REACTION

RESULTS COMPONENT

DESCRIPTION

SEC.POSN

Position down pile

X value

Y value

Z value

RX

RX value

RY

RY value

RZ

RZ value

Type:

PILE FORCE/MOMENT

RESULTS COMPONENT

DESCRIPTION

SEC.POSN

FXX

X force (axial)
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

341

Soil-Pile-Structure Analysis
RESULTS COMPONENT

DESCRIPTION

FQY

Y shear force

FQZ

Z shear force

TXX

Torsional moment

MYY

Y bending moment

MZZ

Z bending moment

Type:

PILE STRESS

RESULTS COMPONENT

DESCRIPTION

SEC.POSN

Position down pile

SAX

Axial stress

SB

Bending stress

SXX.MAX

Maximum axial stress on section

Type:

API WSD PILEED20 UC

API LRFDPILEED1 UC
ISO PILEED1 UC

RESULTS COMPONENT

DESCRIPTION

SEC.POSN

Position down pile

UC.AX+BN

Combined axial + bending UC

15.5. References
The following references are cited in this chapter:
1. Lymon C. Reese, William R. Cox and Francis D. Koop, Analysis of Laterally Loaded Piles in Sand, Paper
No 2080, Offshore Technology Conference, 1974.
2.

Hudson Mattock, Correlations for Design of Laterally Loaded Piles in Soft Clay, Paper No 1024 Offshore
Technology Conference, 1970.

3. American Petroleum Institute, Recommended Practice for Planning, Designing and Constructing Fixed
Offshore Platforms, 15th Edition ,1984.
4. V. N. Vijayvergiya, Load-Movement Characteristics of Piles, PORTS 77.

342

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 16: Coupling to External Aeroelastic Analysis of Wind

Turbines
This chapter describes the facilities available within Mechanical APDL that enable it to perform integrated
wind and wave load analyses for offshore wind turbines with a specialized wind loading software
package. The purpose of using Mechanical APDL is to provide a realistic foundation model that can
accurately model the foundation structural behavior under the influence of wave loading. Note that
foundation here refers to the whole substructure that is under the influence of wave loading, i.e. from
the sea surface down.
Two different methods of analysis to support a coupled aeroelastic-structural analysis can be used with
Mechanical APDL to enable the design of the structures upon which wind turbines are positioned. Each
method has advantages and disadvantages depending upon the data and resources available. The sequentially coupled solution is found in the following section and uses standard Mechanical APDL. There
is an example of implementing a fully coupled solution that can be found in Fully Coupled Wind Turbine
Example in Mechanical APDL in the Mechanical APDL Programmer's Reference. This requires programmer
knowledge, software compilation tools, and customization to enable the coupling to an aeroelastic
analysis.

16.1. Sequential Coupled Wind Turbine Solution in Mechanical APDL

In the sequential wind coupling method, the aeroelastic analysis is performed by the aeroelastic code
with the effects of the supporting structure incorporated as a superelement to the solution. Mechanical
APDL provides the supporting structure substructure matrices and loading data that are required as
input to the aeroelastic code. After the aeroelastic analysis, the results can be fed back to Mechanical
APDL to recover the element forces inside the supporting structure.

16.1.1. Procedure for a Sequentially Coupled Wind Turbine Analysis

The procedure for carrying out an integrated wind and wave load analysis is described as follows:
1. The wind turbine supporting structure is modelled in Mechanical APDL. A substructure model is
created with the top node (i.e. the connection point between the wind turbine and the supporting
structure) set as the master node. This master node must have 6 freedoms: UX, UY, UZ, ROTX, ROTY,
and ROTZ. A substructure generation run is performed with the supporting structure model subjected
to ocean wave and other external loadings. The solution times for this run should tie in with the
times of the solution that will be attempted in the following aeroelastic run.
2. The command OUTAERO can be called after the solution from 1 is obtained to produce the generalized mass, damping, and stiffness matrices of the supporting structure, together with a time series
of the generalized foundation external loading (due to wave loading etc.). The generalized matrices
are written to 3 separate files (mass, damping, and stiffness) and the generalized load time series is
written to another file. See Output from the OUTAERO Command (p. 344).
3. An aeroelastic solution is then carried out with the foundation effects included through utilizing
the generalized matrices and loading vector derived above. The forces and/or displacements at the
supporting structure top node at each solution time are written to a file.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

343

Coupling to External Aeroelastic Analysis of Wind Turbines

4. Another Mechanical APDL run is performed to recover the member forces in the foundation structure
by applying the supporting structure top node force or displacement time series obtained from the
aeroelastic analysis together with the foundation external loading as specified in step 1. The analysis
can be carried out statically or dynamically.

Note
The following points should be noted for the sequential solution approach:
The generalized matrices (stiffness, etc.) are computed based on the initial undeformed geometry
and assuming small displacement, linear behavior. It is thus implicitly assumed that the
foundation is linear with small deformations throughout the entire solution.
Likewise, when computing the generalized foundation external load time history in the substructure generation pass, it is assumed that the structural displacement and velocity are zero
since such information is not available when the loading is generated.
The hydrodynamic mass for the supporting structure is formed based on the water elevation
at the first time at which the solution is attempted.
If the supporting structure internal forces are recovered statically in step 4, the dynamic forces
(e.g. inertial force) in the foundation will be ignored. The dynamic effects can be accounted
for by running this step as a transient job. However, it should be noted that the following
points may affect the accuracy of the solution:
The generalized mass used in the aeroelastic analysis is only an approximation to the true
mass matrix (static reduction is exact but not dynamic).
Different time integration schemes may be adopted by aeroelastic code and Mechanical
APDL. Hence, while the displacement time histories are identical in both runs (for the prescribed displacement case), it may not be the case for the velocity and acceleration time
histories.
There should be little difference between applying forces or displacements to recover the
foundation forces. The two methods should yield identical results in a linear static analysis.
This approach should be much more efficient than the fully coupled approach as there is no
need to keep both the aeroelastic code and Mechanical APDL running simultaneously and keep
exchanging information every time step.

16.1.2. Output from the OUTAERO Command

Four formatted ASCII files will be generated by specifying the OUTAERO command macro. These are:
Jobname.gnm - generalized mass matrix file
Jobname.gnc - generalized damping matrix file
Jobname.gnk - generalized stiffness matrix file
Jobname.gnf - generalized external force time series file
where Jobname is the current job name.
344

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Sequential Coupled Wind Turbine Solution in Mechanical APDL

The generalized mass, damping, and stiffness matrices are formed based on the information at the very
first load step and are written to 3 separate files. These are formatted ASCII files with the full 6 x 6
matrix included. The files are written with the following format:
ngenfr (1x, I6)
(val(i,1), i = 1,ngenfr) (6(1x, E12.5))
(val(i,2), i = 1,ngenfr) (6(1x, E12.5))
.
.
(val(i,ngenfr), i = 1,ngenfr) (6(1x, E12.5))

where ngenfr is the number of generalized freedoms, which is always 6 at present, and val is the
generalized matrix.
The row and column order in the generalized matrix corresponds to the order UX, UY, UZ, ROTX, ROTY,
ROTZ.
A time series of the generalized foundation external loading vector is written to another file. At each
solution time, the time (t) and the associated generalized load vector (f ) will be output to this file. The
force file has the following format:
ngenfr (1x, I6)
t1 (1x, E12.5)
(f1(i), i = 1,ngenfr) (6(1x, E12.5))
t2 (1x, E12.5)
(f2(i), i = 1,ngenfr) (6(1x, E12.5))
.
.
tn (1x, E12.5)
(fn(i), i = 1,ngenfr) (6(1x, E12.5))

The load values are ordered in the same way as the generalized matrices.

16.1.3. Example Substructured Analysis to Write Out Aeroelastic Analysis Input

Data
This is an example of the sequential aeroelastic analysis process. This first analysis is run to create the
matrices/loading. These are generated by the OUTAERO macro near the end of the analysis.
The master node is set as node 9, which will be the interface point to the aeroelastic structure.
/verify,airysublarge
/FILNAME,airysublarge
/prep7
/TITLE,airysublarge, WAVE ON MONOPILE
/com **************************************************************************
/com Substructure with Airy wave
/com use time to determine phase at each step
/com includes current
/com CREATED 08/03/11
/com **************************************************************************
antype,substr
seopt,monopile,3,1
nlgeom,off
et,1,pipe288
keyopt,1,3,3
keyopt,1,12,1
type,1
mat,1
! define pipe section
secnum,1
sectype,1,pipe
secdata,1.0,0.1
! define ocean
matwat=2
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

345

Coupling to External Aeroelastic Analysis of Wind Turbines

idwat=2
idcur=3
idwav=4
depth=30.0
offset = 1.5
socean,idwat
! Ocean basic
octype,idwat,basic,ocean1,idcur,idwav
ocdata,depth,matwat
octable,,0.7,0.7,,2.0
mp,dens,matwat,1000.0
! Ocean current
octype,idcur,curr
octable,0.0,1.0
octable,-depth,0.0
! define geometry of vertical tube
n,
1,
offset,
0.0,
-30.0000
n,
2,
offset,
0.0,
-25.0000
n,
3,
offset,
0.0,
-20.0000
n,
4,
offset,
0.0,
-15.0000
n,
5,
offset,
0.0,
-10.0000
n,
6,
offset,
0.0,
-5.0000
n,
7,
offset,
0.0,
0.0000
n,
8,
offset,
0.0,
5.0000
n,
9,
offset,
0.0,
10.0000
n,
900,
offset, -20.0000, 0.0000
en, 1,
1, 3, 900
en, 3,
3, 5, 900
en, 5,
5, 7, 900
en, 7,
7, 9, 900
MP,EX, 1,2.0e11
MP,PRXY,1,0.3
MP,ALPX,1,0.0
MP,DENS,1,7850.0
! damping factors
alphad,0.1
betad,0.01
! suppressions
d,1,all
! master freedoms
m,9,all
finish
! increase limit of time values to 5000
/config,numsublv,5000
/SOLU
acel,0,0,9.81
wper=10.0
phs=0.0
! Ocean wave
!! Airy wave
octype,idwav,wave
ocdata,0,0.0,0,1,1
octable,2.0,wper,phs
tm_1=1.0e-8
tm_2=1000.0
tm_inc=0.2
*do,tm,tm_1,tm_2,tm_inc
time,tm
solve
*enddo
! print substructure matrices
! outaero,'monopile',tm_1,tm_inc ! This version uses the time defined by tm_1 & tm_inc
outaero,'monopile' ! This version reads the time off the .sub file
finish

At this point the aeroelastic analysis can be run, using the output from the above analysis. Once complete,
a second Mechanical APDL analysis is run with a time series of forces and/or displacements at the interface node. These need to be converted from the aeroelastic output to Mechanical APDL compatible
output by the user (for example, using Excel) or the aeroelastic analysis program. Any ocean loading

346

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Sequential Coupled Wind Turbine Solution in Mechanical APDL

or extra loading included in the substructured analysis should also be applied in the subsequent analysis.

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

347

348

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Chapter 17: Applying Ocean Loading from a Hydrodynamic Analysis

Pressures and motion effects of a rigid body modeled by hydrodynamic analysis can be imported for
further analysis using ocean commands. For this type of analysis, the wave type in the ocean wave
OCDATA command should be set to diffracted wave (KWAVE = 8). The hydrodynamic data is read in
via the OCREAD command.
This topic does not describe wave-theory-derived ocean loading (implemented via KWAVE = 0 through
7 on the OCDATA command).
The following topics related to using hydrodynamic analysis data to apply ocean loading are available:
17.1. How Hydrodynamic Analysis Data Is Used
17.2. Hydrodynamic Load Transfer with Forward Speed
17.3. Hydrodynamic Data File Format
17.4. Example Analysis Using Results from a Hydrodynamic Diffraction Analysis

17.1. How Hydrodynamic Analysis Data Is Used

Pressures and motions due to the waves, current, and forward speed at a given phase are extracted
from a hydrodynamic data file. The hydrodynamic data file information is translated either into surface
element pressures or into line element pressures based on Morisons equation.
The hydrodynamic data (.ahd) file must be prepared in a specific format for it to be read during the
ocean loading procedure. If the Hydrodynamic Diffraction System (AQWA) has been used for the hydrodynamic diffraction/radiation analysis, a utility program is available to translate the results database
into this format automatically.
For ocean loading on pipes, the loading is applied by associating the section of PIPE288/PIPE289 elements
to an ocean load (SOCEAN). For ocean loading on shells or solids, the hydrodynamic loading is applied
via SURF154 elements with KEYOPT(8) = OCID, where OCID is the ocean load ID number.
For loads applied to shells or solids only, the hydrodynamic load components are included. Hydrostatic
loading can be included on SURF154 using face 4 with PL = -(gravity acceleration) * (water density) and
by setting KEYOPT(6) = 1. For loads applied to pipe elements, hydrostatic loading is included automatically.
It is not necessary for the hydrodynamic and structural meshes to be identical, and the loading is
mapped automatically from one to the other; however, it is assumed that the hydrodynamic axis system
is identical to that used for the structural analysis. Small displacements are assumed in the load-mapping
process, and mapping may not be accurate if this condition is not met.
The .ahd file contains data indicating which side of a panel is the wetted side. The program uses the
data to adjust which direction to apply the hydrodynamic pressure on the SURF154 elements.
Each structural integration point is located in its nearest hydrodynamic panel. The program then calculates
the corner pressures of the panel based on a weighted averaging of that panel and its neighboring
panels. Finally, using a bilinear fit across the panel, the program calculates the hydrodynamic pressure
at the structural integration point.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

349

Applying Ocean Loading from a Hydrodynamic Analysis

17.2. Hydrodynamic Load Transfer with Forward Speed

Ocean loading analysis does not allow input of forward speed explicitly. To account for forward speed,
it is necessary to specify an opposite current. If forward speed and a current are required, the specified
current must represent the resultant of the two effects. This current is used only for the calculation of
Morison loads following the diffraction calculation.
If the panel pressures and motions are obtained from AQWA, it is also possible to define a forward
speed in the diffraction calculation, causing the hydrodynamic coefficients to be corrected for the effect
of forward speed. The following restrictions apply:
Forward speed can be along the global X axis only.
Only one wave direction can be specified in a diffraction calculation with forward speed.
It is your responsibility to ensure that forward speed in the diffraction calculation and current in the
load transfer are consistent.

17.3. Hydrodynamic Data File Format

The hydrodynamic data file comprises a number of sections which represent all of the required input
to the ocean wave command:
17.3.1. Comment (Optional)
17.3.2. General Model Data
17.3.3. Hydrodynamic Surface Geometry
17.3.4. Wave Periods
17.3.5. Wave Directions
17.3.6. Panel Pressures
17.3.7. Morison Element Hydrodynamic Definition
17.3.8. Morison Element Wave Kinematics Definition
17.3.9. RAO Definition
17.3.10. Mass Properties

17.3.1. Comment (Optional)

The comment section contains descriptive text for the model.
* text

where
*

keyword to signify that this is a comment line

text

description of the model

17.3.2. General Model Data

This section defines some general data for the model.
GENE
depth density gravity symxy

where
GENE

350

keyword to denote the start of the general model data

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Hydrodynamic Data File Format

depth

water depth

density

water density

gravity

acceleration of gravity

symxy

symmetry integer
0: no symmetry
1: symmetry about x axis
2: symmetry about y axis
3:symmetry about both x and y axes

Example 17.1: General Model Data

GENE
2.500000E+02

1.025000E+03

9.800000E+00

17.3.3. Hydrodynamic Surface Geometry

The hydrodynamic surface geometry provides information related to the panel definition of the model.
Allowance can be made for symmetric models; that is, only one half of the model must be defined for
one plane of symmetry, and only a quarter of the model must be defined for double symmetry.
HYDR
npan
x1(1) y1(1) z1(1) x2(1) y2(1) z2(1) x3(1) y3(1) z3(1) x4(1) y4(1) z4(1)
.
.
x1(npan) y1(npan) z1(npan) x2(1) x4(npan) y4(npan) z4(npan)

where
HYDR

keyword to denote start of hydrodynamic panel description

npan

number of panels to be defined

xk(j)

X coordinate for panel j, node k

yk(j)

Y coordinate for panel j, node k

zk(j)

Z coordinate for panel j, node k

The nodes must be arranged in counterclockwise order when viewing the panel from the wetted side.
To define a triangular panel, specify the same coordinates for nodes 3 and 4.
Example 17.2: Hydrodynamic Surface Geometry
HYDR
12
15.0
15.0
15.0
15.0
7.5
7.5
7.5
7.5
15.0
15.0
15.0
15.0

-15.0
-15.0
-15.0
-7.5
-15.0
-15.0
-15.0
-7.5
-15.0
-15.0
-7.5
-7.5

0.0
-7.5
-15.0
-15.0
0.0
-7.5
-15.0
-15.0
0.0
-7.5
0.0
-7.5

7.5
7.5
7.5
7.5
0.0
0.0
0.0
0.0
15.0
15.0
15.0
15.0

-15.0
-15.0
-15.0
-7.5
-15.0
-15.0
-15.0
-7.5
-15.0
-15.0
-7.5
-7.5

0.0
-7.5
-15.0
-15.0
0.0
-7.5
-15.0
-15.0
-7.5
-15.0
-7.5
-15.0

7.5
7.5
7.5
7.5
0.0
0.0
0.0
0.0
15.0
15.0
15.0
15.0

-15.0
-15.0
-7.5
0.0
-15.0
-15.0
-7.5
0.0
-7.5
-7.5
0.0
0.0

-7.5
-15.0
-15.0
-15.0
-7.5
-15.0
-15.0
-15.0
-7.5
-15.0
-7.5
-15.0

15.0
15.0
15.0
15.0
7.5
7.5
7.5
7.5
15.0
15.0
15.0
15.0

-15.0
-15.0
-7.5
0.0
-15.0
-15.0
-7.5
0.0
-7.5
-7.5
0.0
0.0

-7.5
-15.0
-15.0
-15.0
-7.5
-15.0
-15.0
-15.0
0.0
-7.5
0.0
-7.5

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

351

Applying Ocean Loading from a Hydrodynamic Analysis

17.3.4. Wave Periods

This section defines the wave periods where hydrodynamic pressures have been computed.
PERD
nperd
period(1) period(2) ... period(i)
period(i+1) period(i+2) ... period(nperd)

where
PERD

keyword to denote start of wave period data

nperd

number of wave periods

period(i)

the ith wave period (s)

The data may be specified in one or more lines until all the periods are entered.
Example 17.3: Wave Periods
perd
1
18.0

17.3.5. Wave Directions

This section defines the wave directions where hydrodynamic pressures have been computed.
DIRN
ndirn
heading(1) heading(2) ... heading(i)
heading(i+1) heading(i+2) ... heading(ndirn)

where
DIRN

keyword to denote start of wave direction data

ndirn

number of wave directions

heading(i)

the ith wave direction (in degrees, between -180 and +180)

The data may be specified in one or more lines until all the directions are entered.
Example 17.4: Wave Directions
dirn
2
0.0 90.0

17.3.6. Panel Pressures

This section defines the wave directions where hydrodynamic pressures have been computed.
PRES
iperiod iheading region panel magnitude phase

where
PRES

keyword to denote the start of hydrodynamic pressure values

iperiod

period number (position in the period list)

352

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Hydrodynamic Data File Format

iheading

wave direction number (position in the direction list)

region

If two planes of symmetry (symxy set to 3)

region 1 corresponds to modelled elements
region 2 corresponds to reflection in x-z plane
region 3 corresponds to reflection in y-z plane
region 4 corresponds to reflection in x-z and y-z plans
If one plane of symmetry (symxy set to 1 or 2)
region 1 corresponds to modelled elements
region 2 corresponds to reflected elements

panel

panel number (must be between 1 and npan)

magnitude

pressure amplitude

phase

associated phase angle in degrees (with respect to the wave at the center of gravity of the
vessel)

The associated phase angle (phase) determines the horizontal location of the loading. The magnitude
and phase together define the shape of the harmonic loading curve.
Pressure is positive when acting from the wetted surface towards the inner surface.
If no input pressure is given to a panel at a particular period and direction, the pressure on the panel
is assumed to be zero. If more than one set of pressures are defined, their effects are accumulative; that
is, the real and imaginary parts of each set are summed together.
Example 17.5: Panel Pressures
PRES
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
1

1
2
3
4
5
6

25.0
75.0
100.0
100.0
25.0
75.0

0.0
0.0
0.0
0.0
0.0
0.0

17.3.7. Morison Element Hydrodynamic Definition

The data in this section describes the line elements used to provide slender body loading.
MORI
nmori
x1(1) y1(1) z1(1) x2(1) y2(1) z2(1)
.
.
x1(nmori) y1(nmori) z1(nmori) x2(nmori) y2(nmori) z2(nmori)

where
MORI

keyword to denote start of Morison hydrodynamic element description

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

353

Applying Ocean Loading from a Hydrodynamic Analysis

Nmori

number of Morison elements defined

xk(j)

X coordinate for line element j, node k

yk(j)

Y coordinate for line element j, node k

zk(j)

Z coordinate for line element j, node k

Example 17.6: Morison Element Hydrodynamic Definition

MORI
1
10 0 0 20 0 0

17.3.8. Morison Element Wave Kinematics Definition

The data in this section defines the wave kinematics at the nodes of the Morison elements defined
above.
WKIN
iperiod iheading node element pmag pphs zsmag zsphs
vxmag vxphs vymag vyphs vzmag vzphs
axmag axphs aymag ayphs azmag azphs

where
WKIN

keyword to denote start of Morison element wave kinematics definition

iperiod

wave period number (position in period list)

iheading

wave direction number (position in direction list)

node

local node number on Morison element (1 or 2)

element

element number referencing the Morison element list

pmag

hydrodynamic pressure amplitude

pphs

associated phase angle (degrees)

zsmag

surface elevation amplitude

zsphs

associated phase angle (degrees)

vxmag

x velocity amplitude

vxphs

associated phase angle (degrees)

vymag

y velocity amplitude

vyphs

associated phase angle (degrees)

vzmag

z velocity amplitude

vzphs

associated phase angle (degrees)

axmag

x acceleration amplitude

axphs

associated phase angle (degrees)

aymag

y acceleration amplitude

ayphs

associated phase angle (degrees)

azmag

z acceleration amplitude

azphs

associated phase angle (degrees)

Note that the data is defined for X, Y and Z in the global directions.

354

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Hydrodynamic Data File Format

If no input is given to a Morison element at a particular period and direction, the kinematics data on
this element are assumed to be zero.
Example 17.7: Morison Element Wave Kinematics Definition
WKIN
1 1 1 134 0.1 15 1.2 15
2.2 15 2.2 15 2.3 15
0.2 15 1.5 15 4.3 15

17.3.9. RAO Definition

This section data defines the RAO data. The first line defines the RAO reference position, followed by
blocks of data defining RAOs in various periods and directions.
RAOS
xr yr zr
iperiod iheading xmag xphs ymag yphs zmag zphs
rxmag rxphs rymag ryphs rzmag rzphs

where
RAOS

keyword to denote start of RAO definition

xr yr zr

X, Y and Z coordinate of the RAO reference position

iperiod

wave period number (position in period list)

iheading

wave direction number (position in direction list)

xmag

x RAO amplitude

xphs

associated phase angle (degrees)

ymag

y RAO amplitude

yphs

associated phase angle (degrees)

zmag

z RAO amplitude

zphs

associated phase angle (degrees)

rxmag

rx RAO amplitudee (degrees)

rxphs

associated phase angle (degrees)

rymag

ry RAO amplitude (degrees)

ryphs

associated phase angle (degrees)

rzmag

rz RAO amplitude (degrees)

rzphs

associated phase angle (degrees)

Example 17.8: RAO Definition

RAOS
0.000000E+00 0.000000E+00 -5.000000E+00
1 1 8.972334E-01 8.999010E+01 6.241512E-08 -1.799247E+02
7.862426E-09 0.000000E+00 7.524462E-01 -9.000951E+01
1 2 6.357081E-01 8.999005E+01 6.357080E-01 8.999005E+01
5.293267E-01 8.999037E+01 5.293267E-01 -9.000961E+01

1.042555E+00
2.027182E-07
1.042556E+00
1.605987E-07

4.346394E-02
0.000000E+00
4.340762E-02
0.000000E+00

17.3.10. Mass Properties

This section defines the mass properties of the hydrodynamic model but is currently unused.
MASS
xcg ycg zcg

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

355

Applying Ocean Loading from a Hydrodynamic Analysis

Mass(1,1) mass(1,2) .. mass(1,6)
.
.
Mass(6,1) mass(6,2) .. mass(6,6)

where
MASS

keyword to denote start of mass information

xcg ycg zcg

X, Y and Z coordinate of CoG

mass(j,k)

mass matrix term

Example 17.9: Mass Properties

MASS
0.000000E+00
1.383750E+07
0.000000E+00
0.000000E+00
0.000000E+00
0.000000E+00
0.000000E+00

0.000000E+00 -5.000000E+00
0.000000E+00 0.000000E+00
1.383750E+07 0.000000E+00
0.000000E+00 1.383750E+07
0.000000E+00 0.000000E+00
0.000000E+00 0.000000E+00
0.000000E+00 0.000000E+00

0.000000E+00
0.000000E+00
0.000000E+00
7.780000E+08
0.000000E+00
0.000000E+00

0.000000E+00
0.000000E+00
0.000000E+00
0.000000E+00
7.780000E+08
0.000000E+00

0.000000E+00
0.000000E+00
0.000000E+00
0.000000E+00
0.000000E+00
7.780000E+08

17.4. Example Analysis Using Results from a Hydrodynamic Diffraction

Analysis
In this example, a 30m x 30m x 22.5m box with 7.5m above the water level (z = 0m) was modeled in
a hydrodynamic diffraction analysis.
The following conditions exist:
The diffracting elements are 3m x 3m.
Waves are defined in eight directions and four frequencies (0.031, 0.5, 0.97, and 1.43 rad/s).
Forward speed and hydrostatic pressures are not included.
The pressure results from the hydrodynamic diffraction analysis are 45, 1.43 rad / s, t / T = 0, as shown
in the following figure:

356

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Analysis Using Results from a Hydrodynamic Diffraction Analysis

Figure 17.1: Hydrodynamic Diffraction Analysis Pressure Results

An .ahd file named analysis_1, which contains the data from the hydrodynamic diffraction analysis,
was generated via the AQWA product's AQWA2NEUT utility, as follows:
aqwa2neut analysis 1

The diffracted wave loading data in the analysis_1.ahd file is imported (OCREAD) and applied to
the structural model for further analysis, as shown in the following example input:
/verify,difwaveeg
/FILNAME,difwaveg
/prep7
/TITLE,difwaveeg, EXAMPLE FOR AQWA LOADING
/com **************************************************************************
/com SHELL MODEL OF A HOLLOW BOX TUBE
/com CREATED 20/05/10
/com **************************************************************************
antype,static
nlgeom,off
! define shell section
sectype,11,shell
secdata,0.1
! define ocean
matwat=2
idwat=2
idcur=3
idwav=4
depth=250.0
! define geometry
/prep7
!wpoff,,,-15

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

357

Applying Ocean Loading from a Hydrodynamic Analysis

k,,-15,-15,-15
k,,15,-15,-15
k,,15,15,-15
k,,-15,15,-15
l,1,2
l,2,3
l,3,4
l,4,1
k,,-15,-15,0
l,1,5
k,,-15,-15,7.5
l,5,6
adrag,1,2,3,4,,,5,6
a,1,2,3,4
! Mesh
et,1,181
type,1
mat,1
secnum,11
esize,1
amesh,all
et,2,154
keyopt,2,4,1
keyopt,2,8,idwat
r,2
rmore,0.0001
type,2
mat,1
real,2
esurf
! steel material properties
MP,EX, 1,2.1e11
MP,PRXY,1,0.3
MP,ALPX,1,0.0
MP,DENS,1,7800.0
! water density
mp,dens,matwat,1025.0
! supports
d, 1,ux
d, 2,ux
d,977,ux
d,497,ux
d, 1,uy
d, 2,uy
d,977,uy
d,497,uy
d, 1,uz
d, 2,uz
d,977,uz
d,497,uz
finish
!
/SOLU
! CASE 4 H=1 T=4.384199 PH=0 D=45
acel,0,0,9.8
! ocean
octype,idwat,basic,ocean1,,idwav
ocdata,depth,matwat
!! diffracted wave
octype,idwav,wave
ocdata,8,45.0
!octable,1.0,4.384199,0.0
octable,1.0,200,0.0
!octable,1.0,4.384199,0.0
!
!! ocread to read in external data file
ocread,analysis_1,ahd
!
! list all
slist,all
oclist,all
!! apply inertia relief

358

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

Example Analysis Using Results from a Hydrodynamic Diffraction Analysis

!! irlf,1
NSUBST,1,1
TIME,1.0
OUTPR,all,1
OUTRES,all,1
solve
finish
/post1
etable,x,smisc,13
pletab,x,avg

! 1 SUBSTEPS

The following figure shows the analysis results using the pressures from the diffracted wave:
Figure 17.2: Analysis Results Using Pressure Data from a Hydrodynamic Diffraction Analysis

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

359

360

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

transferring from ADAMS to ANSYS, 289

methodology employed by, 293
modal neutral file, 286, 293
model: building, 284
overview of, 283
rigid-body dynamics, 283
verifying results following export to ADAMS, 288
weak springs, 291, 293-294

Index
Symbols
.mnf suffix, 293
/CYCEXPAND command
limitations, 172
2-D to 3-D analysis
benefits, 135
initial-state-based analysis, 141
introduction to, 135
key commands, 138
map solution variables, 141
performing, 139
requirements, 135
restrictions, 141
selecting a region for remeshing, 140
starting, 140
substep at which to initiate, 139
supported analyses, 135
understanding the process, 137

A
ADAMS (see ANSYS-ADAMS Interface)
ADAMS.MAC macro, 293
adaptive meshing
activating custom subroutines, 88
ADAPT macro: customizing (via UADAPT.MAC), 89
custom loads (via ADAPTBC.MAC), 88
custom meshing (via ADAPTMSH.MAC), 87
custom solution (via ADAPTSOL.MAC), 88
element types for, 85
example problems: finding, 90
hints, 89
modifying the basic process, 86
overview of, 85
prerequisites for, 85
process involved, 85
selective adaptivity, 86
user subroutines: employing, 87
aeroelastic analysis, 343
ANSYS-ADAMS Interface
ADAMS simulation: running, 289
ADAMS.MAC macro, 293
batch mode
exporting to ADAMS, 288
importing loads, 292
example analysis, 294
exporting to ADAMS, 286
interface points: modeling, 285
introduction, 283
loads

B
birth and death of elements
applying loads and solving, 301
building the model, 301
constraint equations, 300
controlled by analysis results, 303
example problems, 303
first load step: defining, 301
in linear models, 300
initial strain, 300-301
material properties: changing, 301
multiple load steps, 300
Newton-Raphson option, 300
nodal constraints and loads, 301
overview of, 299
reviewing results, 302
sample input, 302-303
status of an element: determining, 300
strain, 299
subsequent load steps: defining, 302
supported elements, 299
usage guidelines, 300
using in an analysis, 301

C
Campbell diagram, 199
CDB generic mesh file, 106
CMS (see component mode synthesis)
component mode synthesis
database: saving copy during generation pass, 261
definition of, 259
example analysis, 264
example: expanding a superelement in a transformed
location, 279
expanding superelements in transformed locations,
263
expansion pass, 263
generation pass, 261
introduction to, 259
lumped mass approximation, 261
master DOFs, 261
methods supported, 259
mode shapes: plotting or printing, 264

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

361

Index
offsets: applying to node and element IDs, 263
process (passes) involved, 261
solvers and files used, 260
specifying CMS results files for postprocessing, 264
superelement: creating, 261
understanding, 259
use pass, 263
constraint equations
in a cyclic symmetry analysis, 151
Coriolis effect, 197
cut boundary displacement method (see submodeling)
cyclic symmetry analysis
ANCYC command, 169
automated, 144
availability, 143
basic sector, 145
buckling, 162
commands affected by non-cyclic loading, 155
constraint equations for non-cyclically symmetric
loading, 152
coupling and constraint equations (CEs), 151
CYCLIC command, 146
database: saving after solution, 165
definition of cyclic symmetry, 143
deleting a sector load, 152
Duplicate Sector solution method, 151
edge component pairs, 146
edge node patterns: identical vs. dissimilar, 149
example: buckling, 177
example: modal, 172
EXPAND command, 169
expanding the solution, 167
geometric nonlinearity, 160
graphical picking of non-cyclic boundary conditions,
155
harmonic index, 157
inconsistent constraints, 152
introduction to, 143
large-deflection prestressed modal, 160
limitations, 171
linear buckling, 162
meshing the cyclic sector, 149
modal, 157
model only one sector, 144
model verification during preprocessing, 150
model verification during solution, 165
modeling, 144
most common commands, 144
naming convention for low- and high-edge component pairs, 146
nodal diameter, 157
non-cyclic loading, 155

362

non-cyclically symmetric loading, 152

phase sweep of repeated eigenvector shapes, 169
plotting and listing non-cyclic BCs, 155
postprocessing, 166
prestressed modal, 159
real and imaginary solution components: understanding, 167
sector angle, 146
SECTOR primary variable, 152
solution architecture, 151
solving, 150
static (linear and large-deflection nonlinear), 155
stress-free modal, 158
traveling wave animation, 169
understanding, 143
unmatched nodes on edge-component pairs, 150
vt accelerator, 157, 160

D
damping
substructuring, 240
diffracted wave, 349

E
elements
birth and death of, 299
expansion pass
substructuring, 249

F
files
.mnf, 293
Jobname.LOOP, 3
Jobname.NODE, 223
Jobname.PDS, 3
probabilistic design analysis file, 9
sename.sub, 240
fixed-interface CMS method, 259
flexible body dynamics (see rotating structure analysis)
free-interface CMS method, 259

H
hydrodynamic analysis, 349

J
Jobname
in a submodeling analysis, 221
in a substructure analysis, 239

L
loads
substructuring, 240

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

M
macros
ADAMS.MAC, 293
material properties
changing to deactivate or reactivate elements, 301
meshing
adaptive, 85
modal neutral file, 293
model generation
adaptive meshing, 85

N
natural frequencies, 204
Newton-Raphson
birth and death analysis, 300
nonstandard uses of ANSYS, 307

O
ocean loading, 349
offsets
to node and element IDs in a CMS analysis, 263

P
PDS (see probabilistic design)
pile analysis, 315
probabilistic design
algorithms available, 24, 51
analysis file
creating, 9
specifying, 13
analysis flow, 3
analysis results: reviewing, 41
and traditional analysis methods: comparison of, 2
correlations between random variables, 20
cumulative distribution function: plotting, 42
definition of, 1
employing, 8
example analysis, 69
example problem, 10
histogram: plotting, 42
HTML report: generating, 67
inverse probabilities: requesting, 42
loops: executing, 25
Monte Carlo Simulation method, 51
parallel execution, 27
parameters: establishing, 12
PDS database
clearing, 69
description of, 3
saving, 68
postprocessing, 59
print correlation matrix, 43

Probabilistic Method Determination Wizard, 24

probability value: printing, 42
process involved, 8
random input variables
declaring, 14
visualizing, 20
random output parameters: specifying, 23
report options, 43
response surface analysis: sampling methods, 55
response surfaces: fitting and using, 38
restarting an analysis, 68
results data: reviewing, 41
results report: creating, 44
sample history: plotting, 42
scatter plot, 43
sensitivities, 43
serial execution, 27
statistics: viewing, 42
techniques available, 51
terminology
analysis file, 3
loop file, 3
looping, 3
random input variables (RVs), 3
random output parameters (RPs), 3
response surface set, 3
simulation, 3
simulation loop, 3
solution set, 3
trends: viewing, 43
understanding, 1
using, 8
variables: guidelines for defining, 44
when to use, 3
Probabilistic Design System (PDS) (see probabilistic
design)
Probabilistic Method Determination Wizard (see probabilistic design)

R
residual-flexible free-interface CMS method, 259
rezoning
balancing residual forces, 118
benefits, 93
contract boundaries, 116
convergence difficulties during mapping operation,
120
example, 124, 127
file structures for repeated rezonings, 121
forces and isolated applied displacements, 116
horizontal rezoning, 106
introduction to, 93

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.

363

Index
key commands, 101
limitations, 95
mapping variables, 118
mesh control during, 114
multiframe restart after, 120
multiple regions remeshed at the same substep,106
nodal temperatures, 116
postprocessing results, 121
pressure and contiguous displacements, 116
remeshing, 104
repeating as needed, 121
requirements, 96
restrictions, 123
selecting a region for remeshing, 103
starting, 103
substep at which to initiate, 102
supported analyses, 96
understanding the process, 98
using a generic (third-party) mesh, 106
using manual mesh splitting, 110
variable mapping, 118
rigid-body dynamics, 283
rotating structure analysis
Campbell diagram, 199
Campbell diagram analysis example, 205
choosing the appropriate reference frame option,
204
Coriolis analysis example, 207
flexible body dynamics, 202
harmonic analysis for unbalance forces example,210
introduction to, 197
natural frequencies, 204
rotating reference frame, 202
rotordynamics, 198
stationary reference frame, 198
understanding, 197
rotordynamics (see rotating structure analysis)

S
soil data description, 325
soil-pile-structure analysis, 315
soil data description, 325
solid modeling
adaptive meshing, 85
mesh density, 85
State-Space Matrices
Export of, 309
submodeling, 219
analyzing the submodel, 224
coarse model: creating and analyzing, 221
creating the submodel, 222
cut-boundary interpolation: performing, 223

364

distance between cut boundaries and stress concentration: verifying, 226

example input, 227
example problems, 235
introduction, 219
shell-to-solid, 233
understanding, 219
Submodeling
process involved, 220
substructuring
applying loads, 240
benefits of, 237
bottom-up, 252
damping, 240
example problems, 257
expansion pass, 249
generation pass, 238
introduction to, 237
nested superelements, 256
prestressed, 256
process (passes) involved, 237
sample input, 252
top-down, 253
use pass, 244
superelements
and substructuring, 237
expanding in transformed locations, 263
generating automatically, 255

T
traveling wave animation, 169

U
UPFs (see user-programmable features)
user-programmable features
available types, 306
overview of, 305
process involved, 305

V
Variational Technology
example problem
transmission line, 78
waveguide, 79
forced-frequency simulations: solving, 76
overview, 75

W
wind turbine analysis, 343
sequentially coupled, 343

Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information
of ANSYS, Inc. and its subsidiaries and affiliates.