System Identification Toolbox™ 7

User’s Guide

Lennart Ljung
How to Contact MathWorks

www.mathworks.com Web
comp.soft-sys.matlab Newsgroup
www.mathworks.com/contact_TS.html Technical Support

suggest@mathworks.com Product enhancement suggestions

bugs@mathworks.com Bug reports
doc@mathworks.com Documentation error reports
service@mathworks.com Order status, license renewals, passcodes
info@mathworks.com Sales, pricing, and general information

508-647-7000 (Phone)

508-647-7001 (Fax)

The MathWorks, Inc.

3 Apple Hill Drive
Natick, MA 01760-2098
For contact information about worldwide offices, see the MathWorks Web site.
System Identification Toolbox™ User’s Guide
© COPYRIGHT 1988–2010 by The MathWorks, Inc.
The software described in this document is furnished under a license agreement. The software may be used
or copied only under the terms of the license agreement. No part of this manual may be photocopied or
reproduced in any form without prior written consent from The MathWorks, Inc.
FEDERAL ACQUISITION: This provision applies to all acquisitions of the Program and Documentation
by, for, or through the federal government of the United States. By accepting delivery of the Program
or Documentation, the government hereby agrees that this software or documentation qualifies as
commercial computer software or commercial computer software documentation as such terms are used
or defined in FAR 12.212, DFARS Part 227.72, and DFARS 252.227-7014. Accordingly, the terms and
conditions of this Agreement and only those rights specified in this Agreement, shall pertain to and govern
the use, modification, reproduction, release, performance, display, and disclosure of the Program and
Documentation by the federal government (or other entity acquiring for or through the federal government)
and shall supersede any conflicting contractual terms or conditions. If this License fails to meet the
government’s needs or is inconsistent in any respect with federal procurement law, the government agrees
to return the Program and Documentation, unused, to The MathWorks, Inc.

Trademarks
MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See
www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand
names may be trademarks or registered trademarks of their respective holders.
Patents
MathWorks products are protected by one or more U.S. patents. Please see
www.mathworks.com/patents for more information.
Revision History
April 1988 First printing
July 1991 Second printing
May 1995 Third printing
November 2000 Fourth printing Revised for Version 5.0 (Release 12)
April 2001 Fifth printing
July 2002 Online only Revised for Version 5.0.2 (Release 13)
June 2004 Sixth printing Revised for Version 6.0.1 (Release 14)
March 2005 Online only Revised for Version 6.1.1 (Release 14SP2)
September 2005 Seventh printing Revised for Version 6.1.2 (Release 14SP3)
March 2006 Online only Revised for Version 6.1.3 (Release 2006a)
September 2006 Online only Revised for Version 6.2 (Release 2006b)
March 2007 Online only Revised for Version 7.0 (Release 2007a)
September 2007 Online only Revised for Version 7.1 (Release 2007b)
March 2008 Online only Revised for Version 7.2 (Release 2008a)
October 2008 Online only Revised for Version 7.2.1 (Release 2008b)
March 2009 Online only Revised for Version 7.3 (Release 2009a)
September 2009 Online only Revised for Version 7.3.1 (Release 2009b)
March 2010 Online only Revised for Version 7.4 (Release 2010a)
September 2010 Online only Revised for Version 7.4.1 (Release 2010b)
 About the Developers

About the Developers

System Identification Toolbox™ software is developed in association with the
following leading researchers in the system identification field:

Lennart Ljung. Professor Lennart Ljung is with the Department of

Electrical Engineering at Linköping University in Sweden. He is a recognized
leader in system identification and has published numerous papers and books
in this area.

Qinghua Zhang. Dr. Qinghua Zhang is a researcher at Institut National

de Recherche en Informatique et en Automatique (INRIA) and at Institut de
Recherche en Informatique et Systèmes Aléatoires (IRISA), both in Rennes,
France. He conducts research in the areas of nonlinear system identification,
fault diagnosis, and signal processing with applications in the fields of energy,
automotive, and biomedical systems.

Peter Lindskog. Dr. Peter Lindskog is employed by NIRA Dynamics

AB, Sweden. He conducts research in the areas of system identification,
signal processing, and automatic control with a focus on vehicle industry
applications.

Anatoli Juditsky. Professor Anatoli Juditsky is with the Laboratoire Jean

Kuntzmann at the Université Joseph Fourier, Grenoble, France. He conducts
research in the areas of nonparametric statistics, system identification, and
stochastic optimization.
About the Developers
 Contents

Choosing Your System Identification Approach

1
Linear Model Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2

Nonlinear Model Structures . . . . . . . . . . . . . . . . . . . . . . . . 1-4

Recommended Model Estimation Sequence . . . . . . . . . . 1-5

Supported Models for Time- and Frequency-Domain

Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
Supported Models for Time-Domain Data . . . . . . . . . . . . . . 1-7
Supported Models for Frequency-Domain Data . . . . . . . . . 1-8

Supported Continuous-Time and Discrete-Time

Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10

Commands for Model Estimation . . . . . . . . . . . . . . . . . . . . 1-12

Creating Model Structures at the Command Line . . . . . 1-14

About System Identification Toolbox Model Objects . . . . . . 1-14
When to Construct a Model Structure Independently of
Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-15
Commands for Constructing Model Structures . . . . . . . . . . 1-16
Model Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-17
See Also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-23

Modeling Multiple-Output Systems . . . . . . . . . . . . . . . . . . 1-24

About Modeling Multiple-Output Systems . . . . . . . . . . . . . 1-24
Modeling Multiple Outputs Directly . . . . . . . . . . . . . . . . . . 1-25
Modeling Multiple Outputs as a Combination of
Single-Output Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-25
Improving Multiple-Output Estimation Results by
Weighing Outputs During Estimation . . . . . . . . . . . . . . 1-26

vii
 Data Import and Processing
2
Types of Data You Can Model . . . . . . . . . . . . . . . . . . . . . . . 2-2

Ways to Process Data for System Identification . . . . . . 2-4

Requirements on Data Sampling . . . . . . . . . . . . . . . . . . . . 2-6

Importing Data into the MATLAB Workspace . . . . . . . . 2-7

Importing Time-Domain Data into MATLAB . . . . . . . . . . . 2-7
Importing Time-Series Data into MATLAB . . . . . . . . . . . . 2-8
Importing Frequency-Domain Data into MATLAB . . . . . . 2-9
Importing Frequency-Response Data into MATLAB . . . . . 2-11

Importing Data into the GUI . . . . . . . . . . . . . . . . . . . . . . . . 2-14

Types of Data You Can Import into the GUI . . . . . . . . . . . . 2-14
Importing Time-Domain Data into the GUI . . . . . . . . . . . . 2-16
Importing Frequency-Domain Data into the GUI . . . . . . . . 2-19
Importing Frequency-Response Data into the GUI . . . . . . 2-22
Importing Data Objects into the GUI . . . . . . . . . . . . . . . . . 2-26
Specifying the Data Sampling Interval . . . . . . . . . . . . . . . . 2-30
Specifying Estimation and Validation Data . . . . . . . . . . . . 2-31
Preprocessing Data Using Quick Start . . . . . . . . . . . . . . . . 2-32
Creating Data Sets from a Subset of Signal Channels . . . . 2-33
Creating Multiexperiment Data Sets in the GUI . . . . . . . . 2-35
Viewing Data Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-42
Renaming Data and Changing Display Color . . . . . . . . . . . 2-43
Distinguishing Data Types in the GUI . . . . . . . . . . . . . . . . 2-45
Organizing Data Icons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-45
Deleting Data Sets in the GUI . . . . . . . . . . . . . . . . . . . . . . . 2-46
Exporting Data from the GUI to the MATLAB
Workspace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-47

Representing Time- and Frequency-Domain Data Using

iddata Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-49
iddata Constructor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-49
iddata Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-52
Creating Multiexperiment Data at the Command Line . . . 2-55
Subreferencing iddata Objects . . . . . . . . . . . . . . . . . . . . . . . 2-57
Modifying Time and Frequency Vectors . . . . . . . . . . . . . . . 2-61

viii Contents
 Naming, Adding, and Removing Data Channels . . . . . . . . . 2-64
Concatenating iddata Objects . . . . . . . . . . . . . . . . . . . . . . . 2-66

Representing Frequency-Response Data Using idfrd

Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-69
idfrd Constructor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-69
idfrd Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-70
Subreferencing idfrd Objects . . . . . . . . . . . . . . . . . . . . . . . . 2-72
Concatenating idfrd Objects . . . . . . . . . . . . . . . . . . . . . . . . . 2-73
See Also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-76

Analyzing Data Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-78

Is Your Data Ready for Modeling? . . . . . . . . . . . . . . . . . . . . 2-78
Plotting Data in the GUI Versus at the Command Line . . 2-79
How to Plot Data in the GUI . . . . . . . . . . . . . . . . . . . . . . . . 2-79
How to Plot Data at the Command Line . . . . . . . . . . . . . . . 2-85
How to Analyze Data Using the advice Command . . . . . . . 2-87

Selecting Subsets of Data . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-89

Why Select Subsets of Data? . . . . . . . . . . . . . . . . . . . . . . . . 2-89
Selecting Data Using the GUI . . . . . . . . . . . . . . . . . . . . . . . 2-90
Selecting Data at the Command Line . . . . . . . . . . . . . . . . . 2-92

Handling Missing Data and Outliers . . . . . . . . . . . . . . . . . 2-93

Handling Missing Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-93
Handling Outliers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-94
Example – Extracting and Modeling Specific Data
Segments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-95
See Also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-96

Handling Offsets and Trends in Data . . . . . . . . . . . . . . . . 2-97

When to Detrend Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-97
Alternatives for Detrending Data in GUI or at the
Command-Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-98
How to Detrend Data Using the GUI . . . . . . . . . . . . . . . . . . 2-99
How to Detrend Data at the Command Line . . . . . . . . . . . . 2-99
Next Steps After Detrending . . . . . . . . . . . . . . . . . . . . . . . . 2-101

Resampling Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-102

What Is Resampling? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-102
Resampling Data Using the GUI . . . . . . . . . . . . . . . . . . . . . 2-103

ix
 Resampling Data at the Command Line . . . . . . . . . . . . . . . 2-103
Resampling Data Without Aliasing Effects . . . . . . . . . . . . . 2-105
See Also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-108

Filtering Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-109

Supported Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-109
Choosing to Prefilter Your Data . . . . . . . . . . . . . . . . . . . . . . 2-109
How to Filter Data Using the GUI . . . . . . . . . . . . . . . . . . . . 2-110
How to Filter Data at the Command Line . . . . . . . . . . . . . . 2-113
See Also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-116

Generating Data Using Simulation . . . . . . . . . . . . . . . . . . 2-117

Commands for Generating and Simulating Data . . . . . . . . 2-117
Example – Creating Data with Periodic Inputs . . . . . . . . . 2-118
Example – Generating Data Using Simulation . . . . . . . . . . 2-119
Simulating Data Using Other MathWorks Products . . . . . 2-120

Transforming Between Time- and Frequency-Domain

Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-121
Transforming Data Domain in the GUI . . . . . . . . . . . . . . . . 2-121
Transforming Data Domain at the Command Line . . . . . . 2-128

Manipulating Complex-Valued Data . . . . . . . . . . . . . . . . . 2-133

Supported Operations for Complex Data . . . . . . . . . . . . . . . 2-133
Processing Complex iddata Signals at the Command
Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-133

Linear Model Identification

3
Identifying Frequency-Response Models . . . . . . . . . . . . . 3-2
What Is a Frequency-Response Model? . . . . . . . . . . . . . . . . 3-2
Data Supported by Frequency-Response Models . . . . . . . . 3-3
How to Estimate Frequency-Response Models in the
GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
How to Estimate Frequency-Response Models at the
Command Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5
Options for Computing Spectral Models . . . . . . . . . . . . . . . 3-5
Options for Frequency Resolution . . . . . . . . . . . . . . . . . . . . 3-6

x Contents
 Spectrum Normalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8

Identifying Impulse-Response Models . . . . . . . . . . . . . . . 3-11

What Is Time-Domain Correlation Analysis? . . . . . . . . . . . 3-11
Data Supported by Correlation Analysis . . . . . . . . . . . . . . . 3-12
How to Estimate Impulse and Step Response Models Using
the GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-12
How to Estimate Impulse and Step Response Models at the
Command Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-14
How to Compute Response Values . . . . . . . . . . . . . . . . . . . . 3-15
How to Identify Delay Using Transient-Response Plots . . . 3-16
Algorithm for Correlation Analysis . . . . . . . . . . . . . . . . . . . 3-18

Identifying Low-Order Transfer Functions (Process

Models) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-20
What Is a Process Model? . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-20
Data Supported by a Process Model . . . . . . . . . . . . . . . . . . . 3-21
How to Estimate Process Models Using the GUI . . . . . . . . 3-21
How to Estimate Process Models at the Command Line . . 3-27
Options for Specifying the Process-Model Structure . . . . . 3-33
Options for Multiple-Input Models . . . . . . . . . . . . . . . . . . . 3-34
Options for the Disturbance Model Structure . . . . . . . . . . . 3-35
Options for Frequency-Weighing Focus . . . . . . . . . . . . . . . . 3-36
Options for Initial States . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-37

Identifying Input-Output Polynomial Models . . . . . . . . 3-39

What Are Black-Box Polynomial Models? . . . . . . . . . . . . . . 3-39
Data Supported by Polynomial Models . . . . . . . . . . . . . . . . 3-46
Preliminary Step – Estimating Model Orders and Input
Delays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-48
How to Estimate Polynomial Models in the GUI . . . . . . . . 3-56
How to Estimate Polynomial Models at the Command
Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-59
Options for Multiple-Input and Multiple-Output ARX
Orders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-64
Option for Frequency-Weighing Focus . . . . . . . . . . . . . . . . . 3-65
Options for Initial States . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-66
Algorithms for Estimating Polynomial Models . . . . . . . . . . 3-66
Example – Estimating Models Using armax . . . . . . . . . . . . 3-67

Identifying State-Space Models . . . . . . . . . . . . . . . . . . . . . 3-73

What Are State-Space Models? . . . . . . . . . . . . . . . . . . . . . . 3-73

xi
 Data Supported by State-Space Models . . . . . . . . . . . . . . . . 3-77
Supported State-Space Parameterizations . . . . . . . . . . . . . 3-78
Preliminary Step – Estimating State-Space Model
Orders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-79
How to Estimate State-Space Models in the GUI . . . . . . . . 3-84
How to Estimate State-Space Models at the Command
Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-87
How to Estimate Free-Parameterization State-Space
Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-91
How to Estimate State-Space Models with Canonical
Parameterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-92
How to Estimate State-Space Models with Structured
Parameterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-94
How to Estimate the State-Space Equivalent of ARMAX
and OE Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-100
Options for Frequency-Weighing Focus . . . . . . . . . . . . . . . . 3-101
Options for Initial States . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-102
Algorithms for Estimating State-Space Models . . . . . . . . . 3-103

Refining Linear Parametric Models . . . . . . . . . . . . . . . . . 3-104

When to Refine Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-104
What You Specify to Refine a Model . . . . . . . . . . . . . . . . . . 3-104
How to Refine Linear Parametric Models in the GUI . . . . . 3-105
How to Refine Linear Parametric Models at the Command
Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-106

Extracting Parameter Values from Linear Models . . . . 3-109

Extracting Dynamic Model and Noise Model

Separately . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-111

Transforming Between Discrete-Time and

Continuous-Time Representations . . . . . . . . . . . . . . . . 3-113
Why Transform Between Continuous and Discrete
Time? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-113
Using the c2d, d2c, and d2d Commands . . . . . . . . . . . . . . . 3-113
Specifying Intersample Behavior . . . . . . . . . . . . . . . . . . . . . 3-115
How d2c Handles Input Delays . . . . . . . . . . . . . . . . . . . . . . 3-115
Effects on the Noise Model . . . . . . . . . . . . . . . . . . . . . . . . . . 3-116

Transforming Between Linear Model

Representations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-118

xii Contents
 Subreferencing Model Objects . . . . . . . . . . . . . . . . . . . . . . 3-120
What Is Subreferencing? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-120
Limitation on Supported Models . . . . . . . . . . . . . . . . . . . . . 3-120
Subreferencing Specific Measured Channels . . . . . . . . . . . 3-120
Subreferencing Measured and Noise Models . . . . . . . . . . . 3-121
Treating Noise Channels as Measured Inputs . . . . . . . . . . 3-123

Concatenating Model Objects . . . . . . . . . . . . . . . . . . . . . . . 3-125

About Concatenating Models . . . . . . . . . . . . . . . . . . . . . . . . 3-125
Limitation on Supported Models . . . . . . . . . . . . . . . . . . . . . 3-125
Horizontal Concatenation of Model Objects . . . . . . . . . . . . 3-126
Vertical Concatenation of Model Objects . . . . . . . . . . . . . . . 3-126
Concatenating Noise Spectrum Data of idfrd Objects . . . . 3-127
See Also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-128

Merging Model Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-129

Nonlinear Black-Box Model Identification

4
About Nonlinear Model Identification . . . . . . . . . . . . . . . 4-2
What Are Nonlinear Models? . . . . . . . . . . . . . . . . . . . . . . . . 4-2
When to Fit Nonlinear Models . . . . . . . . . . . . . . . . . . . . . . . 4-2
Available Nonlinear Models . . . . . . . . . . . . . . . . . . . . . . . . . 4-4

Preparing Data for Nonlinear Identification . . . . . . . . . 4-7

Identifying Nonlinear ARX Models . . . . . . . . . . . . . . . . . . 4-8

Nonlinear ARX Model Extends the Linear ARX
Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
Structure of Nonlinear ARX Models . . . . . . . . . . . . . . . . . . 4-9
Nonlinearity Estimators for Nonlinear ARX Models . . . . . 4-10
Ways to Configure Nonlinear ARX Estimation . . . . . . . . . . 4-12
How to Estimate Nonlinear ARX Models in the GUI . . . . . 4-16
How to Estimate Nonlinear ARX Models at the Command
Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-19
Using Linear Model for Nonlinear ARX Estimation . . . . . . 4-28
Validating Nonlinear ARX Models . . . . . . . . . . . . . . . . . . . . 4-35
Using Nonlinear ARX Models . . . . . . . . . . . . . . . . . . . . . . . 4-40

xiii
 Low-Level Manipulation of Nonlinear ARX Components . . 4-41

Identifying Hammerstein-Wiener Models . . . . . . . . . . . . 4-49

Applications of Hammerstein-Wiener Models . . . . . . . . . . . 4-49
Structure of Hammerstein-Wiener Models . . . . . . . . . . . . . 4-50
Nonlinearity Estimators for Hammerstein-Wiener
Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-52
Ways to Configure Hammerstein-Wiener Estimation . . . . 4-53
Estimation Algorithm for Hammerstein-Wiener Models . . 4-55
How to Estimate Hammerstein-Wiener Models in the
GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-55
How to Estimate Hammerstein-Wiener Models at the
Command Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-58
Using Linear Model for Hammerstein-Wiener
Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-64
Validating Hammerstein-Wiener Models . . . . . . . . . . . . . . 4-70
Using Hammerstein-Wiener Models . . . . . . . . . . . . . . . . . . 4-76
Low-Level Manipulation of Hammerstein-Wiener
Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-78

Linear Approximation of Nonlinear Black-Box

Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-81
Why Compute a Linear Approximation of a Nonlinear
Model? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-81
Choosing Your Linear Approximation Approach . . . . . . . . 4-81
Linear Approximation of Nonlinear Black-Box Models for a
Given Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-82
Tangent Linearization of Nonlinear Black-Box Models . . . 4-82
Computing Operating Points for Nonlinear Black-Box
Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-83

ODE Parameter Estimation (Grey-Box

Modeling)
5
Supported Grey-Box Models . . . . . . . . . . . . . . . . . . . . . . . . 5-2

Data Supported by Grey-Box Models . . . . . . . . . . . . . . . . 5-3

xiv Contents
 Choosing idgrey or idnlgrey Model Object . . . . . . . . . . . 5-4

Estimating Linear Grey-Box Models . . . . . . . . . . . . . . . . . 5-6

Specifying the Linear Grey-Box Model Structure . . . . . . . . 5-6
Example – Creating a Function for Representing a Grey-Box
Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-7
Example – Estimating a Continuous-Time Grey-Box Model
for Heat Diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-9
Example – Estimating a Discrete-Time Grey-Box Model
with Parameterized Disturbance . . . . . . . . . . . . . . . . . . . 5-12

Estimating Nonlinear Grey-Box Models . . . . . . . . . . . . . . 5-16

Supported Nonlinear Grey-Box Models . . . . . . . . . . . . . . . . 5-16
Nonlinear Grey-Box Demos and Examples . . . . . . . . . . . . . 5-16
Specifying the Nonlinear Grey-Box Model Structure . . . . . 5-17
Constructing the idnlgrey Object . . . . . . . . . . . . . . . . . . . . . 5-18
Using pem to Estimate Nonlinear Grey-Box Models . . . . . 5-19
Options for the Estimation Algorithm . . . . . . . . . . . . . . . . . 5-20

After Estimating Grey-Box Models . . . . . . . . . . . . . . . . . . 5-23

Time Series Identification

6
What Are Time-Series Models? . . . . . . . . . . . . . . . . . . . . . . 6-2

Preparing Time-Series Data . . . . . . . . . . . . . . . . . . . . . . . . 6-3

Estimating Time-Series Power Spectra . . . . . . . . . . . . . . 6-4

How to Estimate Time-Series Power Spectra Using the
GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
How to Estimate Time-Series Power Spectra at the
Command Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-5

Estimating AR and ARMA Models . . . . . . . . . . . . . . . . . . . 6-7

Definition of AR and ARMA Models . . . . . . . . . . . . . . . . . . 6-7
Estimating Polynomial Time-Series Models in the GUI . . . 6-7

xv
 Estimating AR and ARMA Models at the Command
Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-10

Estimating State-Space Time-Series Models . . . . . . . . . . 6-12

Definition of State-Space Time-Series Model . . . . . . . . . . . 6-12
Estimating State-Space Models at the Command Line . . . 6-12

Example – Identifying Time-Series Models at the

Command Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-14

Estimating Nonlinear Models for Time-Series Data . . . 6-15

Recursive Model Identification

7
What Is Recursive Estimation? . . . . . . . . . . . . . . . . . . . . . . 7-2

Commands for Recursive Estimation . . . . . . . . . . . . . . . . 7-3

Algorithms for Recursive Estimation . . . . . . . . . . . . . . . . 7-6

Types of Recursive Estimation Algorithms . . . . . . . . . . . . . 7-6
General Form of Recursive Estimation Algorithm . . . . . . . 7-6
Kalman Filter Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-8
Forgetting Factor Algorithm . . . . . . . . . . . . . . . . . . . . . . . . 7-10
Unnormalized and Normalized Gradient Algorithms . . . . . 7-11

Data Segmentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-14

Model Analysis
8
Overview of Model Validation and Plots . . . . . . . . . . . . . 8-2
When to Validate Models . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2
Ways to Validate Models . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2

xvi Contents
 Data for Validating Models . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
Supported Model Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-4
Plotting Models in the GUI . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5
Getting Advice About Models . . . . . . . . . . . . . . . . . . . . . . . . 8-7

Simulating and Predicting Model Output . . . . . . . . . . . . 8-8

Using a Model for Simulation and Prediction . . . . . . . . . . . 8-8
Simulation and Prediction in the GUI . . . . . . . . . . . . . . . . . 8-11
Simulation and Prediction at the Command Line . . . . . . . . 8-16

Residual Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-21

What Is Residual Analysis? . . . . . . . . . . . . . . . . . . . . . . . . . 8-21
Supported Model Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-22
What Residual Plots Show for Different Data Domains . . . 8-22
Displaying the Confidence Interval . . . . . . . . . . . . . . . . . . . 8-23
How to Plot Residuals Using the GUI . . . . . . . . . . . . . . . . . 8-24
How to Plot Residuals at the Command Line . . . . . . . . . . . 8-26
Example – Examining Model Residuals . . . . . . . . . . . . . . . 8-26

Impulse and Step Response Plots . . . . . . . . . . . . . . . . . . . 8-30

Supported Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-30
How Transient Response Helps to Validate Models . . . . . . 8-30
What Does a Transient Response Plot Show? . . . . . . . . . . . 8-31
How to Plot Impulse and Step Response Using the GUI . . 8-32
Displaying the Confidence Interval . . . . . . . . . . . . . . . . . . . 8-35
How to Plot Impulse and Step Response at the Command
Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-36

Frequency Response Plots . . . . . . . . . . . . . . . . . . . . . . . . . . 8-38

What Is Frequency Response? . . . . . . . . . . . . . . . . . . . . . . . 8-38
How Frequency Response Helps to Validate Models . . . . . 8-39
What Does a Frequency-Response Plot Show? . . . . . . . . . . 8-40
How to Plot Bode Plots Using the GUI . . . . . . . . . . . . . . . . 8-41
How to Plot Bode and Nyquist Plots at the Command
Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-43

Noise Spectrum Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-46

Supported Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-46
What Does a Noise Spectrum Plot Show? . . . . . . . . . . . . . . 8-46
Displaying the Confidence Interval . . . . . . . . . . . . . . . . . . . 8-47
How to Plot the Noise Spectrum Using the GUI . . . . . . . . . 8-48
How to Plot the Noise Spectrum at the Command Line . . . 8-51

xvii
 Pole and Zero Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-53
Supported Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-53
What Does a Pole-Zero Plot Show? . . . . . . . . . . . . . . . . . . . 8-53
How to Plot Model Poles and Zeros Using the GUI . . . . . . 8-55
How to Plot Poles and Zeros at the Command Line . . . . . . 8-56
Reducing Model Order Using Pole-Zero Plots . . . . . . . . . . . 8-57

Akaike’s Criteria for Model Validation . . . . . . . . . . . . . . . 8-58

Definition of FPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-58
Computing FPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-59
Definition of AIC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-59
Computing AIC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-60

Computing Model Uncertainty . . . . . . . . . . . . . . . . . . . . . . 8-61

Why Analyze Model Uncertainty? . . . . . . . . . . . . . . . . . . . . 8-61
What Is Model Covariance? . . . . . . . . . . . . . . . . . . . . . . . . . 8-61
Viewing Model Uncertainty Information . . . . . . . . . . . . . . . 8-62

Troubleshooting Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-64

About Troubleshooting Models . . . . . . . . . . . . . . . . . . . . . . . 8-64
Model Order Is Too High or Too Low . . . . . . . . . . . . . . . . . . 8-64
Nonlinearity Estimator Produces a Poor Fit . . . . . . . . . . . . 8-65
Substantial Noise in the System . . . . . . . . . . . . . . . . . . . . . 8-66
Unstable Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-66
Missing Input Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-68
Complicated Nonlinearities . . . . . . . . . . . . . . . . . . . . . . . . . 8-68

Next Steps After Getting an Accurate Model . . . . . . . . . 8-69

Control Design Applications

9
Using Models with Control System Toolbox Software . . 9-2
How Control System Toolbox Software Works with
Identified Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
Using balred to Reduce Model Order . . . . . . . . . . . . . . . . . . 9-3
Compensator Design Using Control System Toolbox
Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-3
Converting Models to LTI Objects . . . . . . . . . . . . . . . . . . . . 9-4

xviii Contents
 Viewing Model Response Using the LTI Viewer . . . . . . . . . 9-5
Combining Model Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-6
Example – Using System Identification Toolbox Software
with Control System Toolbox Software . . . . . . . . . . . . . . 9-6

System Identification Toolbox Blocks

10
System Identification Toolbox Block Library . . . . . . . . . 10-2

Opening the System Identification Toolbox Block

Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-3

Preparing Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-4

Identifying Linear Models . . . . . . . . . . . . . . . . . . . . . . . . . . 10-5

Simulating Model Output . . . . . . . . . . . . . . . . . . . . . . . . . . 10-6

When to Use Simulation Blocks . . . . . . . . . . . . . . . . . . . . . . 10-6
Summary of Simulation Blocks . . . . . . . . . . . . . . . . . . . . . . 10-6
Specifying Initial Conditions for Simulation . . . . . . . . . . . . 10-7

Example – Simulating a Model Using Simulink

Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-9

System Identification Tool GUI

11
Steps for Using the System Identification Tool GUI . . . 11-2

Starting and Managing GUI Sessions . . . . . . . . . . . . . . . . 11-3

What Is a System Identification Tool Session? . . . . . . . . . . 11-3
Starting a New Session in the GUI . . . . . . . . . . . . . . . . . . . 11-4
Description of the System Identification Tool Window . . . . 11-5

xix
 Opening a Saved Session . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-6
Saving, Merging, and Closing Sessions . . . . . . . . . . . . . . . . 11-6
Deleting a Session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-7
Getting Help in the GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-7
Exiting the System Identification Tool GUI . . . . . . . . . . . . 11-8

Managing Models in the GUI . . . . . . . . . . . . . . . . . . . . . . . . 11-9

Importing Models into the GUI . . . . . . . . . . . . . . . . . . . . . . 11-9
Viewing Model Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-10
Renaming Models and Changing Display Color . . . . . . . . . 11-11
Organizing Model Icons . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-11
Deleting Models in the GUI . . . . . . . . . . . . . . . . . . . . . . . . . 11-12
Exporting Models from the GUI to the MATLAB
Workspace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-13

Working with Plots in the System Identification Tool

GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-15
Identifying Data Sets and Models on Plots . . . . . . . . . . . . . 11-15
Changing and Restoring Default Axis Limits . . . . . . . . . . . 11-16
Selecting Measured and Noise Channels in Plots . . . . . . . . 11-18
Grid and Line Styles in Plots . . . . . . . . . . . . . . . . . . . . . . . . 11-19
Opening a Plot in a MATLAB Figure Window . . . . . . . . . . 11-19
Printing Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-19

Customizing the System Identification Tool GUI . . . . . 11-20

Types of GUI Customization . . . . . . . . . . . . . . . . . . . . . . . . 11-20
Displaying Warnings While You Work . . . . . . . . . . . . . . . . 11-20
Saving Session Preferences . . . . . . . . . . . . . . . . . . . . . . . . . 11-20
Modifying idlayout.m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-21

Index

xx Contents
 1

Choosing Your System

Identification Approach

• “Linear Model Structures” on page 1-2

• “Nonlinear Model Structures” on page 1-4
• “Recommended Model Estimation Sequence” on page 1-5
• “Supported Models for Time- and Frequency-Domain Data” on page 1-7
• “Supported Continuous-Time and Discrete-Time Models” on page 1-10
• “Commands for Model Estimation” on page 1-12
• “Creating Model Structures at the Command Line” on page 1-14
• “Modeling Multiple-Output Systems” on page 1-24
 1 Choosing Your System Identification Approach

Linear Model Structures

A linear model is often sufficient to accurately describe the system dynamics
and, in most cases, you should first try to fit linear models. Available linear
structures include transfer functions and state-space models, summarized
in the following table.

Model Type Usage Learn More

Process model Use this structure to “Identifying Low-Order

(idproc object) represent low-order Transfer Functions
transfer function that (Process Models)” on
include integrator, page 3-20
delay, zero, and up
to 3 poles. You can
also specify parameter
bounds.
State-space model Use this structure “Identifying
(idss object) to represent known State-Space Models”
state-space structures on page 3-73
and black-box
structures. You can
fix certain parameters
to known values and
estimate the remaining
parameters. If you need
to specify parameter
dependencies or
constraints, use
the grey-box model
structure.

1-2
 Linear Model Structures

Model Type Usage Learn More

Generalized transfer Use to represent linear “Identifying

function transfer functions Input-Output
(idpoly object) based on the general Polynomial Models”
form input-output on page 3-39
polynomial form:

B C
Ay = u+ e
F D

where A, B, C, D and
F are polynomials with
coefficients that the
toolbox estimates from
data.
Typically, you begin
modeling using
simpler forms of
this generalized
structure (such as
ARX: Ay = Bu + e and
B
OE: y = u + e ) and, if
F increase the
necessary,
model complexity.
Grey-box model Use to represent “Estimating Linear
(idgrey object) arbitrary Grey-Box Models” on
parameterizations page 5-6
of state-space models.
For example, you can
use this structure
to represent your
ordinary differential
or difference equation
(ODE) and to
define parameter
dependencies.

1-3
 1 Choosing Your System Identification Approach

Nonlinear Model Structures

System Identification Toolbox provides several nonlinear black-box model
structures, which have traditionally been useful for representing dynamic
systems.

Model Type Usage Learn More

Nonlinear ARX model Use to represent “Identifying Nonlinear
(idnlarx object) nonlinear extensions ARX Models” on page
of linear models. This 4-8
structure allows you
to model complex
nonlinear behavior
using flexible nonlinear
functions, such as
wavelet and sigmoid
networks.
Linear models Use to represent linear “Identifying
with input/output models with static Hammerstein-Wiener
nonlinearities, or nonlinearities. Models” on page 4-49
Hammerstein-Wiener
model
(idnlhw object)
Nonlinear grey-box Use to represent “Estimating Nonlinear
model nonlinear ODEs with Grey-Box Models” on
(idnlgrey object) unknown parameters. page 5-16

1-4
 Recommended Model Estimation Sequence

Recommended Model Estimation Sequence

System identification is an iterative process, where you identify models
with different structures from data and compare model performance. You
start by estimating the parameters of simple model structures. If the model
performance is poor, you gradually increase the complexity of the model
structure. Ultimately, you choose the simplest model that best describes
the dynamics of your system.

Another reason to start with simple model structures is that higher-order

models are not always more accurate. Increasing model complexity increases
the uncertainties in parameter estimates and typically requires more data
(which is common in the case of nonlinear models).

Note Model structure is not the only factor that determines model accuracy.
If your model is poor, you might need to preprocess your data by removing
outliers or filtering noise. For more information, see “Ways to Process Data
for System Identification” on page 2-4.

Estimate impulse-response and frequency-response models first to gain

insight into the system dynamics and assess whether a linear model is
sufficient. Then, estimate parametric models in the following order:

1 ARX polynomial and state-space models provide the simplest structures.

These models let you estimate the model order and noise dynamics.

In the System Identification Tool GUI. Select to estimate the ARX

linear parametric model and the state-space model using the N4SID
method.

At the command line. Use the arx and the n4sid commands.

For more information, see “Identifying Input-Output Polynomial Models”

on page 3-39 and “Identifying State-Space Models” on page 3-73.

2 ARMAX and BJ polynomial models provide more complex structures and

require iterative estimation. Try several model orders and keep the model
orders as low as possible.

1-5
 1 Choosing Your System Identification Approach

In the System Identification Tool GUI. Select to estimate the BJ and

ARMAX linear parametric models.

At the command line. Use the bj or armax commands.

For more information, see “Identifying Input-Output Polynomial Models”

on page 3-39.

3 Nonlinear ARX or Hammerstein-Wiener models provide nonlinear

structures. For more information, see Chapter 4, “Nonlinear Black-Box
Model Identification”.

For general information about choosing you model strategy, see “About
System Identification”. For information about validating models, see
“Overview of Model Validation and Plots” on page 8-2.

1-6
 Supported Models for Time- and Frequency-Domain Data

Supported Models for Time- and Frequency-Domain Data

In this section...
“Supported Models for Time-Domain Data” on page 1-7
“Supported Models for Frequency-Domain Data” on page 1-8

Supported Models for Time-Domain Data

Continuous-Time Models
You can directly estimate the following types of continuous-time models:

• Low-order transfer functions. See “Identifying Low-Order Transfer

Functions (Process Models)” on page 3-20.
• Input-output polynomial models. See “Identifying Input-Output Polynomial
Models” on page 3-39.
• State-space models. See “Identifying State-Space Models” on page 3-73.

To get a linear, continuous-time model of arbitrary structure for time-domain

data, you can estimate a discrete-time model, and then use d2c to transform
it to a continuous-time model.

Discrete-Time Models
You can estimate all linear and nonlinear models supported by the
System Identification Toolbox product as discrete-time models, except the
continuous-time transfer functions (process models). For more information
about process models, see “Identifying Low-Order Transfer Functions (Process
Models)” on page 3-20.

ODEs (Grey-Box Models)

You can estimate both continuous-time and discrete-time models from
time-domain data for linear and nonlinear differential and difference
equations. See Chapter 5, “ODE Parameter Estimation (Grey-Box Modeling)”.

1-7
 1 Choosing Your System Identification Approach

Nonlinear Models
You can estimate discrete-time Hammerstein-Wiener and nonlinear ARX
models from time-domain data. See Chapter 4, “Nonlinear Black-Box Model
Identification”.

You can also estimate nonlinear grey-box models from time-domain data. See
“Estimating Nonlinear Grey-Box Models” on page 5-16.

Supported Models for Frequency-Domain Data

There are two types of frequency-domain data:

• Continuous-time data
• Discrete-time data

You specify frequency-domain data as continuous- or discrete-time when you

either import data into the System Identification Tool GUI or create a System
Identification Toolbox data object. For more information about representing
your data as System Identification Toolbox data objects, see Chapter 2, “Data
Import and Processing”.

To designate discrete-time data, you set the sampling interval of the data to
the experimental data sampling interval. To designate continuous-time data,
you must set the sampling interval of the data to zero. Setting the sampling
interval to zero corresponds to taking a Fourier transform of continuous-time
data.

Continuous-Time Models
You can estimate the following types of continuous-time models directly:

• Low-order transfer functions. See “Identifying Low-Order Transfer

Functions (Process Models)” on page 3-20.
• Input-output polynomial models. See “Identifying Input-Output Polynomial
Models” on page 3-39.
• State-space models.
From continuous-time frequency-domain data, you can estimate
continuous-time state-space models. From discrete-time frequency-domain

1-8
 Supported Models for Time- and Frequency-Domain Data

data, you can estimate continuous-time black-box models with canonical

parameterization. See “Identifying State-Space Models” on page 3-73.

To get a linear, continuous-time model of arbitrary structure for

frequency-domain data, you can estimate a discrete-time model and use d2c
to transform it to a continuous-time model.

Discrete-Time Models
You can estimate only output-error (OE) polynomial models using
frequency-domain data. See “Identifying Input-Output Polynomial Models”
on page 3-39.

Other linear model structures include noise models, which are not supported
for frequency-domain data.

ODEs (Grey-Box Models)

For linear grey-box models, you can estimate both continuous-time and
discrete-time models from frequency-domain data.

Nonlinear grey-box models are supported only for time-domain data.

See Chapter 5, “ODE Parameter Estimation (Grey-Box Modeling)”.

Nonlinear Black-Box Models

Frequency-domain data is not relevant to nonlinear black-box models, which
support only time-domain data.

1-9
 1 Choosing Your System Identification Approach

Supported Continuous-Time and Discrete-Time Models

For linear and nonlinear ODEs (grey-box models), you can specify any
ordinary differential or difference equation to represent your continuous-time
or discrete-time model in state-space form, respectively. In the linear case,
both time-domain and frequency-domain data are supported. In the nonlinear
case, only time-domain data is supported.

For black-box models, the following tables summarize supported

continuous-time and discrete-time models.

Supported Continuous-Time Models

Model Type Description

Low-order transfer functions Estimate low-order process models for up to three free poles
(process models) from either time- or frequency-domain data.
Linear input-output polynomial To get a linear, continuous-time model of arbitrary
models structure from time-domain data, you can estimate a
discrete-time model, and then use d2c to transform it into a
continuous-time model.
For frequency-domain data, you can directly estimate only
the ARX and output-error (OE) continuous-time polynomial
models by setting the sampling interval of the data to 0.
Other structures include noise models and are not supported
for frequency-domain data.
State-space models To get a linear, continuous-time model of arbitrary
structure for time-domain data, you can estimate a
discrete-time model, and then use d2c to transform it into a
continuous-time model.
For frequency-domain data, you can estimate
continuous-time state-space models directly.
Linear ODEs (grey-box models Estimate ordinary differential equations (ODEs) from either
time- or frequency-domain data.
Nonlinear ODEs (grey-box) Estimate arbitrary differential equations (ODEs) from
models time-domain data.

1-10
 Supported Continuous-Time and Discrete-Time Models

Supported Discrete-Time Models

Model Type Description

Linear, input-output polynomial Estimate arbitrary-order, linear parametric models from
models time- or frequency-domain data.
To get a discrete-time model, your data sampling interval
must be set to the (nonzero) value you used to sample in
your experiment.
Nonlinear black-box models Estimate from time-domain data only.
Linear ODEs (grey-box) models Estimate ordinary difference equations from time- or
frequency-domain data.
Nonlinear ODEs (grey-box) Estimate ordinary difference equations from time-domain
models data.

1-11
 1 Choosing Your System Identification Approach

Commands for Model Estimation

The quickest way to both construct a model object and estimate the model
parameters is to use estimation commands.

Note For ODEs (grey-box models), you must first construct the model
structure and then apply an estimation command to the resulting model
object.

For ARMAX, Box-Jenkins, and Output-Error Models—which you can only

estimate using the iterative prediction-error method—use the armax, bj, and
oe estimation commands, respectively. For more information about choosing
the models to estimate first, see “Recommended Model Estimation Sequence”
on page 1-5.

The following table summarizes System Identification Toolbox estimation

commands. For detailed information about using each command, see the
corresponding reference page.

Commands for Constructing and Estimating Models

Model Type Estimation Commands

Continuous-time low-order pem
transfer functions (process
models)
Linear input-output armax (ARMAX only)
polynomial models arx (ARX only)
bj (BJ only)
iv4 (ARX only)
oe (OE only)
pem (for all models)
State-space models n4sid
pem

1-12
 Commands for Model Estimation

Commands for Constructing and Estimating Models (Continued)

Model Type Estimation Commands

Linear time-series models ar
arx (for multiple outputs)
ivar
Nonlinear ARX models nlarx
Hammerstein-Wiener nlhw
models

1-13
 1 Choosing Your System Identification Approach

Creating Model Structures at the Command Line

In this section...
“About System Identification Toolbox Model Objects” on page 1-14
“When to Construct a Model Structure Independently of Estimation” on
page 1-15
“Commands for Constructing Model Structures” on page 1-16
“Model Properties” on page 1-17
“See Also” on page 1-23

About System Identification Toolbox Model Objects

Objects are based on model classes. Each class is a blueprint that defines the
following information about your model:

• How the object stores data

• Which operations you can perform on the object

This toolbox includes nine classes for representing models. For example,
idpoly represents linear input-output polynomial models, and idss
represents linear state-space models. For a complete list of available model
objects, see “Commands for Constructing Model Structures” on page 1-16.

Model properties define how a model object stores information. Model objects
store information about a model, including the mathematical form of a
model, names of input and output channels, units, names and values of
estimated parameters, parameter uncertainties, algorithm specifications, and
estimation information. For example, the idpoly model class has a property
called InputName for storing one or more input channel names. Different
model objects have different properties.

The allowed operations on an object are called methods. In the System

Identification Toolbox product, some methods have the same name but apply
to multiple model objects. For example, the method bode creates a bode plot
for all linear model objects. However, other methods are unique to a specific

1-14
 Creating Model Structures at the Command Line

model object. For example, the estimation method n4sid is unique to the
state-space model object idss.

Every class has a special method for creating objects of that class, called the
constructor. Using a constructor creates an instance of the corresponding
class or instantiates the object. The constructor name is the same as the class
name. For example, idpoly is both the name of the class representing linear
black-box polynomial models and the name of the constructor for instantiating
the model object.

For a tutorial about estimating models at the command line, see “Tutorial –
Identifying Linear Models Using the Command Line” in System Identification
Toolbox Getting Started Guide.

When to Construct a Model Structure Independently

of Estimation
You use model constructors to create a model object at the command line by
specifying all required model properties explicitly.

You must construct the model object independently of estimation when you
want to:

• Simulate a model
• Analyze a model
• Specify an initial guess for specific model parameter values before
estimation

In most cases, you can use the estimation commands to both construct
and estimate the model—without having to construct the model object
independently. For example, the estimation command pem lets you specify
both the model structure with unknown parameters and the estimation
algorithm. For information about how to both construct and estimate models
with a single command, see “Commands for Model Estimation” on page 1-12.

In case of grey-box models, you must always construct the model object first
and then estimate the parameters of the ordinary differential or difference
equation. For more information, see Chapter 5, “ODE Parameter Estimation
(Grey-Box Modeling)”.

1-15
 1 Choosing Your System Identification Approach

Commands for Constructing Model Structures

The following table summarizes the model constructors available in the
System Identification Toolbox product for representing various types of
models.

After model estimation, you can recognize the corresponding model objects
in the MATLAB® Workspace browser by their class names. The name of the
constructor matches the name of the object it creates.

For information about how to both construct and estimate models with a
single command, see “Commands for Model Estimation” on page 1-12.

Summary of Model Constructors

Model Constructor Resulting Model Class Single or Multiple Outputs?

idarx Parametric multiple-output Single- or multiple-output
ARX models. Also models.
represents nonparametric
transient-response models.
idfrd Nonparametric Single- or multiple-output
frequency-response model. models.
idproc Continuous-time, low-order Single-output models only.
transfer functions (process
models).
idpoly Linear input-output polynomial Single-output models only.
models:

• ARX
• ARMAX
• Output-Error
• Box-Jenkins
idss Linear state-space models. Single- or multiple-output
models.

1-16
 Creating Model Structures at the Command Line

Summary of Model Constructors (Continued)

Model Constructor Resulting Model Class Single or Multiple Outputs?

idgrey Linear ordinary differential or Single- and multiple-output
difference equations (grey-box models.
models). You write a function
that translates user parameters
to state-space matrices.
idnlgrey Nonlinear ordinary differential Supports single- and
or difference equation (grey-box multiple-output models.
models). You write a function
or MEX-file to represent the
set of first-order differential or
difference equations.
idnlarx Nonlinear ARX models, which Single- or multiple-output
define the predicted output as a models.
nonlinear function of past inputs
and outputs.
idnlhw Nonlinear Hammerstein-Wiener Single- or multiple-output
models, which include a linear models. Does not support time
dynamic system with nonlinear series.
static transformations of inputs
and outputs.

For more information about when to use these commands, see “When to
Construct a Model Structure Independently of Estimation” on page 1-15.

Model Properties
• “Categories of Model Properties” on page 1-18
• “Specifying Model Properties for Estimation” on page 1-19
• “Viewing Model Properties and Estimated Parameters” on page 1-20
• “Getting Help on Model Properties at the Command Line” on page 1-22

1-17
 1 Choosing Your System Identification Approach

Categories of Model Properties

The way a model object stores information is defined by the properties of the
corresponding model class.

Each model object has properties for storing information that are relevant
only to that specific model type. However, the idarx, idgrey, idpoly, idproc,
and idss model objects are based on the idmodel superclass and inherit all
idmodel properties.

Similarly, the nonlinear models idnlarx, idnlhw, and idnlgrey are based on
the idnlmodel superclass and inherit all idnlmodel properties.

In general, all nonlinear model objects have properties that belong to the
following categories:

• Names of input and output channels, such as InputName and OutputName

• Sampling interval of the model, such as Ts
• Units for time or frequency
• Model order and mathematical structure (for example, ODE or
nonlinearities)
• Properties that store estimation results and model uncertainty
• User comments, such as Notes and Userdata
• Estimation algorithm information
- Algorithm
Structure includes fields that specify the estimation method. Algorithm
includes another structure, called Advanced, which provides additional
flexibility for setting the search algorithm. Different fields apply to
different estimation techniques.
For linear parametric models, Algorithm specifies the frequency
weighing of the estimation using the Focus property.

Note Algorithm does not apply to idfrd models.

1-18
 Creating Model Structures at the Command Line

- EstimationInfo
Structure includes read-only fields that describe the estimation data set,
quantitative model quality measures, search termination conditions,
how the initial states are handled, and any warnings encountered during
the estimation.

For information about getting help on object properties, see “Getting Help on
Model Properties at the Command Line” on page 1-22.

Specifying Model Properties for Estimation

If you are estimating a new model, you can specify model properties directly
in the estimator syntax. For a complete list of model estimation commands,
see “Commands for Model Estimation” on page 1-12.

When using the commands that let you both construct and estimate a model,
you can specify all top-level model properties in the estimator syntax.
Top-level properties are those listed when you type get(object_name). You
can also specify the top-level fields of the Algorithm structure directly in
the estimator using property-value pairs—such as focus in the previous
example—without having to define the structure fields first.

The following commands load the sample data, z8, construct an ARMAX
model, and estimate the model parameters. The arguments of the armax
estimator specify model properties as property-value pairs.

load iddata8
m_armax=armax(z8,'na',4,...
'nb',[3 2 3],...
'nc',4,...
'nk',[0 0 0],...
'focus', 'simulation',...
'covariance', 'none',...
'tolerance',1e-5,...
'maxiter',50);

focus, covariance, tolerance, and maxiter are fields in the Algorithm

model property and specify aspects of the estimation algorithm.

1-19
 1 Choosing Your System Identification Approach

For linear models, you can use a shortcut to specify the second-level
Algorithm properties, such as Advanced. With this syntax, you can reference
the structure fields by name without specifying the structure to which these
fields belong.

However, when estimating nonlinear black-box models, you must set the
specific fields of the Advanced Algorithm structure and the nonlinearity
estimators before estimation. For example, suppose you want to set the value
of the wavenet object property Options, which is a structure. The following
commands set the Options values before estimation and include the modified
wavenet object in the estimator:

% Define wavenet object with defaul properties

W = wavenet;
% Specify variable to represent Options field
O = W.Options;
% Modify values of specific Options fields
O.MaxLevels = 5 ;
O.DilationStep = 2;
% Estimate model using new Options settings
M = nlarx(data,[2 2 1],wavenet('options',O))

where O specifies the values of the Options structure fields and M is the
estimated model. For more information about these and other commands,
see the corresponding reference page.

Viewing Model Properties and Estimated Parameters

To view all the properties and values of any model object, use the get
command. For example, type the following at the prompt to load sample data,
compute an ARX model, and list the model properties:

load iddata8
m_arx=arx(z8,[4 3 2 3 0 0 0]);
get(m_arx)

To access a specific property, use dot notation. For example, to view the A
matrix containing the estimated parameters in the previous model, type the
following command:

m_arx.a

1-20
 Creating Model Structures at the Command Line

ans =
1.0000 -0.8441 -0.1539 0.2278 0.1239

Similarly, to access the uncertainties in these parameter estimates, type

the following command:

m_arx.da
ans =
0 0.0357 0.0502 0.0438 0.0294

Property names are not case sensitive. You do not need to type the entire
property name if the first few letters uniquely identify the property.

To change property values for an existing model object, use the set command
or dot notation. For example, to change the input delays for all three input
channels to [1 1 1], type the following at the prompt:

set(m_arx,'nk',[1 1 1])

or equivalently

m_arx.nk = [1 1 1]

Some model properties, such as Algorithm, are structures. To access the

fields in this structure, use the following syntax:

model.algorithm.PropertyName

where PropertyName represents any of the Algorithm fields. For example,

to change the maximum number of iterations using the MaxIter property,
type the following command:

m_arx.algorithm.MaxIter=50

To verify the new property value, type the following:

m_arx.algorithm.MaxIter

Note PropertyName refers to fields in a structure and is case sensitive. You

must type the entire property name. Use the Tab key when typing property
names to get completion suggestions.

1-21
 1 Choosing Your System Identification Approach

Getting Help on Model Properties at the Command Line

If you need to learn more about model properties while working at the
command line, you can use the idprops command to list the properties and
values for each object.

Some model objects are based on the superclasses idmodel and idnlmodel
and inherit the properties of these superclasses. For such model objects, you
must independently look up the properties for both the model object and for
its superclass.

The following table summarizes the commands for getting help on object
properties.

Help Commands for Model Properties

Model Class Help Commands

idarx idprops idarx
Also inherits properties from idmodel.
idfrd idprops idfrd
idnlmodel idprops idnlmodel
idmodel idprops idmodel
idprops idmodel Algorithm
idprops idmodel EstimationInfo
Also see the Algorithm and EstimationInfo reference page.
idproc idprops idproc
Also inherits properties from idmodel.
idpoly idprops idpoly
Also inherits properties from idmodel.
idss idprops idss
Also inherits properties from idmodel.
idgrey idprops idgrey
Also inherits properties from idmodel.
idnlgrey idprops idnlgrey
idprops idnlgrey Algorithm
idprops idnlgrey EstimationInfo
Also inherits properties from idnlmodel.

1-22
 Creating Model Structures at the Command Line

Help Commands for Model Properties (Continued)

Model Class Help Commands

idnlarx idprops idnlarx
idprops idnlarx Algorithm
idprops idnlarx EstimationInfo
Also inherits properties from idnlmodel.
idnlhw idprops idnlhw
idprops idnlhw Algorithm
idprops idnlhw EstimationInfo
Also inherits properties from idnlmodel.

See Also
Validate each model directly after estimation to help fine-tune your modeling
strategy. When you do not achieve a satisfactory model, you can try a different
model structure and order, or try another identification algorithm. For more
information about validating and troubleshooting models, see “Overview of
Model Validation and Plots” on page 8-2.

1-23
 1 Choosing Your System Identification Approach

Modeling Multiple-Output Systems

In this section...
“About Modeling Multiple-Output Systems” on page 1-24
“Modeling Multiple Outputs Directly” on page 1-25
“Modeling Multiple Outputs as a Combination of Single-Output Models”
on page 1-25
“Improving Multiple-Output Estimation Results by Weighing Outputs
During Estimation” on page 1-26

About Modeling Multiple-Output Systems

You can estimate multiple-output model directly using all the inputs and
outputs, or you can try building models for subsets of the most important
input and output channels. To learn more about each approach, see:

• “Modeling Multiple Outputs Directly” on page 1-25

• “Modeling Multiple Outputs as a Combination of Single-Output Models”
on page 1-25

Modeling multiple-output systems is more challenging because input/output

couplings require additional parameters to obtain a good fit and involve more
complex models. In general, a model is better when more data inputs are
included during modeling. Including more outputs typically leads to worse
simulation results because it is harder to reproduce the behavior of several
outputs simultaneously.

If you know that some of the outputs have poor accuracy and should be
less important during estimation, you can control how much each output
is weighed in the estimation. For more information, see “Improving
Multiple-Output Estimation Results by Weighing Outputs During
Estimation” on page 1-26.

1-24
 Modeling Multiple-Output Systems

Modeling Multiple Outputs Directly

You can estimate the following types of models for multiple-output data:

• Impulse- and step-response models

• Frequency-response models
• Linear ARX models
• State-space models
• Nonlinear ARX and Hammerstein-Wiener models
• Linear and nonlinear ODEs

Tip Estimating multiple-output state-space models directly generally

produces better results than estimating other types of multiple-output models
directly.

Modeling Multiple Outputs as a Combination of

Single-Output Models
You may find that it is harder for a single model to explain the behavior of
several outputs. If you get a poor fit estimating a multiple-output model
directly, you can try building models for subsets of the most important input
and output channels.

Use this approach when no feedback is present in the dynamic system and
there are no couplings between the outputs. If you are unsure about the
presence of feedback, see “How to Analyze Data Using the advice Command”
on page 2-87.

To construct partial models, use subreferencing to create partial data sets,

such that each data set contains all inputs and one output. For more
information about creating partial data sets, see the following sections in the
System Identification Toolbox User’s Guide:

• For working in the System Identification Tool GUI, see “Creating Data Sets
from a Subset of Signal Channels” on page 2-33.

1-25
 1 Choosing Your System Identification Approach

• For working at the command line, see the “Subreferencing iddata Objects”
on page 2-57.

After validating the single-output models, use vertical concatenation to

combine these partial models into a single multiple-output model. For more
information about concatenation, see “Concatenating iddata Objects” on page
2-66 or “Concatenating idfrd Objects” on page 2-73.

You can try refining the concatenated multiple-output model using the
original (multiple-output) data set.

Improving Multiple-Output Estimation Results by

Weighing Outputs During Estimation
When estimating linear and nonlinear black-box models for multiple-output
systems, you can control the relative importance of output channels during
the estimation process. The ability to control how much each output is
weighed during estimation is useful when some of the measured outputs
have poor accuracy or should be treated as less important during estimation.
For example, if you have already modeled one output well, you might want
to focus the estimation on modeling the remaining outputs. Similarly, you
might want to refine a model for a subset of outputs.

You can specify output weights directly in the estimation command using
the Criterion and Weighting fields of the Algorithm property. You must
set the Criterion field to Trace, and set the Weighting field to the matrix
that contains the output weights. The Trace criterion minimizes the weighted
sum of the prediction errors using the weights specified by Weighting.

The following code snippet shows how to specify the Criterion and Weighting
Algorithm fields as part of the pem command:

model=pem(z,2,'criterion','trace','weighting',diag(Q,1))

where Q is a vector of positive values and the higher values for outputs to be
emphasized more during estimation.

You set Weighting to a positive semi-definite symmetric matrix of size equal

to number of outputs. By default, Weighting is an identity matrix, which
means that all outputs are weighed equally during estimation.

1-26
 Modeling Multiple-Output Systems

For more information about these Algorithm fields for linear estimation, see
the Algorithm Properties reference page. For more information about
the Algorithm fields for nonlinear estimation, see the idnlarx and idnlhw
reference pages.

Note For multiple-output idnlarx models containing neuralnet or

treepartition nonlinearity estimators, output weighting is ignored because
each output is estimated independently.

1-27
 1 Choosing Your System Identification Approach

1-28
 2

Data Import and Processing

• “Types of Data You Can Model” on page 2-2

• “Ways to Process Data for System Identification” on page 2-4
• “Requirements on Data Sampling” on page 2-6
• “Importing Data into the MATLAB Workspace” on page 2-7
• “Importing Data into the GUI” on page 2-14
• “Representing Time- and Frequency-Domain Data Using iddata Objects”
on page 2-49
• “Representing Frequency-Response Data Using idfrd Objects” on page 2-69
• “Analyzing Data Quality” on page 2-78
• “Selecting Subsets of Data” on page 2-89
• “Handling Missing Data and Outliers” on page 2-93
• “Handling Offsets and Trends in Data” on page 2-97
• “Resampling Data” on page 2-102
• “Filtering Data” on page 2-109
• “Generating Data Using Simulation” on page 2-117
• “Transforming Between Time- and Frequency-Domain Data” on page 2-121
• “Manipulating Complex-Valued Data” on page 2-133
 2 Data Import and Processing

Types of Data You Can Model

System Identification Toolbox software supports estimation of linear models
from both time- and frequency-domain data. For nonlinear models, this
toolbox supports only time-domain data.

The data can have single or multiple inputs and outputs, and can be either
real or complex.

Your data should be sampled at discrete and uniformly spaced time instants
to obtain an input sequence

u={u(T),u(2T),...,u(NT)}

and a corresponding output sequence

y={y(T),y(2T),...,y(NT)}

u(t) and y(t) are the values of the input and output signals at time t,
respectively.

This toolbox supports modeling both single- or multiple-channel input-output

data or time-series data.

Supported Data Description

Time-domain I/O data One or more input variables u(t) and one
or more output variables y(t), sampled as a
function of time. Time-domain data can be
either real or complex
Time-series data Contains one or more outputs y(t) and no
measured input. Can be time-domain or
frequency-domain data.

2-2
 Types of Data You Can Model

Supported Data Description

Frequency-domain data Fourier transform of the input and output
time-domain signals.
Frequency-response data Complex frequency-response values for a
linear system characterized by its transfer
function G, measurable directly using a
spectrum analyzer. Also called frequency
function data.

Note If your data is complex valued, see “Manipulating Complex-Valued

Data” on page 2-133 for information about supported operations for complex
data.

2-3
 2 Data Import and Processing

Ways to Process Data for System Identification

Before you can perform any task in this toolbox, your data must be in the
MATLAB workspace. You can import the data from external data files or
manually create data arrays at the command line. For more information
about importing data, see “Importing Data into the MATLAB Workspace”
on page 2-7.

The following tasks help to prepare your data for identifying models from data:

Represent data for system identification

You can represent data in the format of this toolbox by doing one of the
following:

• For working in the GUI, import data into the System Identification Tool
GUI.
See “Importing Data into the GUI” on page 2-14.
• For working at the command line, create an iddata or idfrd object.
For time-domain or frequency-domain data, see “Representing Time- and
Frequency-Domain Data Using iddata Objects” on page 2-49.
For frequency-response data, see “Representing Frequency-Response Data
Using idfrd Objects” on page 2-69.
• To simulate data with and without noise, see “Generating Data Using
Simulation” on page 2-117.

Analyze data quality

You can analyze your data by doing either of the following:

• Plotting data to examine both time- and frequency-domain behavior.

See “Analyzing Data Quality” on page 2-78.
• Using the advice command to analyze the data for the presence of constant
offsets and trends, delay, possible feedback, and signal excitation levels.
See “How to Analyze Data Using the advice Command” on page 2-87.

2-4
 Ways to Process Data for System Identification

Preprocess data

Review the data characteristics for any of the following features to determine
if there is a need for preprocessing:

• Missing or faulty values (also known as outliers). For example, you might
see gaps that indicate missing data, values that do not fit with the rest of
the data, or noninformative values.
See “Handling Missing Data and Outliers” on page 2-93.
• Offsets and drifts in signal levels (low-frequency disturbances).
See “Handling Offsets and Trends in Data” on page 2-97 for information
about subtracting means and linear trends, and “Filtering Data” on page
2-109 for information about filtering.
• High-frequency disturbances above the frequency interval of interest for
the system dynamics.
See “Resampling Data” on page 2-102 for information about decimating and
interpolating values, and “Filtering Data” on page 2-109 for information
about filtering.

Select a subset of your data

You can use data selection as a way to clean the data and exclude parts
with noisy or missing information. You can also use data selection to create
independent data sets for estimation and validation.

To learn more about selecting data, see “Selecting Subsets of Data” on page
2-89.

Combine data from multiple experiments

You can combine data from several experiments into a single data set. The
model you estimate from a data set containing several experiments describes
the average system that represents these experiments.

To learn more about creating multiple-experiment data sets, see “Creating

Multiexperiment Data Sets in the GUI” on page 2-35 or “Creating
Multiexperiment Data at the Command Line” on page 2-55.

2-5
 2 Data Import and Processing

Requirements on Data Sampling

A sampling interval is the time between successive data samples.

The System Identification Tool GUI only supports uniformly sampled data.

The System Identification Toolbox product provides limited support for

nonuniformly sampled data. For more information about specifying uniform
and nonuniform time vectors, see “Constructing an iddata Object for
Time-Domain Data” on page 2-50.

2-6
 Importing Data into the MATLAB® Workspace

Importing Data into the MATLAB Workspace

In this section...
“Importing Time-Domain Data into MATLAB” on page 2-7
“Importing Time-Series Data into MATLAB” on page 2-8
“Importing Frequency-Domain Data into MATLAB” on page 2-9
“Importing Frequency-Response Data into MATLAB” on page 2-11

Importing Time-Domain Data into MATLAB

Time-domain data consists of one or more input variables u(t) and one or more
output variables y(t), sampled as a function of time. If there is no output data,
see “Importing Time-Series Data into MATLAB” on page 2-8.

You must import your time-domain data into the MATLAB workspace as
the following variables:

• Input data
- For single-input/single-output (SISO) data, the input must be a column
vector.
- For a data set with Nu inputs and NT samples (measurements), the input
is an NT-by-Nu matrix.
• Output data
- For single-input/single-output (SISO) data, the output must be a column
vector.
- For a data set with Ny outputs and NT samples (measurements), the
output is an NT-by-Ny matrix.

• Sampling time interval

If you are working with uniformly sampled data, use the actual sampling
interval from your experiment. Each data value is assigned a sample time,
which is calculated from the start time and the sampling interval. If you
are working with nonuniformly sampled data at the command line, you
can specify a vector of time instants using the iddata SamplingInstants

2-7
 2 Data Import and Processing

property, as described in “Constructing an iddata Object for Time-Domain

Data” on page 2-50.

For more information about importing data into the MATLAB workspace, see
MATLAB Data Import and Export.

After you import data, you can import it into the System Identification Tool
GUI or create a data object for working at the command line. For more
information about importing data into the GUI, see “Importing Time-Domain
Data into the GUI” on page 2-16. To learn more about creating a data object,
see “Representing Time- and Frequency-Domain Data Using iddata Objects”
on page 2-49.

Importing Time-Series Data into MATLAB

Time-series data is time-domain or frequency-domain data that consist of one
or more outputs y(t) with no corresponding input.

You must import your time-series data into the MATLAB workspace as the
following variables:

• Output data
- For single-input/single-output (SISO) data, the output must be a column
vector.
- For a data set with Ny outputs and NT samples (measurements), the
output is an NT-by-Ny matrix.

• Sampling time interval

- If you are working with uniformly sampled data, use the actual
sampling interval in your experiment. Each data value is assigned a
sample time, which is calculated from the start time and the sampling
interval. If you are working with nonuniformly sampled data at the
command line, you can specify a vector of time instants using the iddata
SamplingInstants property, as described in “Constructing an iddata
Object for Time-Domain Data” on page 2-50.

For more information about importing data into the MATLAB workspace, see
MATLAB Data Import and Export.

2-8
 Importing Data into the MATLAB® Workspace

After you import data, you can import it into the System Identification Tool
GUI or create a data object for working at the command line. For more
information about importing data into the GUI, see “Importing Time-Domain
Data into the GUI” on page 2-16. To learn more about creating a data object,
see “Representing Time- and Frequency-Domain Data Using iddata Objects”
on page 2-49.

For information about estimating time-series model parameters, see Chapter

6, “Time Series Identification”.

Importing Frequency-Domain Data into MATLAB

• “What Is Frequency-Domain Data?” on page 2-9
• “How to Import Frequency-Domain Data into MATLAB” on page 2-10

What Is Frequency-Domain Data?

Frequency-domain data is the Fourier transform of the input and output
time-domain signals. For continuous-time signals, the Fourier transform over
the entire time axis is defined as follows:

∞
Y (iw) = ∫ y(t) e−iwt dt
−∞
∞
− iwt
U (iw) = 1
2π ∫ u(t)e dt
−∞

In the context of numerical computations, continuous equations are replaced

by their discretized equivalents to handle discrete data values. For a
discrete-time system with a sampling interval T, the frequency-domain output
Y(eiw) and input U(eiw) is the time-discrete Fourier transform (TDFT):

N
Y (eiwT ) = T ∑ y(kT ) e−iwkT
k=1

In this example, k = 1,2,...,N, where N is the number of samples in the

sequence.

2-9
 2 Data Import and Processing

Note This form only discretizes the time. The frequency is continuous.

In practice, the Fourier transform cannot be handled for all continuous

frequencies and you must specify a finite number of frequencies. The discrete
Fourier transform (DFT) of time-domain data for N equally spaced frequencies
between 0 and the sampling frequency 2π/N is:

N
Y (eiwnT ) = ∑ y(kT )e−iw kT
n

k=1
2πn
wn = n = 0, 1, 2,…, N − 1
T

The DFT is useful because it can be calculated very efficiently using the fast
Fourier transform (FFT) method. Fourier transforms of the input and output
data are complex numbers.

How to Import Frequency-Domain Data into MATLAB

You must import your frequency-domain data as the following variables:

• Input data
- For single-input/single-output (SISO) data, the input u must be a column

(
vector containing the values u ei kT ) for k=1, 2, ..., N , where N is the
f f
number of frequencies.
- For a data set with Nu inputs and Nf frequencies, the input is an
Nf-by-Nu matrix.
• Output data
- For single-input/single-output (SISO) data, the output y must be a

(
column vector containing the values y ei kT ) for k=1, 2, ..., N , where
f
Nf is the number of frequencies.
- For a data set with Ny outputs and Nf frequencies, the output is an
Nf-by-Ny matrix.

2-10
 Importing Data into the MATLAB® Workspace

• Frequency values
Must be a column vector.

For more information about importing data into the MATLAB workspace, see
MATLAB Data Import and Export.

After you import data, you can import it into the System Identification
Tool GUI or create a data object for working at the command line. For
more information about importing data into the GUI, see “Importing
Frequency-Domain Data into the GUI” on page 2-19. To learn more about
creating a data object, see “Representing Time- and Frequency-Domain Data
Using iddata Objects” on page 2-49.

Importing Frequency-Response Data into MATLAB

• “What Is Frequency-Response Data?” on page 2-11
• “How to Import Frequency-Response Data into the Software” on page 2-12

What Is Frequency-Response Data?

Frequency-response data, also called frequency-function data, consists of
complex frequency-response values for a linear system characterized by
its transfer function G. Frequency-response data tells you how the system
handles sinusoidal inputs. You can measure frequency-response data values
directly using a spectrum analyzer, for example, which provides a compact
representation of the input-output relationship (compared to storing input
and output independently).

The transfer function G is an operator that takes the input u of a linear

system to the output y:

y = Gu

For a continuous-time system, the transfer function relates the Laplace

transforms of the input U(s) and output Y(s):

Y (s) = G(s)U (s)

2-11
 2 Data Import and Processing

In this case, the frequency function G(iw) is the transfer function evaluated
on the imaginary axis s=iw.

For a discrete-time system sampled with a time interval T, the transfer

function relates the Z-transforms of the input U(z) and output Y(z):

Y ( z) = G( z)U ( z)

In this case, the frequency function G(eiwT) is the transfer function G(z)
evaluated on the unit circle. The argument of the frequency function G(eiwT)
is scaled by the sampling interval T to make the frequency function periodic
with the sampling frequency 2π T .

When the input to the system is a sinusoid of a specific frequency, the output
is also a sinusoid with the same frequency. The amplitude of the output is G
times the amplitude of the input. The phase of the shifted from the input by
ϕ = arg G . G is evaluated at the frequency of the input sinusoid.

Frequency-response data represents a (nonparametric) model of the

relationship between the input and the outputs as a function of frequency.
You might use such a model, which consists of a table or plot of values, to
study the system frequency response. However, this model is not suitable for
simulation and prediction. You should create parametric model from the
frequency-response data.

How to Import Frequency-Response Data into the Software

There are two ways to represent frequency-response data for system
identification. The first approach lets you manipulate the data using both
System Identification Tool GUI and the command line. The second approach
is only used for working with data in the System Identification Tool GUI.

You must import your frequency-response data into the MATLAB workspace
as the following variables:

• In System Identification Tool GUI or MATLAB Command Window,

represent complex-valued G(eiw).

2-12
 Importing Data into the MATLAB® Workspace

For single-input single-output (SISO) systems, the frequency function is

a column vector.
For a data set with Nu inputs, Ny outputs, and Nf frequencies, the frequency
function is an Ny-by-Nu-by-Nf array.

• In System Identification Tool GUI only, represent amplitude G and phase

shift ϕ = arg G .
For single-input single-output (SISO) systems, the amplitude and the
phase must each be a column vector.
For a data set with Nu inputs, Ny outputs, and Nf frequencies, the amplitude
and the phase must each be an Ny-by-Nu-by-Nf array.
• Frequency values must be a column vector.

For more information about importing data into the MATLAB workspace, see
MATLAB Data Import and Export.

After you import data into the MATLAB workspace, you can import it into
the System Identification Tool GUI or create a data object for working at the
command line. For more information about importing data into the GUI, see
“Importing Frequency-Response Data into the GUI” on page 2-22. To learn
more about creating a data object, see “Representing Frequency-Response
Data Using idfrd Objects” on page 2-69.

2-13
 2 Data Import and Processing

Importing Data into the GUI

In this section...
“Types of Data You Can Import into the GUI” on page 2-14
“Importing Time-Domain Data into the GUI” on page 2-16
“Importing Frequency-Domain Data into the GUI” on page 2-19
“Importing Frequency-Response Data into the GUI” on page 2-22
“Importing Data Objects into the GUI” on page 2-26
“Specifying the Data Sampling Interval” on page 2-30
“Specifying Estimation and Validation Data” on page 2-31
“Preprocessing Data Using Quick Start” on page 2-32
“Creating Data Sets from a Subset of Signal Channels” on page 2-33
“Creating Multiexperiment Data Sets in the GUI” on page 2-35
“Viewing Data Properties” on page 2-42
“Renaming Data and Changing Display Color” on page 2-43
“Distinguishing Data Types in the GUI” on page 2-45
“Organizing Data Icons” on page 2-45
“Deleting Data Sets in the GUI” on page 2-46
“Exporting Data from the GUI to the MATLAB Workspace” on page 2-47

Types of Data You Can Import into the GUI

You can import the following types of data from the MATLAB workspace into
the System Identification Tool GUI:

• “Importing Time-Domain Data into the GUI” on page 2-16

• “Importing Frequency-Domain Data into the GUI” on page 2-19
• “Importing Frequency-Response Data into the GUI” on page 2-22
• “Importing Data Objects into the GUI” on page 2-26

2-14
 Importing Data into the GUI

To open the GUI, type the following command in the MATLAB Command
Window:

ident

In the Import data list, select the type of data to import from the MATLAB
workspace, as shown in the following figure.

For an example of importing data into the System Identification Tool GUI, see
the Getting Started documentation.

2-15
 2 Data Import and Processing

Importing Time-Domain Data into the GUI

Before you can import time-domain data into the System Identification Tool
GUI, you must import the data into the MATLAB workspace, as described in
“Importing Time-Domain Data into MATLAB” on page 2-7.

Note Your time-domain data must be sampled at equal time intervals. The
input and output signals must have the same number of data samples.

To import data into the GUI:

1 Type the following command in the MATLAB Command Window to open

the GUI:

ident

2 In the System Identification Tool window, select Import data > Time
domain data. This action opens the Import Data dialog box.

2-16
 Importing Data into the GUI

3 Specify the following options:

Note For time series, only import the output signal and enter [] for the
input.

• Input — Enter the MATLAB variable name (column vector or matrix)

or a MATLAB expression that represents the input data. The expression
must evaluate to a column vector or matrix.
• Output — Enter the MATLAB variable name (column vector or
matrix) or a MATLAB expression that represents the output data. The
expression must evaluate to a column vector or matrix.
• Data name — Enter the name of the data set, which appears in
the System Identification Tool window after the import operation is
completed.
• Starting time — Enter the starting value of the time axis for time plots.
• Sampling interval — Enter the actual sampling interval in the
experiment. For more information about this setting, see “Specifying the
Data Sampling Interval” on page 2-30.

Tip The System Identification Toolbox product uses the sampling

interval during model estimation and to set the horizontal axis on time
plots. If you transform a time-domain signal to a frequency-domain
signal, the Fourier transforms are computed as discrete Fourier
transforms (DFTs) using this sampling interval.

2-17
 2 Data Import and Processing

4 (Optional) In the Data Information area, click More to expand the dialog
box and enter the following settings:

Input Properties
• InterSample — This options specifies the behavior of the input signals
between samples during data acquisition. It is used when transforming
models from discrete-time to continuous-time and when resampling the
data.
– zoh (zero-order hold) indicates that the input was piecewise-constant
during data acquisition.
– foh (first-order hold) indicates that the output was piecewise-linear
during data acquisition.
– bl (bandwidth-limited behavior) specifies that the continuous-time
input signal has zero power above the Nyquist frequency (equal to the
inverse of the sampling interval).

Note See the d2c and c2d reference pages for more information about
transforming between discrete-time and continuous-time models.

• Period — Enter Inf to specify a nonperiodic input. If the underlying

time-domain data was periodic over an integer number of periods, enter
the period of the input signal.

Note If your data is periodic, always include a whole number of periods

for model estimation.

Channel Names
• Input — Enter a string to specify the name of one or more input
channels.

2-18
 Importing Data into the GUI

Tip Naming channels helps you to identify data in plots. For

multivariable input-output signals, you can specify the names of
individual Input and Output channels, separated by commas.

• Output — Enter a string to specify the name of one or more output

channels.

Physical Units of Variables

• Input — Enter a string to specify the input units.

Tip When you have multiple inputs and outputs, enter a

comma-separated list of Input and Output units corresponding to each
channel.

• Output — Enter a string to specify the output units.

Notes — Enter comments about the experiment or the data. For

example, you might enter the experiment name, date, and a description
of experimental conditions. Models you estimate from this data inherit
your data notes.

5 Click Import. This action adds a new data icon to the System Identification
Tool window.

6 Click Close to close the Import Data dialog box.

Importing Frequency-Domain Data into the GUI

Frequency-domain data consists of Fourier transforms of time-domain data (a
function of frequency).

Before you can import frequency-domain data into the System Identification
Tool GUI, you must import the data into the MATLAB workspace, as
described in “Importing Frequency-Domain Data into MATLAB” on page 2-9.

2-19
 2 Data Import and Processing

Note The input and output signals must have the same number of data
samples.

To import data into the GUI:

1 Type the following command in the MATLAB Command Window to open

the GUI:

ident

2 In the System Identification Tool window, select Import data > Freq.
domain data. This action opens the Import Data dialog box.

3 Specify the following options:

• Input — Enter the MATLAB variable name (column vector or matrix)

or a MATLAB expression that represents the input data. The expression
must evaluate to a column vector or matrix.
• Output — Enter the MATLAB variable name (column vector or
matrix) or a MATLAB expression that represents the output data. The
expression must evaluate to a column vector or matrix.
• Frequency — Enter the MATLAB variable name of a vector or a
MATLAB expression that represents the frequencies. The expression
must evaluate to a column vector.
The frequency vector must have the same number of rows as the input
and output signals.
• Data name — Enter the name of the data set, which appears in
the System Identification Tool window after the import operation is
completed.
• Frequency unit — Enter Hz for Hertz or keep the rad/s default value.
• Sampling interval — Enter the actual sampling interval in the
experiment. For continuous-time data, enter 0. For more information
about this setting, see “Specifying the Data Sampling Interval” on page
2-30.

2-20
 Importing Data into the GUI

4 (Optional) In the Data Information area, click More to expand the dialog
box and enter the following optional settings:

Input Properties
• InterSample — This options specifies the behavior of the input signals
between samples during data acquisition. It is used when transforming
models from discrete-time to continuous-time and when resampling the
data.
– zoh (zero-order hold) indicates that the input was piecewise-constant
during data acquisition.
– foh (first-order hold) indicates that the output was piecewise-linear
during data acquisition.
– bl (bandwidth-limited behavior) specifies that the continuous-time
input signal has zero power above the Nyquist frequency (equal to the
inverse of the sampling interval).

Note See the d2c and c2d reference page for more information about
transforming between discrete-time and continuous-time models.

• Period — Enter Inf to specify a nonperiodic input. If the underlying

time-domain data was periodic over an integer number of periods, enter
the period of the input signal.

Note If your data is periodic, always include a whole number of periods

for model estimation.

Channel Names
• Input — Enter a string to specify the name of one or more input
channels.

2-21
 2 Data Import and Processing

Tip Naming channels helps you to identify data in plots. For

multivariable input and output signals, you can specify the names of
individual Input and Output channels, separated by commas.

• Output — Enter a string to specify the name of one or more output

channels.

Physical Units of Variables

• Input — Enter a string to specify the input units.

Tip When you have multiple inputs and outputs, enter a

comma-separated list of Input and Output units corresponding to each
channel.

• Output — Enter a string to specify the output units.

Notes — Enter comments about the experiment or the data. For

example, you might enter the experiment name, date, and a description
of experimental conditions. Models you estimate from this data inherit
your data notes.

5 Click Import. This action adds a new data icon to the System Identification
Tool window.

6 Click Close to close the Import Data dialog box.

Importing Frequency-Response Data into the GUI

• “Prerequisite” on page 2-23
• “Importing Complex-Valued Frequency-Response Data” on page 2-23
• “Importing Amplitude and Phase Frequency-Response Data” on page 2-24

2-22
 Importing Data into the GUI

Prerequisite
Before you can import frequency-response data into the System Identification
Tool GUI, you must import the data into the MATLAB workspace, as described
in “Importing Frequency-Response Data into MATLAB” on page 2-11.

Importing Complex-Valued Frequency-Response Data

To import frequency-response data consisting of complex-valued frequency
values at specified frequencies:

1 Type the following command in the MATLAB Command Window to open

the GUI:

ident

2 In the System Identification Tool window, select Import data > Freq.
domain data. This action opens the Import Data dialog box.

3 In the Data Format for Signals list, select Freq. Function (Complex).

4 Specify the following options:

• Freq. Func. — Enter the MATLAB variable name or a MATLAB

expression that represents the complex frequency-response data G(eiw).
• Frequency — Enter the MATLAB variable name of a vector or a
MATLAB expression that represents the frequencies. The expression
must evaluate to a column vector.
• Data name — Enter the name of the data set, which appears in
the System Identification Tool window after the import operation is
completed.
• Frequency unit — Enter Hz for Hertz or keep the rad/s default value.
• Sampling interval — Enter the actual sampling interval in the
experiment. For continuous-time data, enter 0. For more information
about this setting, see “Specifying the Data Sampling Interval” on page
2-30.

5 (Optional) In the Data Information area, click More to expand the dialog
box and enter the following optional settings:

Channel Names

2-23
 2 Data Import and Processing

• Input — Enter a string to specify the name of one or more input

channels.

Tip Naming channels helps you to identify data in plots. For

multivariable input and output signals, you can specify the names of
individual Input and Output channels, separated by commas.

• Output — Enter a string to specify the name of one or more output

channels.

Physical Units of Variables

• Input — Enter a string to specify the input units.

Tip When you have multiple inputs and outputs, enter a

comma-separated list of Input and Output units corresponding to each
channel.

• Output — Enter a string to specify the output units.

Notes — Enter comments about the experiment or the data. For

example, you might enter the experiment name, date, and a description
of experimental conditions. Models you estimate from this data inherit
your data notes.

6 Click Import. This action adds a new data icon to the System Identification
Tool window.

7 Click Close to close the Import Data dialog box.

Importing Amplitude and Phase Frequency-Response Data

To import frequency-response data consisting of amplitude and phase values
at specified frequencies:

1 Type the following command in theMATLAB Command Window to open

the GUI:

2-24
 Importing Data into the GUI

ident

2 In the System Identification Tool window, select Import data > Freq.
domain data. This action opens the Import Data dialog box.

3 In the Data Format for Signals list, select Freq. Function

(Amp/Phase).

4 Specify the following options:

• Amplitude — Enter the MATLAB variable name or a MATLAB

expression that represents the amplitude G .
• Phase (deg) — Enter the MATLAB variable name or a MATLAB
expression that represents the phase ϕ = arg G .
• Frequency — Enter the MATLAB variable name of a vector or a
MATLAB expression that represents the frequencies. The expression
must evaluate to a column vector.
• Data name — Enter the name of the data set, which appears in
the System Identification Tool window after the import operation is
completed.
• Frequency unit — Enter Hz for Hertz or keep the rad/s default value.
• Sampling interval — Enter the actual sampling interval in the
experiment. For continuous-time data, enter 0. For more information
about this setting, see “Specifying the Data Sampling Interval” on page
2-30.

5 (Optional) In the Data Information area, click More to expand the dialog
box and enter the following optional settings:

Channel Names
• Input — Enter a string to specify the name of one or more input
channels.

2-25
 2 Data Import and Processing

Tip Naming channels helps you to identify data in plots. For

multivariable input and output signals, you can specify the names of
individual Input and Output channels, separated by commas.

• Output — Enter a string to specify the name of one or more output

channels.

Physical Units of Variables

• Input — Enter a string to specify the input units.

Tip When you have multiple inputs and outputs, enter a

comma-separated list of Input and Output units corresponding to each
channel.

• Output — Enter a string to specify the output units.

Notes — Enter comments about the experiment or the data. For

example, you might enter the experiment name, date, and a description
of experimental conditions. Models you estimate from this data inherit
your data notes.

6 Click Import. This action adds a new data icon to the System Identification
Tool window.

7 Click Close to close the Import Data dialog box.

Importing Data Objects into the GUI

You can import the System Identification Toolbox iddata and idfrd data
objects into the System Identification Tool GUI.

Before you can import a data object into the System Identification Tool GUI,
you must create the data object in the MATLAB workspace, as described in
“Representing Time- and Frequency-Domain Data Using iddata Objects” on
page 2-49 or “Representing Frequency-Response Data Using idfrd Objects”
on page 2-69.

2-26
 Importing Data into the GUI

Note You can also import a Control System Toolbox™ frd object. Importing
an frd object converts it to an idfrd object.

Select Import data > Data object to open the Import Data dialog box.

Import iddata, idfrd, or frd data object in the MATLAB workspace.

To import a data object into the GUI:

1 Type the following command in the MATLAB Command Window to open

the GUI:

ident

2 In the System Identification Tool window, select Import data > Data
object.

This action opens the Import Data dialog box. IDDATA or IDFRD/FRD is
already selected in the Data Format for Signals list.

2-27
 2 Data Import and Processing

3 Specify the following options:

• Object — Enter the name of the MATLAB variable that represents the
data object in the MATLAB workspace. Press Enter.
• Data name — Enter the name of the data set, which appears in
the System Identification Tool window after the import operation is
completed.
• (Only for time-domain iddata object) Starting time — Enter the
starting value of the time axis for time plots.
• (Only for frequency domain iddata or idfrd object) Frequency unit
— Enter the frequency unit for response plots.
• Sampling interval — Enter the actual sampling interval in the
experiment. For more information about this setting, see “Specifying the
Data Sampling Interval” on page 2-30.

Tip The System Identification Toolbox product uses the sampling

interval during model estimation and to set the horizontal axis on time
plots. If you transform a time-domain signal to a frequency-domain
signal, the Fourier transforms are computed as discrete Fourier
transforms (DFTs) using this sampling interval.

4 (Optional) In the Data Information area, click More to expand the dialog
box and enter the following optional settings:

(Only for iddata object) Input Properties

2-28
 Importing Data into the GUI

• InterSample — This options specifies the behavior of the input signals

between samples during data acquisition. It is used when transforming
models from discrete-time to continuous-time and when resampling the
data.
– zoh (zero-order hold) indicates that the input was piecewise-constant
during data acquisition.
– foh (first-order hold) indicates that the input was piecewise-linear
during data acquisition.
– bl (bandwidth-limited behavior) specifies that the continuous-time
input signal has zero power above the Nyquist frequency (equal to the
inverse of the sampling interval).

Note See the d2c and c2d reference page for more information about
transforming between discrete-time and continuous-time models.

• Period — Enter Inf to specify a nonperiodic input. If the underlying

time-domain data was periodic over an integer number of periods, enter
the period of the input signal.

Note If your data is periodic, always include a whole number of periods

for model estimation.

Channel Names
• Input — Enter a string to specify the name of one or more input
channels.

Tip Naming channels helps you to identify data in plots. For

multivariable input and output signals, you can specify the names of
individual Input and Output channels, separated by commas.

• Output — Enter a string to specify the name of one or more output

channels.

2-29
 2 Data Import and Processing

Physical Units of Variables

• Input — Enter a string to specify the input units.

Tip When you have multiple inputs and outputs, enter a

comma-separated list of Input and Output units corresponding to each
channel.

• Output — Enter a string to specify the output units.

Notes — Enter comments about the experiment or the data. For

example, you might enter the experiment name, date, and a description
of experimental conditions. Models you estimate from this data inherit
your data notes.

5 Click Import. This action adds a new data icon to the System Identification
Tool window.

6 Click Close to close the Import Data dialog box.

Specifying the Data Sampling Interval

When you import data into the GUI, you must specify the data sampling
interval.

The sampling interval is the time between successive data samples in your
experiment and must be the numerical time interval at which your data is
sampled in any units. For example, enter 0.5 if your data was sampled every
0.5 s, and enter 1 if your data was sampled every 1 s.

You can also use the sampling interval as a flag to specify continuous-time
data. When importing continuous-time frequency domain or
frequency-response data, set the Sampling interval to 0.

The sampling interval is used during model estimation. For time-domain

data, the sampling interval is used together with the start time to calculate
the sampling time instants. When you transform time-domain signals to
frequency-domain signals (see the fft reference page), the Fourier transforms
are computed as discrete Fourier transforms (DFTs) for this sampling

2-30
 Importing Data into the GUI

interval. In addition, the sampling instants are used to set the horizontal
axis on time plots.

Sampling Interval in the Import Data dialog box

Specifying Estimation and Validation Data

You should use different data sets to estimate and validate your model for
best validation results.

In the System Identification Tool GUI, Working Data refers to estimation

data. Similarly, Validation Data refers to the data set you use to validate a
model. For example, when you plot the model output, the input to the model
is the input signal from the validation data set. This plot compares model
output to the measured output in the validation data set. Selecting Model
resids performs residual analysis using the validation data.

To specify Working Data, drag and drop the corresponding data icon into the
Working Data rectangle, as shown in the following figure.

2-31
 2 Data Import and Processing

Similarly, to specify Validation Data, drag and drop the corresponding data
icon into the Validation Data rectangle.

Preprocessing Data Using Quick Start

As a preprocessing shortcut for time-domain data, select Preprocess > Quick
start to simultaneously perform the following four actions:

• Subtract the mean value from each channel.

Note For information about when to subtract mean values from the data,
see “Handling Offsets and Trends in Data” on page 2-97.

• Split data into two parts.

• Specify the first part as estimation data for models (or Working Data).
• Specify the second part as Validation Data.

2-32
 Importing Data into the GUI

Creating Data Sets from a Subset of Signal Channels

You can create a new data set in the System Identification Tool GUI by
extracting subsets of input and output channels from an existing data set.

To create a new data set from selected channels:

1 In the System Identification Tool GUI, drag the icon of the data from which
you want to select channels to the Working Data rectangle.

2 Select Preprocess > Select channels to open the Select Channels dialog
box.

The Inputs list displays the input channels and the Outputs list displays
the output channels in the selected data set.

2-33
 2 Data Import and Processing

3 In the Inputs list, select one or more channels in any of following ways:

• Select one channel by clicking its name.

• Select adjacent channels by pressing the Shift key while clicking the
first and last channel names.
• Select nonadjacent channels by pressing the Ctrl key while clicking
each channel name.

Tip To exclude input channels and create time-series data, clear all
selections by holding down the Ctrl key and clicking each selection. To
reset selections, click Revert.

4 In the Outputs list, select one or more channels in any of following ways:

• Select one channel by clicking its name.

• Select adjacent channels by pressing the Shift key while clicking the
first and last channel names.
• Select nonadjacent channels by pressing the Ctrl key while clicking
each channel name.

Tip To reset selections, click Revert.

5 In the Data name field, type the name of the new data set. Use a name
that is unique in the Data Board.

6 Click Insert to add the new data set to the Data Board in the System
Identification Tool GUI.

7 Click Close.

2-34
 Importing Data into the GUI

Creating Multiexperiment Data Sets in the GUI

• “Why Create Multiexperiment Data?” on page 2-35
• “Limitations on Data Sets” on page 2-35
• “Merging Data Sets” on page 2-35
• “Extracting Specific Experiments from a Multiexperiment Data Set into a
New Data Set” on page 2-39

Why Create Multiexperiment Data?

You can create a time-domain or frequency-domain data set in the System
Identification Tool GUI that includes several experiments. Identifying models
for multiexperiment data results in an average model.

Experiments can mean data that was collected during different sessions, or
portions of the data collected during a single session. In the latter situation,
you can create multiexperiment data by splitting a single data set into
multiple segments that exclude corrupt data, and then merge the good data
segments.

Limitations on Data Sets

You can only merge data sets that have all of the following characteristics:

• Same number of input and output channels.

• Different names. The name of each data set becomes the experiment name
in the merged data set.
• Same input and output channel names.
• Same data domain (that is, time-domain data or frequency-domain data
only).

Merging Data Sets

You can merge data sets using the System Identification Tool GUI.

2-35
 2 Data Import and Processing

For example, suppose that you want to combine the data sets tdata, tdata2,
tdata3, tdata4 shown in the following figure.

GUI Contains Four Data Sets to Merge

2-36
 Importing Data into the GUI

To merge data sets in the GUI:

1 In the Operations area, select <--Preprocess > Merge experiments

from the drop-down menu to open the Merge Experiments dialog box.

2-37
 2 Data Import and Processing

2 In the System Identification Tool window, drag a data set icon to the Merge
Experiments dialog box (to the drop them here to be merged rectangle).

The name of the data set is added to the List of sets.

tdata and tdata2 to Be Merged

Tip To empty the list, click Revert.

3 Repeat step 2 for each data set you want to merge. Go to the next step
after adding data sets.

4 In the Data name field, type the name of the new data set. This name
must be unique in the Data Board.

2-38
 Importing Data into the GUI

5 Click Insert to add the new data set to the Data Board in the System
Identification Tool window.

Data Board Now Contains tdatam with Merged Experiments

6 Click Close to close the Merge Experiments dialog box.

Tip To get information about a data set in the System Identification Tool
GUI, right-click the data icon to open the Data/model Info dialog box.

Extracting Specific Experiments from a Multiexperiment Data

Set into a New Data Set
When a data set already consists of several experiments, you can extract
one or more of these experiments into a new data set, using the System
Identification Tool GUI.

For example, suppose that tdatam consists of four experiments.

To create a new data set that includes only the first and third experiments
in this data set:

2-39
 2 Data Import and Processing

1 In the System Identification Tool window, drag and drop the tdatam data
icon to the Working Data rectangle.

tdatam Is Set to Working Data

2 In the Operations area, select Preprocess > Select experiments from

the drop-down menu to open the Select Experiment dialog box.

2-40
 Importing Data into the GUI

3 In the Experiments list, select one or more data sets in either of the
following ways:
• Select one data set by clicking its name.
• Select adjacent data sets by pressing the Shift key while clicking the
first and last names.
• Select nonadjacent data sets by pressing the Ctrl key while clicking
each name.

4 In the Data name field, type the name of the new data set. This name
must be unique in the Data Board.

5 Click Insert to add the new data set to the Data Board in the System
Identification Tool GUI.

6 Click Close to close the Select Experiment dialog box.

2-41
 2 Data Import and Processing

Viewing Data Properties

You can get information about each data set in the System Identification Tool
GUI by right-clicking the corresponding data icon.

The Data/model Info dialog box opens. This dialog box describes the contents
and the properties of the corresponding data set. It also displays any
associated notes and the command-line equivalent of the operations you used
to create this data.

Tip To view or modify properties for several data sets, keep this window
open and right-click each data set in the System Identification Tool GUI. The
Data/model Info dialog box updates as you select each data set.

Data object
description

History of
syntax that
created this
object

2-42
 Importing Data into the GUI

To displays the data properties in the MATLAB Command Window, click

Present.

Renaming Data and Changing Display Color

You can rename data and change its display color by double-clicking the data
icon in the System Identification Tool GUI.

The Data/model Info dialog box opens. This dialog box describes both the
contents and the properties of the data. The object description area displays
the syntax of the operations you used to create the data in the GUI.

The Data/model Info dialog box also lets you rename the data by entering a
new name in the Data name field.

You can also specify a new display color using three RGB values in the Color
field. Each value is between 0 to 1 and indicates the relative presence of
red, green, and blue, respectively. For more information about specifying
default data color, see “Customizing the System Identification Tool GUI” on
page 11-20.

Tip As an alternative to using three RGB values, you can enter any one of
the following letters in single quotes:

'y' 'r' 'b' 'c' 'g' 'm' 'k'

These strings represent yellow, red, blue, cyan, green, magenta, and black,
respectively.

2-43
 2 Data Import and Processing

Data object
description

History of
syntax that
created this
object

Information About the Data

You can enter comments about the origin and state of the data in the Diary
And Notes area. For example, you might want to include the experiment
name, date, and the description of experimental conditions. When you
estimate models from this data, these notes are associated with the models.

Clicking Present display portions of this information in the MATLAB

Command Window.

2-44
 Importing Data into the GUI

Distinguishing Data Types in the GUI

The background color of a data icon is color-coded, as follows:

• White background represents time-domain data.

• Blue background represents frequency-domain data.
• Yellow background represents frequency-response data.

Time-domain data

Frequency-domain
data

Frequency-response
data

Colors Representing Type of Data

Organizing Data Icons

You can rearrange data icons in the System Identification Tool GUI by
dragging and dropping the icons to empty Data Board rectangles in the GUI.

Note You cannot drag and drop a data icon into the model area on the right.

When you need additional space for organizing data or model icons, select
Options > Extra model/data board in the System Identification Tool GUI.
This action opens an extra session window with blank rectangles for data and
models. The new window is an extension of the current session and does
not represent a new session.

2-45
 2 Data Import and Processing

Tip When you import or create data sets and there is insufficient space for
the icons, an additional session window opens automatically.

You can drag and drop data between the main System Identification Tool
GUI and any extra session windows.

Type comments in the Notes field to describe the data sets. When you save a
session, as described in “Saving, Merging, and Closing Sessions” on page 11-6,
all additional windows and notes are also saved.

Deleting Data Sets in the GUI

To delete data sets in the System Identification Tool GUI, drag and drop the
corresponding icon into Trash. Moving items to Trash does not permanently
delete these items.

Note You cannot delete a data set that is currently designated as Working
Data or Validation Data. You must first specify a different data set in the
System Identification Tool GUI to be Working Data or Validation Data, as
described in “Specifying Estimation and Validation Data” on page 2-31.

2-46
 Importing Data into the GUI

To restore a data set from Trash, drag its icon from Trash to the Data or
Model Board in the System Identification Tool window. You can view the
Trash contents by double-clicking the Trash icon.

Note You must restore data to the Data Board; you cannot drag data icons to
the Model Board.

To permanently delete all items in Trash, select Options > Empty trash.

Exiting a session empties the Trash automatically.

Exporting Data from the GUI to the MATLAB

Workspace
The data you create in the System Identification Tool GUI is not available
in the MATLAB workspace until you export the data set. Exporting to the
MATLAB workspace is necessary when you need to perform an operation on
the data that is only available at the command line.

To export a data set to the MATLAB workspace, drag and drop the
corresponding icon to the To Workspace rectangle.

2-47
 2 Data Import and Processing

When you export data to the MATLAB workspace, the resulting variables
have the same name as in the System Identification Tool GUI. For example,
the following figure shows how to export the time-domain data object datad.

Exporting Data to the MATLAB® Workspace

In this example, the MATLAB workspace contains a variable named data

after export.

2-48
 Representing Time- and Frequency-Domain Data Using iddata Objects

Representing Time- and Frequency-Domain Data Using

iddata Objects
In this section...
“iddata Constructor” on page 2-49
“iddata Properties” on page 2-52
“Creating Multiexperiment Data at the Command Line” on page 2-55
“Subreferencing iddata Objects” on page 2-57
“Modifying Time and Frequency Vectors” on page 2-61
“Naming, Adding, and Removing Data Channels” on page 2-64
“Concatenating iddata Objects” on page 2-66

iddata Constructor
• “Requirements for Constructing an iddata Object” on page 2-49
• “Constructing an iddata Object for Time-Domain Data” on page 2-50
• “Constructing an iddata Object for Frequency-Domain Data” on page 2-51

Requirements for Constructing an iddata Object

To construct an iddata object, you must have already imported data into the
MATLAB workspace, as described in “Importing Data into the MATLAB
Workspace” on page 2-7.

2-49
 2 Data Import and Processing

Constructing an iddata Object for Time-Domain Data

Use the following syntax to create a time-domain iddata object data:

data = iddata(y,u,Ts)

You can also specify additional properties, as follows:

data = iddata(y,u,Ts,'Property1',Value1,...,'PropertyN',ValueN)

For more information about accessing object properties, see “iddata

Properties” on page 2-52.

In this example, Ts is the sampling time, or the time interval, between

successive data samples. For uniformly sampled data, Ts is a scalar value
equal to the sampling interval of your experiment. The default time unit is
seconds, but you can specify any unit string using the TimeUnit property.
For more information about iddata time properties, see “Modifying Time
and Frequency Vectors” on page 2-61.

For nonuniformly sampled data, specify Ts as [], and set the value of the
SamplingInstants property as a column vector containing individual time
values. For example:

data = iddata(y,u,Ts,[],'SamplingInstants',TimeVector)

Where TimeVector represents a vector of time values.

Note You can modify the property SamplingInstants by setting it to a new

vector with the length equal to the number of data samples.

To represent time-series data, use the following syntax:

ts_data = iddata(y,[],Ts)

where y is the output data, [] indicates empty input data, and Ts is the
sampling interval.

2-50
 Representing Time- and Frequency-Domain Data Using iddata Objects

The following example shows how to create an iddata object using

single-input/single-output (SISO) data from dryer2.mat. The input and
output each contain 1000 samples with the sampling interval of 0.08 second.

load dryer2 % Load input u2 and output y2.

data = iddata(y2,u2,0.08) % Create iddata object.

MATLAB returns the following output:

Time domain data set with 1000 samples.

Sampling interval: 0.08

Outputs Unit (if specified)

y1

Inputs Unit (if specified)

u1

The default channel name 'y1' is assigned to the first and only output
channel. When y2 contains several channels, the channels are assigned
default names 'y1','y2','y2',...,'yn'. Similarly, the default channel
name 'u1' is assigned to the first and only input channel. For more
information about naming channels, see “Naming, Adding, and Removing
Data Channels” on page 2-64.

Constructing an iddata Object for Frequency-Domain Data

Frequency-domain data is the Fourier transform of the input and output
signals at specific frequency values. To represent frequency-domain data, use
the following syntax to create the iddata object:

data = iddata(y,u,Ts,'Frequency',w)

'Frequency' is an iddata property that specifies the frequency values w,

where w is the frequency column vector that defines the frequencies at which
the Fourier transform values of y and u are computed. Ts is the time interval
between successive data samples in seconds for the original time-domain
data. w, y, and u have the same number of rows.

2-51
 2 Data Import and Processing

Note You must specify the frequency vector for frequency-domain data.

For more information about iddata time and frequency properties, see
“Modifying Time and Frequency Vectors” on page 2-61.

To specify a continuous-time system, set Ts to 0.

You can specify additional properties when you create the iddata object, as
follows:

data = iddata(y,u,Ts,'Property1',Value1,...,'PropertyN',ValueN)

For more information about accessing object properties, see “iddata

Properties” on page 2-52.

iddata Properties
To view the properties of the iddata object, use the get command. For
example, type the following commands at the prompt:

load dryer2 % Load input u2 and output y2

data = iddata(y2,u2,0.08); % Create iddata object
get(data) % Get property values of data

2-52
 Representing Time- and Frequency-Domain Data Using iddata Objects

MATLAB returns the following object properties and values:

Domain: 'Time'
Name: []
OutputData: [1000x1 double]
y: 'Same as OutputData'
OutputName: {'y1'}
OutputUnit: {''}
InputData: [1000x1 double]
u: 'Same as InputData'
InputName: {'u1'}
InputUnit: {''}
Period: Inf
InterSample: 'zoh'
Ts: 0.0800
Tstart: []
SamplingInstants: [1000x0 double]
TimeUnit: ''
ExperimentName: 'Exp1'
Notes: []
UserData: []

For a complete description of all properties, see the iddata reference page or
type idprops iddata at the prompt.

You can specify properties when you create an iddata object using the
constructor syntax:

data = iddata(y,u,Ts,'Property1',Value1,...,'PropertyN',ValueN)

To change property values for an existing iddata object, use the set command
or dot notation. For example, to change the sampling interval to 0.05, type
the following at the prompt:

set(data,'Ts',0.05)

or equivalently:

data.ts = 0.05

Property names are not case sensitive. You do not need to type the entire
property name if the first few letters uniquely identify the property.

2-53
 2 Data Import and Processing

Tip You can use data.y as an alternative to data.OutputData to access the

output values, or use data.u as an alternative to data.InputData to access
the input values.

An iddata object containing frequency-domain data includes

frequency-specific properties, such as Frequency for the frequency vector and
Units for frequency units (instead of Tstart and SamplingIntervals).

To view the property list, type the following command sequence at the prompt:

% Load input u2 and output y2

load dryer2;
% Create iddata object
data = iddata(y2,u2,0.08);
% Take the Fourier transform of the data
% transforming it to frequency domain
data = fft(data)
% Get property values of data
get(data)

2-54
 Representing Time- and Frequency-Domain Data Using iddata Objects

MATLAB returns the following object properties and values:

Domain: 'Frequency'
Name: []
OutputData: [501x1 double]
y: 'Same as OutputData'
OutputName: {'y1'}
OutputUnit: {''}
InputData: [501x1 double]
u: 'Same as InputData'
InputName: {'u1'}
InputUnit: {''}
Period: Inf
InterSample: 'zoh'
Ts: 0.0800
Units: 'rad/s'
Frequency: [501x1 double]
TimeUnit: ''
ExperimentName: 'Exp1'
Notes: []
UserData: []

Creating Multiexperiment Data at the Command Line

• “Why Create Multiexperiment Data Sets?” on page 2-55
• “Limitations on Data Sets” on page 2-56
• “Entering Multiexperiment Data Directly” on page 2-56
• “Merging Data Sets” on page 2-56
• “Adding Experiments to an Existing iddata Object” on page 2-57

Why Create Multiexperiment Data Sets?

You can create iddata objects that contain several experiments. Identifying
models for an iddata object with multiple experiments results in an average
model.

In the System Identification Toolbox product, experiments can either

mean data collected during different sessions, or portions of the data

2-55
 2 Data Import and Processing

collected during a single session. In the latter situation, you can create a
multiexperiment iddata object by splitting the data from a single session into
multiple segments to exclude bad data, and merge the good data portions.

Note The idfrd object does not support the iddata equivalent of
multiexperiment data.

Limitations on Data Sets

You can only merge data sets that have all of the following characteristics:

• Same number of input and output channels.

• Same input and output channel names.
• Same data domain (that is, time-domain data or frequency-domain data).

Entering Multiexperiment Data Directly

To construct an iddata object that includes N data sets, you can use this
syntax:

data = iddata(y,u,Ts)

where y, u, and Ts are 1-by-N cell arrays containing data from the different
experiments. Similarly, when you specify Tstart, Period, InterSample, and
SamplingInstants properties of the iddata object, you must assign their
values as 1-by-N cell arrays.

Merging Data Sets

Create a multiexperiment iddata object by merging iddata objects, where
each contains data from a single experiment or is a multiexperiment data set.
For example, you can use the following syntax to merge data:

load iddata1 % Loads iddata object z1

load iddata3 % Loads iddata object z3
z = merge(z1,z3) % Merges experiments z1 and z3 into
% the iddata object z

2-56
 Representing Time- and Frequency-Domain Data Using iddata Objects

These commands create an iddata object that conatains two experiments,

where the experiments are assigned default names 'Exp1' and 'Exp2',
respectively.

Adding Experiments to an Existing iddata Object

You can add experiments individually to an iddata object as an alternative
approach to merging data sets.

For example, to add the experiments in the iddata object dat4 to data, use
the following syntax:

data(:,:,:,'Run4') = dat4

This syntax explicitly assigns the experiment name 'Run4' to the new
experiment. The Experiment property of the iddata object stores experiment
names.

For more information about subreferencing experiments in a multiexperiment

data set, see “Subreferencing Experiments” on page 2-60.

Subreferencing iddata Objects

• “Subreferencing Input and Output Data” on page 2-57
• “Subreferencing Data Channels” on page 2-59
• “Subreferencing Experiments” on page 2-60

Subreferencing Input and Output Data

Subreferencing data and its properties lets you select data values and assign
new data and property values.

2-57
 2 Data Import and Processing

Use the following general syntax to subreference specific data values in

iddata objects:

data(samples,outputchannels,inputchannels,experimentname)

In this syntax, samples specify one or more sample indexes, outputchannels

and inputchannels specify channel indexes or channel names, and
experimentname specifies experiment indexes or names.

For example, to retrieve samples 5 through 30 in the iddata object data and
store them in a new iddata object data_sub, use the following syntax:

data_sub = data(5:30)

You can also use logical expressions to subreference data. For example, to
retrieve all data values from a single-experiment data set that fall between
sample instants 1.27 and 9.3 in the iddata object data and assign them to
data_sub, use the following syntax:

data_sub = data(data.sa>1.27&data.sa<9.3)

Note You do not need to type the entire property name. In this example, sa
in data.sa uniquely identifies the SamplingInstants property.

You can retrieve the input signal from an iddata object using the following
commands:

u = get(data,'InputData')

or

data.InputData

or

data.u % u is the abbreviation for InputData

2-58
 Representing Time- and Frequency-Domain Data Using iddata Objects

Similarly, you can retrieve the output data using

data.OutputData

or

data.y % y is the abbreviation for OutputData

Subreferencing Data Channels

Use the following general syntax to subreference specific data channels in
iddata objects:

data(samples,outputchannels,inputchannels,experiment)

In this syntax, samples specify one or more sample indexes, outputchannels

and inputchannels specify channel indexes or channel names, and
experimentname specifies experiment indexes or names.

To specify several channel names, you must use a cell array of name strings.

For example, suppose the iddata object data contains three output channels
(named y1, y2, and y3), and four input channels (named u1, u2, u3, and u4).
To select all data samples in y3, u1, and u4, type the following command at
the prompt:

% Use a cell array to reference

% input channels 'u1' and 'u4'
data_sub = data(:,'y3',{'u1','u4'})

or equivalently

% Use channel indexes 1 and 4

% to reference the input channels
data_sub = data(:,3,[1 4])

Tip Use a colon (:) to specify all samples or all channels, and the empty
matrix ([]) to specify no samples or no channels.

2-59
 2 Data Import and Processing

If you want to create a time-series object by extracting only the output data
from an iddata object, type the following command:

data_ts = data(:,:,[])

You can assign new values to subreferenced variables. For example, the
following command assigns the first 10 values of output channel 1 of data to
values in samples 101 through 110 in the output channel 2 of data1. It also
assigns the values in samples 101 through 110 in the input channel 3 of data1
to the first 10 values of input channel 1 of data.

data(1:10,1,1) = data1(101:110,2,3)

Subreferencing Experiments
Use the following general syntax to subreference specific experiments in
iddata objects:

data(samples,outputchannels,inputchannels,experimentname)

In this syntax, samples specify one or more sample indexes, outputchannels

and inputchannels specify channel indexes or channel names, and
experimentname specifies experiment indexes or names.

When specifying several experiment names, you must use a cell array of name
strings. The iddata object stores experiments name in the ExperimentName
property.

For example, suppose the iddata object data contains five experiments with
default names, Exp1, Exp2, Exp3, Exp4, and Exp5. Use the following syntax to
subreference the first and fifth experiment in data:

data_sub = data(:,:,:,{'Exp1','Exp5'}) % Using experiment name

or

data_sub = data(:,:,:,[1 5]) % Using experiment index

Tip Use a colon (:) to denote all samples and all channels, and the empty
matrix ([]) to specify no samples and no channels.

2-60
 Representing Time- and Frequency-Domain Data Using iddata Objects

Alternatively, you can use the getexp command. The following example
shows how to subreference the first and fifth experiment in data:

data_sub = getexp(data,{'Exp1','Exp5'}) % Using experiment name

or

data_sub = getexp(data,[1 5]) % Using experiment index

The following example shows how to retrieve the first 100 samples of output
channels 2 and 3 and input channels 4 to 8 of Experiment 3:

dat(1:100,[2,3],[4:8],3)

Modifying Time and Frequency Vectors

The iddata object stores time-domain data or frequency-domain data and has
several properties that specify the time or frequency values. To modify the
time or frequency values, you must change the corresponding property values.

Note You can modify the property SamplingInstants by setting it to a

new vector with the length equal to the number of data samples. For more
information, see “Constructing an iddata Object for Time-Domain Data” on
page 2-50.

The following tables summarize time-vector and frequency-vector properties,

respectively, and provides usage examples. In each example, data is an
iddata object.

Note Property names are not case sensitive. You do not need to type the
entire property name if the first few letters uniquely identify the property.

2-61
 2 Data Import and Processing

iddata Time-Vector Properties

Property Description Syntax Example

Ts Sampling time interval. To set the sampling
interval to 0.05:
• For a single
experiment, Ts is a set(data,'ts',0.05)
scalar value.
or
• For multiexperiement
data with Ne data.ts = 0.05
experiments, Ts is
a 1-by-Ne cell array,
and each cell contains
the sampling interval
of the corresponding
experiment.
Tstart Starting time of the To change starting time
experiment. of the first data sample to
24:
• For a single
experiment, Ts is a data.Tstart = 24
scalar value.
Time units are set by the
• For multiexperiement property TimeUnit.
data with Ne
experiments, Ts is
a 1-by-Ne cell array,
and each cell contains
the sampling interval
of the corresponding
experiment.

2-62
 Representing Time- and Frequency-Domain Data Using iddata Objects

iddata Time-Vector Properties (Continued)

Property Description Syntax Example

SamplingInstants Time values in the time To retrieve the time
vector, computed from the vector for iddata object
properties Tstart and Ts. data, use:
get(data,'sa')
• For a single
experiment, To plot the input data as
SamplingInstants a function of time:
is an N-by-1 vector.
plot(data.sa,data.u)
• For multiexperiement
data with Ne
experiments, this
Note sa is the first
property is a 1-by-Ne
two letters of the
cell array, and each
SamplingInstants
cell contains the
property that uniquely
sampling instants
identifies this property.
of the corresponding
experiment.
TimeUnit Unit of time. To change the unit of the
time vector to msec:

data.ti = 'msec'

iddata Frequency-Vector Properties

Property Description Syntax Example

Frequency Frequency values at To specify 100 frequency
which the Fourier values in log space,
transforms of the signals ranging between 0.1 and
are defined. 100, use the following
syntax:
• For a single
experiment, Frequency data.freq =
is a scalar value.

2-63
 2 Data Import and Processing

iddata Frequency-Vector Properties (Continued)

Property Description Syntax Example

logspace(-1,2,100)
• For multiexperiement
data with Ne
experiments,
Frequency is a
1-by-Ne cell array,
and each cell contains
the frequencies of
the corresponding
experiment.

Units Frequency unit must If you specified the

have the following values: TimeUnit as msec, your
frequency units must be:
• If the TimeUnit is
empty or s (seconds), data.unit=
enter rad/s or Hz 'rad/msec'

• If the TimeUnit is any

string unit (other than
s), enter rad/unit.

For multiexperiement
data with Ne experiments,
Units is a 1-by-Ne cell
array, and each cell
contains the frequency
unit for each experiment.

Naming, Adding, and Removing Data Channels

• “What Are Input and Output Channels?” on page 2-65
• “Naming Channels” on page 2-65
• “Adding Channels” on page 2-66

2-64
 Representing Time- and Frequency-Domain Data Using iddata Objects

• “Modifying Channel Data” on page 2-66

What Are Input and Output Channels?

A multivariate system might contain several input variables or several output
variables, or both. When an input or output signal includes several measured
variables, these variables are called channels.

Naming Channels
The iddata properties InputName and OutputName store the channel names
for the input and output signals. When you plot the data, you use channel
names to select the variable displayed on the plot. If you have multivariate
data, it is helpful to assign a name to each channel that describes the
measured variable. For more information about selecting channels on a plot,
see “Selecting Measured and Noise Channels in Plots” on page 11-18.

You can use the set command to specify the names of individual channels.
For example, suppose data contains two input channels (voltage and current)
and one output channel (temperature). To set these channel names, use the
following syntax:

set(data,'InputName',{'Voltage','Current'},
'OutputName','Temperature')

Tip You can also specify channel names as follows:

data.una = {'Voltage','Current')
data.yna = 'Temperature'

una is equivalent to the property InputName, and yna is equivalent to

OutputName.

If you do not specify channel names when you create the iddata object,
the toolbox assigns default names. By default, the output channels
are named 'y1','y2',...,'yn', and the input channels are named
'u1','u2',...,'un'.

2-65
 2 Data Import and Processing

Adding Channels
You can add data channels to an iddata object.

For example, consider an iddata object named data that contains an input
signal with four channels. To add a fifth input channel, stored as the vector
Input5, use the following syntax:

data.u(:,5) = Input5;

Input5 must have the same number of rows as the other input channels. In
this example, data.u(:,5) references all samples as (indicated by :) of the
input signal u and sets the values of the fifth channel. This channel is created
when assigning its value to Input5.

You can also combine input channels and output channels of several iddata
objects into one iddata object using concatenation. For more information, see
“Concatenating iddata Objects” on page 2-66.

Modifying Channel Data

After you create an iddata object, you can modify or remove specific input
and output channels, if needed. You can accomplish this by subreferencing
the input and output matrices and assigning new values.

For example, suppose the iddata object data contains three output channels
(named y1, y2, and y3), and four input channels (named u1, u2, u3, and u4).
To replace data such that it only contains samples in y3, u1, and u4, type
the following at the prompt:

data = data(:,3,[1 4])

The resulting data object contains one output channel and two input channels.

Concatenating iddata Objects

• “iddata Properties Storing Input and Output Data” on page 2-67
• “Horizontal Concatenation” on page 2-67
• “Vertical Concatenation” on page 2-68

2-66
 Representing Time- and Frequency-Domain Data Using iddata Objects

iddata Properties Storing Input and Output Data

The InputData iddata property stores column-wise input data, and the
OutputData property stores column-wise output data. For more information
about accessing iddata properties, see “iddata Properties” on page 2-52.

Horizontal Concatenation
Horizontal concatenation of iddata objects creates a new iddata object
that appends all InputData information and all OutputData. This type of
concatenation produces a single object with more input and output channels.
For example, the following syntax performs horizontal concatenation on the
iddata objects data1,data2,...,dataN:

data = [data1,data2,...,dataN]

This syntax is equivalent to the following longer syntax:

data.InputData =
[data1.InputData,data2.InputData,...,dataN.InputData]
data.OutputData =
[data1.OutputData,data2.OutputData,...,dataN.OutputData]

For horizontal concatenation, data1,data2,...,dataN must have the same

number of samples and experiments , and the sameTs and Tstart values.

The channels in the concatenated iddata object are named according to the
following rules:

• Combining default channel names. If you concatenate iddata objects

with default channel names, such as u1 and y1, channels in the new iddata
object are automatically renamed to avoid name duplication.
• Combining duplicate input channels. If data1,data2,...,dataN
have input channels with duplicate user-defined names, such that dataK
contains channel names that are already present in dataJ with J < K, the
dataK channels are ignored.
• Combining duplicate output channels. If data1,data2,...,dataN
have input channels with duplicate user-defined names, only the output
channels with unique names are added during the concatenation.

2-67
 2 Data Import and Processing

Vertical Concatenation
Vertical concatenation of iddata objects creates a new iddata object that
vertically stacks the input and output data values in the corresponding data
channels. The resulting object has the same number of channels, but each
channel contains more data points. For example, the following syntax creates
a data object such that its total number of samples is the sum of the samples
in data1,data2,...,dataN.

data = [data1;data2;... ;dataN]

This syntax is equivalent to the following longer syntax:

data.InputData =
[data1.InputData;data2.InputData;...;dataN.InputData]
data.OutputData =
[data1.OutputData;data2.OutputData;...;dataN.OutputData]

For vertical concatenation, data1,data2,...,dataN must have the same

number of input channels, output channels, and experiments.

2-68
 Representing Frequency-Response Data Using idfrd Objects

Representing Frequency-Response Data Using idfrd

Objects
In this section...
“idfrd Constructor” on page 2-69
“idfrd Properties” on page 2-70
“Subreferencing idfrd Objects” on page 2-72
“Concatenating idfrd Objects” on page 2-73
“See Also” on page 2-76

idfrd Constructor
The idfrd represents complex frequency-response data. Before you can
create an idfrd object, you must import your data as described in “Importing
Frequency-Response Data into MATLAB” on page 2-11.

Note The idfrd object can only encapsulate one frequency-response data
set. It does not support the iddata equivalent of multiexperiment data.

Use the following syntax to create the data object fr_data:

fr_data = idfrd(response,f,Ts)

Suppose that ny is the number of output channels, nu is the number of

input channels, and nf is a vector of frequency values. response is an
ny-by-nu-by-nf 3-D array. f is the frequency vector that contains the
frequencies of the response.Ts is the sampling time, which is used when
measuring or computing the frequency response. If you are working with a
continuous-time system, set Ts to 0.

response(ky,ku,kf), where ky, ku, and kf reference the kth output, input,
and frequency value, respectively, is interpreted as the complex-valued
frequency response from input ku to output ky at frequency f(kf).

2-69
 2 Data Import and Processing

Note When you work at the command line, you can only create idfrd objects
from complex values of G(eiw). For a SISO system, response can be a vector.

You can specify object properties when you create the idfrd object using
the constructor syntax:

fr_data = idfrd(response,f,Ts,
'Property1',Value1,...,'PropertyN',ValueN)

idfrd Properties
To view the properties of the idfrd object, you can use the get command.
The following example shows how to create an idfrd object that contains
100 frequency-response values with a sampling time interval of 0.08 s and
get its properties:

% Create the idfrd data object

fr_data = idfrd(response,f,0.08)
% Get property values of data
get(fr_data)

2-70
 Representing Frequency-Response Data Using idfrd Objects

response and f are variables in the MATLAB Workspace browser,

representing the frequency-response data and frequency values, respectively.

MATLAB returns the following object properties and values:

ans =

Name: ''
Frequency: [100x1 double]
ResponseData: [1x1x100 double]
SpectrumData: []
CovarianceData: []
NoiseCovariance: []
Units: 'rad/s'
Ts: 0.0800
InputDelay: 0
EstimationInfo: [1x1 struct]
InputName: {'u1'}
OutputName: {'y1'}
InputUnit: {''}
OutputUnit: {''}
Notes: []
UserData: []

For a complete description of all idfrd object properties, see the idfrd
reference page or type idprops idfrd at the prompt.

To change property values for an existing idfrd object, use the set command
or dot notation. For example, to change the name of the idfrd object, type
the following command sequence at the prompt:

% Set the name of the f_data object

set(fr_data,'name','DC_Converter')
% Get fr_data properties and values
get(fr_data)

Property names are not case sensitive. You do not need to type the entire
property name if the first few letters uniquely identify the property.

2-71
 2 Data Import and Processing

If you import fr_data into the System Identification Tool GUI, this data has
the name DC_Converter in the GUI, and not the variable name fr_data.

MATLAB returns the following object properties and values:

ans =

Name: 'DC_Converter'
Frequency: [100x1 double]
ResponseData: [1x1x100 double]
SpectrumData: []
CovarianceData: []
NoiseCovariance: []
Units: 'rad/s'
Ts: 0.0800
InputDelay: 0
EstimationInfo: [1x1 struct]
InputName: {'u1'}
OutputName: {'y1'}
InputUnit: {''}
OutputUnit: {''}
Notes: []
UserData: []

Subreferencing idfrd Objects

You can reference specific data values in the idfrd object using the following
syntax:

fr_data(outputchannels,inputchannels)

Reference specific channels by name or by channel index.

Tip Use a colon (:) to specify all channels, and use the empty matrix ([]) to
specify no channels.

2-72
 Representing Frequency-Response Data Using idfrd Objects

For example, the following command references frequency-response data from

input channel 3 to output channel 2:

fr_data(2,3)

You can also access the data in specific channels using channel names. To list
multiple channel names, use a cell array. For example, to retrieve the power
output, and the voltage and speed inputs, use the following syntax:

fr_data('power',{'voltage','speed'})

To retrieve only the responses corresponding to frequency values between 200

and 300, use the following command:

fr_data_sub = fselect(fr_data,[200:300])

You can also use logical expressions to subreference data. For example, to
retrieve all frequency-response values between frequencies 1.27 and 9.3 in
the idfrd object fr_data, use the following syntax:

fr_data_sub = fselect(fr_data,fr_data.f>1.27&fr_data.f<9.3)

Tip Use end to reference the last sample number in the data. For example,
data(77:end).

Note You do not need to type the entire property name. In this example, f in
fr_data.f uniquely identifies the Frequency property of the idfrd object.

Concatenating idfrd Objects

• “About Concatenating idfrd Models” on page 2-74
• “Horizontal Concatenation of idfrd Objects” on page 2-74
• “Vertical Concatenation of idfrd Objects” on page 2-75
• “Concatenating Noise Spectrum Data of idfrd Objects” on page 2-75

2-73
 2 Data Import and Processing

About Concatenating idfrd Models

The horizontal and vertical concatenation of idfrd objects combine
information in the ResponseData properties of these objects. ResponseData is
an ny-by-nu-by-nf array that stores the response of the system, where ny is
the number of output channels, nu is the number of input channels, and nf is
a vector of frequency values (see “idfrd Properties” on page 2-70).

Horizontal Concatenation of idfrd Objects

The following syntax creates a new idfrd object data that contains the
horizontal concatenation of data1,data2,...,dataN:

data = [data1,data2,...,dataN]

data contains the frequency responses from all of the inputs in

data1,data2,...,dataN to the same outputs. The following diagram is a
graphical representation of horizontal concatenation of frequency-response
data. The (j,i,:) vector of the resulting response data represents the
frequency response from the ith input to the jth output at all frequencies.

u1 Data 1 y1 Data 2 y1
2-by-2-by-nf u3 2-by-1-by-nf
u2 y2 y2

u1 Horizonal Concatenation y1
Combined of Data 1 and Data 2 Same
inputs u2 outputs
2-by-3-by-nf y2
u3

Note Horizontal concatenation of idfrd objects requires that they have

the same outputs and frequency vectors. If the output channel names are
different and their dimensions are the same, the concatenation operation uses
the names of output channels in the first idfrd object. Input channels must
have unique names.

2-74
 Representing Frequency-Response Data Using idfrd Objects

Vertical Concatenation of idfrd Objects

The following syntax creates a new idfrd object data that contains the
vertical concatenation of data1,data2,...,dataN:

data = [data1;data2;... ;dataN]

The resulting idfrd object data contains the frequency responses from
the same inputs in data1,data2,...,dataN to all the outputs. The
following diagram is a graphical representation of vertical concatenation of
frequency-response data. The (j,i,:) vector of the resulting response data
represents the frequency response from the ith input to the jth output at all
frequencies.

u1 Data 1 y1 u1 Data 1 y3
u2 2-by-2-by-nf y2 u2 1-by-2-by-nf

u1 Vertical Concatenation y1
Same y2 Combined
inputs of Data 1 and Data 2 outputs
u2 3-by-2-by-nf y3

Note Vertical concatenation of idfrd objects requires that they have the
same inputs and frequency vectors. If the input channel names are different
and their dimensions are the same, the concatenation operation uses the
names of input channels in the first idfrd object you listed. Output channels
must have unique names.

Concatenating Noise Spectrum Data of idfrd Objects

When the idfrd objects contain the frequency-response data you measured
or constructed manually, the concatenation operation combines only the
ResponseData properties. Because the noise spectrum data does not exist
(unless you also entered it manually), SpectrumData is empty in both the
individual idfrd objects and the concatenated idfrd object.

2-75
 2 Data Import and Processing

However, when the idfrd objects are spectral models that you estimated, the
SpectrumData property is not empty and contains the power spectra and
cross spectra of the output noise in the system. For each output channel, the
toolbox estimates one noise channel to explain the difference between the
output of the model and the measured output.

When the SpectrumData property of individual idfrd objects is not empty,

horizontal and vertical concatenation handle SpectrumData, as follows.

In case of horizontal concatenation, there is no meaningful way to combine the

SpectrumData of individual idfrd objects and the resulting SpectrumData
property is empty. An empty property results because each idfrd object has
its own set of noise channels, where the number of noise channels equals the
number of outputs. When the resulting idfrd object contains the same output
channels as each of the individual idfrd objects, it cannot accommodate the
noise data from all the idfrd objects.

In case of vertical concatenation, the toolbox concatenates individual noise

models diagonally. The following shows that data.SpectrumData is a block
diagonal matrix of the power spectra and cross spectra of the output noise in
the system:

⎛ data1.s
⎜ 0 ⎞⎟
data.s = ⎜  ⎟
⎜⎜ ⎟
⎝ 0 dataN .s ⎟⎠

s in data.s is the abbreviation for the SpectrumData property name.

See Also
The following operations also create idfrd objects:

• Transforming iddata objects. For more information, see “Transforming

Between Frequency-Domain and Frequency-Response Data” on page 2-130.
• Estimating nonparametric models using etfe, spa, and spafdr. For more
information, see “Identifying Frequency-Response Models” on page 3-2.

2-76
 Representing Frequency-Response Data Using idfrd Objects

• Converting the Control System Toolbox frd object. For more information,
see “Using Models with Control System Toolbox Software” on page 9-2.

2-77
 2 Data Import and Processing

Analyzing Data Quality

In this section...
“Is Your Data Ready for Modeling?” on page 2-78
“Plotting Data in the GUI Versus at the Command Line” on page 2-79
“How to Plot Data in the GUI” on page 2-79
“How to Plot Data at the Command Line” on page 2-85
“How to Analyze Data Using the advice Command” on page 2-87

Is Your Data Ready for Modeling?

Before you start estimating models from data, you should check your data for
the presence of any undesirable characteristics. For example, you might plot
the data to identify drifts and outliers. You plot analysis might lead you to
preprocess your data before model estimation.

The following data plots are available in the toolbox:

• Time plot — Shows data values as a function of time.

Tip You can infer time delays from time plots, which are required inputs
to most parametric models. A time delay is the time interval between the
change in input and the corresponding change in output.

• Spectral plot — Shows a periodogram that is computed by taking the

absolute squares of the Fourier transforms of the data, dividing by the
number of data points, and multiplying by the sampling interval.
• Frequency-response plot — For frequency-response data, shows the
amplitude and phase of the frequency-response function on a Bode plot. For
time- and frequency-domain data, shows the empirical transfer function
estimate (see etfe) .

See Also
“How to Analyze Data Using the advice Command” on page 2-87

2-78
 Analyzing Data Quality

“Ways to Process Data for System Identification” on page 2-4

Plotting Data in the GUI Versus at the Command Line

The plots you create using the System Identification Tool GUI provide
options that are specific to the System Identification Toolbox product, such
as selecting specific channel pairs in a multivariate signals or converting
frequency units between Hertz and radians per second. For more information,
see “How to Plot Data in the GUI” on page 2-79.

The plots you create using the plot commands, such as plot, bode, and
ffplot, are displayed in the standard MATLAB Figure window, which
provides options for formatting, saving, printing, and exporting plots to a
variety of file formats. To learn about plotting at the command line, see “How
to Plot Data at the Command Line” on page 2-85. For more information about
working with Figure window, see the MATLAB Graphics documentation.

How to Plot Data in the GUI

• “How to Plot Data in the GUI” on page 2-79
• “Manipulating a Time Plot” on page 2-81
• “Manipulating Data Spectra Plot” on page 2-82
• “Manipulating a Frequency Function Plot” on page 2-84

How to Plot Data in the GUI

After importing data into the System Identification Tool GUI, as described in
“Importing Data into the GUI” on page 2-14, you can plot the data.

To create one or more plots, select the corresponding check box in the Data
Views area of the System Identification Tool GUI.

An active data icon has a thick line in the icon, while an inactive data set
has a thin line. Only active data sets appear on the selected plots. To toggle
including and excluding data on a plot, click the corresponding icon in the
System Identification Tool GUI. Clicking the data icon updates any plots
that are currently open.

2-79
 2 Data Import and Processing

When you have several data sets, you can view different input-output channel
pair by selecting that pair from the Channel menu. For more information
about selecting different input and output pairs, see “Selecting Measured and
Noise Channels in Plots” on page 11-18.

Thick lines indicate

active data sets
included in plots.

All three available

data plots are
selected.

In this example, data and dataff are active and appear on the three selected
plots.

To close a plot, clear the corresponding check box in the System Identification
Tool GUI.

Tip To get information about working with a specific plot, select a help topic
from the Help menu in the plot window.

2-80
 Analyzing Data Quality

Manipulating a Time Plot

The Time plot only shows time-domain data. In this example, data1 is
displayed on the time plot because, of the three data sets, it is the only one
that contains time-domain input and output.

Time Plot of data1

The following table summarizes options that are specific to time plots, which
you can select from the plot window menus. For general information about
working with System Identification Toolbox plots, see “Working with Plots in
the System Identification Tool GUI” on page 11-15.

2-81
 2 Data Import and Processing

Time Plot Options

Action Command
Toggle input display between Select Style > Staircase input for
piece-wise continuous (zero-order zero-order hold or Style > Regular
hold) and linear interpolation input for first-order hold.
(first-order hold) between samples.

Note This option only affects the

display and not the intersample
behavior specified when importing
the data.

Manipulating Data Spectra Plot

The Data spectra plot shows a periodogram or a spectral estimate of data1
and data3fd.

The periodogram is computed by taking the absolute squares of the Fourier

transforms of the data, dividing by the number of data points, and multiplying
by the sampling interval. The spectral estimate for time-domain data is a
smoothed spectrum calculated using spa. For frequency-domain data, the
Data spectra plot shows the square of the absolute value of the actual data,
normalized by the sampling interval.

The top axes show the input and the bottom axes show the output. The
vertical axis of each plot is labeled with the corresponding channel name.

2-82
 Analyzing Data Quality

Periodograms of data1 and data3fd

Data Spectra Plot Options

Action Command
Toggle display between periodogram Select Options > Periodogram or
and spectral estimate. Options > Spectral analysis.
Change frequency units. Select Style > Frequency (rad/s)
or Style > Frequency (Hz).
Toggle frequency scale between Select Style > Linear frequency
linear and logarithmic. scale or Style > Log frequency
scale.
Toggle amplitude scale between Select Style > Linear amplitude
linear and logarithmic. scale or Style > Log amplitude
scale.

2-83
 2 Data Import and Processing

Manipulating a Frequency Function Plot

For time-domain data, the Frequency function plot shows the empirical
transfer function estimate (etfe). For frequency-domain data, the plot shows
the ratio of output to input data.

The frequency-response plot shows the amplitude and phase plots of

the corresponding frequency response. For more information about
frequency-response data, see “Importing Frequency-Response Data into
MATLAB” on page 2-11.

Frequency Functions of data1 and data3fd

2-84
 Analyzing Data Quality

Frequency Function Plot Options

Action Command
Change frequency units. Select Style > Frequency (rad/s)
or Style > Frequency (Hz).
Toggle frequency scale between Select Style > Linear frequency
linear and logarithmic. scale or Style > Log frequency
scale.
Toggle amplitude scale between Select Style > Linear amplitude
linear and logarithmic. scale or Style > Log amplitude
scale.

How to Plot Data at the Command Line

The following table summarizes the commands available for plotting
time-domain, frequency-domain, and frequency-response data.

Commands for Plotting Data

Command Description Example

bode For frequency-response data To plot idfrd data:
only. Shows the magnitude
and phase of the frequency bode(idfrd_data)
response on a logarithmic
frequency scale of a Bode
plot.
ffplot For frequency-response To plot idfrd data:
data only. Shows the
magnitude and phase of ffplot(idfrd_data)
the frequency response on
a linear frequency scale
(hertz).
plot The type of plot corresponds To plot iddata or idfrd
to the type of data. data:
For example, plotting
time-domain data generates plot(data)
a time plot, and plotting

2-85
 2 Data Import and Processing

Commands for Plotting Data (Continued)

Command Description Example

frequency-response
data generates a
frequency-response plot. Note For idfrd data, this
When plotting time- or command is equivalent to
frequency-domain inputs ffplot(data).
and outputs, the top axes
show the output and the
bottom axes show the input.

All plot commands display the data in the standard MATLAB Figure window.
For more information about working with the Figure window, see the
MATLAB Graphics documentation.

To plot portions of the data, you can subreference specific samples (see
“Subreferencing iddata Objects” on page 2-57 and “Subreferencing idfrd
Objects” on page 2-72. For example:

plot(data(1:300))

For time-domain data, to plot only the input data as a function of time, use
the following syntax:

plot(data(:,[],:)

When data.intersample = 'zoh', the input is piece-wise constant between

sampling points on the plot. For more information about properties, see the
iddata reference page.

You can generate plots of the input data in the time domain using:

plot(data.sa,data.u)

To plot frequency-domain data, you can use the following syntax:

semilogx(data.fr,abs(data.u))

2-86
 Analyzing Data Quality

In this case, sa is an abbreviation of the iddata property SamplingInstants.

Similarly, fr is an abbreviation of Frequency. u is the input signal.

Note The frequencies are linearly spaced on the plot.

When you specify to plot a multivariable iddata object, each input-output

combination is displayed one at a time in the same MATLAB Figure window.
You must press Enter to update the Figure window and view the next
channel combination. To cancel the plotting operation, press Ctrl+C.

Tip To plot specific input and output channels, use plot(data(:,ky,ku)),

where ky and ku are specific output and input channel indexes or names. For
more information about subreferencing channels, see “Subreferencing Data
Channels” on page 2-59.

To plot several iddata sets d1,...,dN, use plot(d1,...,dN). Input-output

channels with the same experiment name, input name, and output name
are always plotted in the same plot.

How to Analyze Data Using the advice Command

You can use the advice command to analyze time- or frequency- domain
data before estimating a model. The resulting report informs you about the
possible need to preprocess the data and identifies potential restrictions on
the model accuracy. You should use these recommendations in combination
with plotting the data and validating the models estimated from this data.

Note advice does not support frequency-response data.

Before applying the advice command to your data, you must have represented
your data as an iddata object. For more information, see “Representing Time-
and Frequency-Domain Data Using iddata Objects” on page 2-49.

2-87
 2 Data Import and Processing

If you are using the System Identification Tool GUI, you must export your
data to the MATLAB workspace before you can use the advice command on
this data. For more information about exporting data, see “Exporting Models
from the GUI to the MATLAB Workspace” on page 11-13.

Use the following syntax to get advice about an iddata object data:

advice(data)

For more information about the advice syntax, see the advice reference page.

Advice provide guidance for these kinds of questions:

• Does it make sense to remove constant offsets and linear trends from the
data?
• What are the excitation levels of the signals and how does this affects the
model orders?
• Is there an indication of output feedback in the data? When feedback
is present in the system, only prediction-error methods work well for
estimating closed-loop data.
• What is the estimated input-output delay in the system (dead time)?

See Also
advice

delayest

detrend

feedback

pexcit

2-88
 Selecting Subsets of Data

Selecting Subsets of Data

In this section...
“Why Select Subsets of Data?” on page 2-89
“Selecting Data Using the GUI” on page 2-90
“Selecting Data at the Command Line” on page 2-92

Why Select Subsets of Data?

You can use data selection to create independent data sets for estimation
and validation.

You can also use data selection as a way to clean the data and exclude parts
with noisy or missing information. For example, when your data contains
missing values, outliers, level changes, and disturbances, you can select one
or more portions of the data that are suitable for identification and exclude
the rest.

If you only have one data set and you want to estimate linear models, you
should split the data into two portions to create two independent data sets for
estimation and validation, respectively. Splitting the data is selecting parts of
the data set and saving each part independently.

You can merge several data segments into a single multiexperiment data set
and identify an average model. For more information, see “Importing Data
into the GUI” on page 2-14 or “Representing Time- and Frequency-Domain
Data Using iddata Objects” on page 2-49.

Note Subsets of the data set must contain enough samples to adequately
represent the system, and the inputs must provide suitable excitation to the
system.

Selecting potions of frequency-domain data is equivalent to filtering the data.

For more information about filtering, see “Filtering Data” on page 2-109.

2-89
 2 Data Import and Processing

Selecting Data Using the GUI

• “Ways to Select Data in the GUI” on page 2-90
• “Selecting a Range for Time-Domain Data” on page 2-90
• “Selecting a Range of Frequency-Domain Data” on page 2-92

Ways to Select Data in the GUI

You can use System Identification Tool GUI to select ranges of data on a
time-domain or frequency-domain plot. Selecting data in the frequency
domain is equivalent to passband-filtering the data.

After you select portions of the data, you can specify to use one data segment
for estimating models and use the other data segment for validating models.
For more information, see “Specifying Estimation and Validation Data” on
page 2-31.

Note Selecting <--Preprocess > Quick start performs the following actions
simultaneously:

• Remove the mean value from each channel.

• Split the data into two parts.
• Specify the first part as estimation data (or Working Data).
• Specify the second part as Validation Data.

Selecting a Range for Time-Domain Data

You can select a range of data values on a time plot and save it as a new data
set in the System Identification Tool GUI.

Note Selecting data does not extract experiments from a data set containing
multiple experiments. For more information about multiexperiment data, see
“Creating Multiexperiment Data Sets in the GUI” on page 2-35.

2-90
 Selecting Subsets of Data

To extract a subset of time-domain data and save it as a new data set:

1 Import time-domain data into the System Identification Tool GUI, as

described in “Importing Data into the GUI” on page 2-14.

2 Drag the data set you want to subset to the Working Data area.

3 If your data contains multiple I/O channels, in the Channel menu, select
the channel pair you want to view. The upper plot corresponds to the input
signal, and the lower plot corresponds to the output signal.

Although you view only one I/O channel pair at a time, your data selection
is applied to all channels in this data set.

4 Select the data of interest in either of the following ways:

• Graphically — Draw a rectangle on either the input-signal or the

output-signal plot with the mouse to select the desired time interval.
Your selection appears on both plots regardless of the plot on which you
draw the rectangle. The Time span and Samples fields are updated to
match the selected region.
• By specifying the Time span — Edit the beginning and the end times
in seconds. The Samples field is updated to match the selected region.
For example:
28.5 56.8
• By specifying the Samples range — Edit the beginning and the end
indices of the sample range. The Time span field is updated to match
the selected region. For example:
342 654

Note To clear your selection, click Revert.

5 In the Data name field, enter the name of the data set containing the
selected data.

6 Click Insert. This action saves the selection as a new data set and adds
it to the Data Board.

2-91
 2 Data Import and Processing

7 To select another range, repeat steps 4 to 6.

Selecting a Range of Frequency-Domain Data

Selecting a range of values in frequency domain is equivalent to filtering
the data. For more information about data filtering, see “Filtering
Frequency-Domain or Frequency-Response Data in the GUI” on page 2-112.

Selecting Data at the Command Line

Selecting ranges of data values is equivalent to subreferencing the data.

For more information about subreferencing time-domain and

frequency-domain data, see “Subreferencing iddata Objects” on page 2-57.

For more information about subreferencing frequency-response data, see

“Subreferencing idfrd Objects” on page 2-72.

2-92
 Handling Missing Data and Outliers

Handling Missing Data and Outliers

In this section...
“Handling Missing Data” on page 2-93
“Handling Outliers” on page 2-94
“Example – Extracting and Modeling Specific Data Segments” on page 2-95
“See Also” on page 2-96

Handling Missing Data

Data acquisition failures sometimes result in missing measurements both in
the input and the output signals. When you import data that contains missing
values using the MATLAB Import Wizard, these values are automatically set
to NaN (“Not-a-Number”). NaN serves as a flag for nonexistent or undefined
data. When you plot data on a time-plot that contains missing values, gaps
appear on the plot where missing data exists.

You can use misdata to estimate missing values. This command linearly
interpolates missing values to estimate the first model. Then, it uses this
model to estimate the missing data as parameters by minimizing the output
prediction errors obtained from the reconstructed data. You can specify the
model structure you want to use in the misdata argument or estimate a
default-order model using the n4sid method. For more information, see the
misdata reference page.

Note You can only use misdata on time-domain data stored in an iddata
object. For more information about creating iddata objects, see “Representing
Time- and Frequency-Domain Data Using iddata Objects” on page 2-49.

For example, suppose y and u are output and input signals that contain NaNs.
This data is sampled at 0.2 s. The following syntax creates a new iddata
object with these input and output signals.

dat = iddata(y,u,0.2) % y and u contain NaNs

% representing missing data

2-93
 2 Data Import and Processing

Apply the misdata command to the new data object. For example:

dat1 = misdata(dat);
plot(dat,dat1) % Check how the missing data
% was estimated on a time plot

Handling Outliers
Malfunctions can produce errors in measured values, called outliers. Such
outliers might be caused by signal spikes or by measurement malfunctions.
If you do not remove outliers from your data, this can adversely affect the
estimated models.

To identify the presence of outliers, perform one of the following tasks:

• Before estimating a model, plot the data on a time plot and identify values
that appear out of range.
• After estimating a model, plot the residuals and identify unusually large
values. For more information about plotting residuals, see “Residual
Analysis” on page 8-21. Evaluate the original data that is responsible for
large residuals. For example, for the model Model and validation data
Data, you can use the following commands to plot the residuals:

% Compute the residuals

E = resid(Model,Data)
% Plot the residuals
plot(E)

Next, try these techniques for removing or minimizing the effects of outliers:

• Extract the informative data portions into segments and merge them into
one multiexperiment data set (see “Example – Extracting and Modeling
Specific Data Segments” on page 2-95). For more information about
selecting and extracting data segments, see “Selecting Subsets of Data”
on page 2-89.

2-94
 Handling Missing Data and Outliers

Tip The inputs in each of the data segments must be consistently exciting
the system. Splitting data into meaningful segments for steady-state
data results in minimum information loss. Avoid making data segments
too small.

• Manually replace outliers with NaNs and then use the misdata command
to reconstruct flagged data. This approach treats outliers as missing data
and is described in “Handling Missing Data” on page 2-93. Use this method
when your data contains several inputs and outputs, and when you have
difficulty finding reliable data segments in all variables.
• Remove outliers by prefiltering the data for high-frequency content because
outliers often result from abrupt changes. For more information about
filtering, see “Filtering Data” on page 2-109.

Note The estimation algorithm handles outliers automatically by assigning

a smaller weight to outlier data. A robust error criterion applies an error
penalty that is quadratic for small and moderate prediction errors, and is
linear for large prediction errors. Because outliers produce large prediction
errors, this approach gives a smaller weight to the corresponding data points
during model estimation. The value LimitError (see Algorithm Properties)
quantitatively distinguishes between moderate and large outliers.

Example – Extracting and Modeling Specific Data

Segments
The following example shows how to create a multiexperiment, time-domain
data set by merging only the accurate-data segments and ignoring the rest.
Modeling multiexperiment data sets produces an average model for the
different experiments.

You cannot simply concatenate the good data segments because the transients
at the connection points compromise the model. Instead, you must create a
multiexperiment iddata object, where each experiment corresponds to a good
segment of data, as follows:

2-95
 2 Data Import and Processing

% Plot the data in a MATLAB Figure window

plot(data)

% Create multiexperiment data set

% by merging data segments
datam = merge(data(1:340),...
data(500:897),...
data(1001:1200),...
data(1550:2000));

% Model the multiexperiment data set

% using "experiments" 1, 2, and 4
m =pem(getexp(datam,[1,2,4]))

% Validate the model by comparing its output to

% the output data of experiment 3
compare(getexp(datam,3),m)

See Also
To learn more about the theory of handling missing data and outliers, see the
chapter on preprocessing data in System Identification: Theory for the User,
Second Edition, by Lennart Ljung, Prentice Hall PTR, 1999.

2-96
 Handling Offsets and Trends in Data

Handling Offsets and Trends in Data

In this section...
“When to Detrend Data” on page 2-97
“Alternatives for Detrending Data in GUI or at the Command-Line” on
page 2-98
“How to Detrend Data Using the GUI” on page 2-99
“How to Detrend Data at the Command Line” on page 2-99
“Next Steps After Detrending” on page 2-101

When to Detrend Data

Detrending is removing means, offsets, or linear trends from regularly
sampled time-domain input-output data signals. This data processing
operation helps you estimate more accurate linear models because linear
models cannot capture arbitrary differences between the input and output
signal levels. The linear models you estimate from detrended data describe
the relationship between the change in input signals and the change in output
signals.

For steady-state data, you should remove mean values and linear trends
from both input and output signals.

For transient data, you should remove physical-equilibrium offsets measured

prior to the excitation input signal.

Remove one linear trend or several piecewise linear trends when the levels
drift during the experiment. Signal drift is considered a low-frequency
disturbance and can result in unstable models.

You should not detrend data before model estimation when you want:

• Linear models that capture offsets essential for describing important

system dynamics. For example, when a model contains integration
behavior, you could estimate a low-order transfer function (process model)
from nondetrended data. For more information, see “Identifying Low-Order
Transfer Functions (Process Models)” on page 3-20.

2-97
 2 Data Import and Processing

• Nonlinear black-box models, such as nonlinear ARX or

Hammerstein-Wiener models. For more information, see Chapter 4,
“Nonlinear Black-Box Model Identification”.

Tip When signals vary around a large signal level, you can improve
computational accuracy of nonlinear models by detrending the signal
means.

• Nonlinear ODE parameters (nonlinear grey-box models). For more

information, see “Estimating Nonlinear Grey-Box Models” on page 5-16.

To simulate or predict the linear model response at the system operating

conditions, you can restore the removed trend to the simulated or predicted
model output using the retrend command.

For more information about handling drifts in the data, see the chapter on
preprocessing data in System Identification: Theory for the User, Second
Edition, by Lennart Ljung, Prentice Hall PTR, 1999.

Alternatives for Detrending Data in GUI or at the

Command-Line
You can detrend data using the System Identification Tool GUI and at the
command line using the detrend command.

Both the GUI and the command line let you subtract the mean values and one
linear trend from steady-state time-domain signals.

However, the detrend command provides the following additional

functionality (not available in the GUI):

• Subtracting piecewise linear trends at specified breakpoints. A breakpoint

is a time value that defines the discontinuities between successive linear
trends.
• Subtracting arbitrary offsets and linear trends from transient data signals.
• Saving trend information to a variable so that you can apply it to multiple
data sets.

2-98
 Handling Offsets and Trends in Data

To learn how to detrend data, see:

• “How to Detrend Data Using the GUI” on page 2-99

• “How to Detrend Data at the Command Line” on page 2-99

How to Detrend Data Using the GUI

Before you can perform this task, you must have regularly-sampled,
steady-state time-domain data imported into the System Identification Tool
GUI. See “Importing Time-Domain Data into the GUI” on page 2-16). For
transient data, see “How to Detrend Data at the Command Line” on page 2-99.

Tip You can use the shortcut Preprocess > Quick start to perform several
operations: remove the mean value from each signal, split data into two
halves, specify the first half as model estimation data (or Working Data),
and specify the second half as model Validation Data.

1 In the System Identification Tool, drag the data set you want to detrend to
the Working Data rectangle.

2 Detrend the data.

• To remove linear trends, select Preprocess > Remove trends.

• To remove mean values from each input and output data signal, select
Preprocess > Remove means.

How to Detrend Data at the Command Line

• “Detrending Steady-State Data” on page 2-99
• “Detrending Transient Data” on page 2-100
• “See Also” on page 2-101

Detrending Steady-State Data

Before you can perform this task, you must have time-domain data as an
iddata object. See “Representing Time- and Frequency-Domain Data Using
iddata Objects” on page 2-49.

2-99
 2 Data Import and Processing

Note If you plan to estimate models from this data, your data must be
regularly sampled.

Use the detrend command to remove the signal means or linear trends:

[data_d,T]=detrend(data,Type)

where data is the data to be detrended. The second input argument

Type=0 removes signal means or Type=1 removes linear trends. data_d is
the detrended data. T is a TrendInfo object that stores the values of the
subtracted offsets and slopes of the removed trends.

Detrending Transient Data

Before you can perform this task, you must have

• Time-domain data as an iddata object. See “Representing Time- and

Frequency-Domain Data Using iddata Objects” on page 2-49.

Note If you plan to estimate models from this data, your data must be
regularly sampled.

• Values of the offsets you want to remove from the input and output data.
If you do not know these values, visually inspect a time plot of your data.
For more information, see “How to Plot Data at the Command Line” on
page 2-85.

1 Create a default object for storing input-output offsets that you want to
remove from the data.

T = getTrend(data)

where T is a TrendInfo object.

2 Assign offset values to T.

T.InputOffset=I_value;

2-100
 Handling Offsets and Trends in Data

T.OutputOffset=O_value;

where I_value is the input offset value, and O_value is the input offset
value.

3 Remove the specified offsets from data.

data_d = detrend(data,T)

where the second input argument T stores the offset values as its properties.

See Also
detrend

TrendInfo

Next Steps After Detrending

After detrending your data, you might do the following:

• Perform other data preprocessing operations. See “Ways to Process Data

for System Identification” on page 2-4.
• Estimate a linear model. See Chapter 3, “Linear Model Identification”.

2-101
 2 Data Import and Processing

Resampling Data
In this section...
“What Is Resampling?” on page 2-102
“Resampling Data Using the GUI” on page 2-103
“Resampling Data at the Command Line” on page 2-103
“Resampling Data Without Aliasing Effects” on page 2-105
“See Also” on page 2-108

What Is Resampling?
Resampling data signals in the System Identification Toolbox product applies
an antialiasing (lowpass) FIR filter to the data and changes the sampling rate
of the signal by decimation or interpolation.

If your data is sampled faster than needed during the experiment, you can
decimate it without information loss. If your data is sampled more slowly
than needed, there is a possibility that you miss important information
about the dynamics at higher frequencies. Although you can resample the
data at a higher rate, the resampled values occurring between measured
samples do not represent measured information about your system. Instead
of resampling, repeat the experiment using a higher sampling rate.

Tip You should decimate your data when it contains high-frequency noise
outside the frequency range of the system dynamics.

Resampling takes into account how the data behaves between samples, which
you specify when you import the data into the System Identification Tool
GUI (zero-order or first-order hold). For more information about the data
properties you specify before importing the data, see “Importing Data into
the GUI” on page 2-14.

You can resample data using the System Identification Tool GUI or the
resample command. You can only resample time-domain data at uniform
time intervals.

2-102
 Resampling Data

Resampling Data Using the GUI

Use the System Identification Tool GUI to resample time-domain data. To
specify additional options, such as the prefilter order, see “Resampling Data
at the Command Line” on page 2-103.

The System Identification Tool GUI uses idresamp to interpolate or decimate

the data. For more information about this command, type help idresamp at
the prompt.

To create a new data set by resampling the input and output signals:

1 Import time-domain data into the System Identification Tool GUI, as

described in “Importing Data into the GUI” on page 2-14.

2 Drag the data set you want to resample to the Working Data area.

3 In the Resampling factor field, enter the factor by which to multiply the
current sampling interval:
• For decimation (fewer samples), enter a factor greater than 1 to increase
the sampling interval by this factor.
• For interpolation (more samples), enter a factor less than 1 to decrease
the sampling interval by this factor.

Default = 1.

4 In the Data name field, type the name of the new data set. Choose a name
that is unique in the Data Board.

5 Click Insert to add the new data set to the Data Board in the System
Identification Toolbox window.

6 Click Close to close the Resample dialog box.

Resampling Data at the Command Line

Use resample to decimate and interpolate time-domain iddata objects. You
can specify the order of the antialiasing filter as an argument.

2-103
 2 Data Import and Processing

Note resample uses the Signal Processing Toolbox™ command, when this
toolbox is installed on your computer. If this toolbox is not installed, use
idresamp instead. idresamp only lets you specify the filter order, whereas
resample also lets you specify filter coefficients and the design parameters
of the Kaiser window.

To create a new iddata object datar by resampling data, use the following
syntax:

datar = resample(data,P,Q,filter_order)

In this case, P and Q are integers that specify the new sampling interval: the
new sampling interval is Q/P times the original one. You can also specify the
order of the resampling filter as a fourth argument filter_order, which is
an integer (default is 10). For detailed information about resample, see the
corresponding reference page.

For example, resample(data,1,Q) results in decimation with the sampling

interval modified by a factor Q.

The next example shows how you can increase the sampling rate by a factor of
1.5 and compare the signals:

plot(u)
ur = resample(u,3,2);
plot(u,ur)

When the Signal Processing Toolbox product is not installed, using resample
calls idresamp instead.

idresamp uses the following syntax:

datar = idresamp(data,R,filter_order)

In this case, R=Q/P, which means that data is interpolated by a factor P and
then decimated by a factor Q. To learn more about idresamp, type help
idresamp.

2-104
 Resampling Data

The data.InterSample property of the iddata object is taken into account

during resampling (for example, first-order hold or zero-order hold). For more
information, see “iddata Properties” on page 2-52.

Resampling Data Without Aliasing Effects

Typically, you decimate a signal to remove the high-frequency contributions
that result from noise from the total energy. Ideally, you want to remove the
energy contribution due to noise and preserve the energy density of the signal.

The command resample performs the decimation without aliasing effects.

This command includes a factor of T to normalize the spectrum and preserve
the energy density after decimation. For more information about spectrum
normalization, see “Spectrum Normalization” on page 3-8.

If you use manual decimation instead of resample—by picking every fourth

sample from the signal, for example—the energy contributions from higher
frequencies are folded back into the lower frequencies. Because the total
signal energy is preserved by this operation and this energy must now be
squeezed into a smaller frequency range, the amplitude of the spectrum at
each frequency increases. Thus, the energy density of the decimated signal
is not constant.

2-105
 2 Data Import and Processing

The following example illustrates how resample avoids folding effects:

% Construct fourth-order MA-process

m0 = idpoly(1,[ ],[1 1 1 1]);
% Generate error signal
e = idinput(2000,'rgs');
e = iddata([],e,'Ts',1);
% Simulate the output using error signal
y = sim(m0,e);
% Estimate signal spectrum
g1 = spa(y);
% Estimate spectrum of modified signal including
% every fourth sample of the original signal.
% This command automatically sets Ts to 4.
g2 = spa(y(1:4:2000));
% Plot frequency response to view folding effects
ffplot(g1,g2)
% Estimate spectrum after prefiltering that does not
% introduce folding effects
g3 = spa(resample(y,1,4));
figure
ffplot(g1,g3)

2-106
 Resampling Data

Folding Effects with Manual Decimation

2-107
 2 Data Import and Processing

Use resample to decimate the signal before estimating the spectrum and plot
the frequency response, as follows:

g3 = spa(resample(y,1,4));
figure
ffplot(g1,g3)

The following figure shows that the estimated spectrum of the resampled
signal has the same amplitude as the original spectrum. Thus, there is no
indication of folding effects when you use resample to eliminate aliasing.

No Folding Effects When Using resample

See Also
For a detailed discussion about handling disturbances, see the chapter on
preprocessing data in System Identification: Theory for the User, Second
Edition, by Lennart Ljung, Prentice Hall PTR, 1999.

2-108
 Filtering Data

Filtering Data
In this section...
“Supported Filters” on page 2-109
“Choosing to Prefilter Your Data” on page 2-109
“How to Filter Data Using the GUI” on page 2-110
“How to Filter Data at the Command Line” on page 2-113
“See Also” on page 2-116

Supported Filters
You can filter the input and output signals through a linear filter before
estimating a model in the System Identification Tool GUI or at the command
line. How you want to handle the noise in the system determines whether it
is appropriate to prefilter the data.

The filter available in the System Identification Tool GUI is a fifth-order

(passband) Butterworth filter. If you need to specify a custom filter, use the
idfilt command.

Choosing to Prefilter Your Data

Prefiltering data can help remove high-frequency noise or low-frequency
disturbances (drift). The latter application is an alternative to subtracting
linear trends from the data, as described in “Handling Offsets and Trends
in Data” on page 2-97.

In addition to minimizing noise, prefiltering lets you focus your model on

specific frequency bands. The frequency range of interest often corresponds
to a passband over the breakpoints on a Bode plot. For example, if you are
modeling a plant for control-design applications, you might prefilter the data
to specifically enhance frequencies around the desired closed-loop bandwidth.

Prefiltering the input and output data through the same filter does not change
the input-output relationship for a linear system. However, prefiltering does
change the noise characteristics and affects the estimated model of the system.

2-109
 2 Data Import and Processing

To get a reliable noise model, avoid prefiltering the data. Instead, set
the Focus property of the estimation algorithm to Simulation. For more
information about the Focus property, see the Algorithm Properties
reference page.

Note When you prefilter during model estimation, the filtered data is used to
only model the input-to-output dynamics. However, the disturbance model is
calculated from the unfiltered data.

How to Filter Data Using the GUI

• “Filtering Time-Domain Data in the GUI” on page 2-110
• “Filtering Frequency-Domain or Frequency-Response Data in the GUI” on
page 2-112

Filtering Time-Domain Data in the GUI

The System Identification Tool GUI lets you filter time-domain data using a
fifth-order Butterworth filter by enhancing or selecting specific passbands.

To create a filtered data set:

1 Import time-domain data into the System Identification Tool GUI, as

described in “Importing Data into the GUI” on page 2-14.

2 Drag the data set you want you want to filter to the Working Data area.

3 Select <--Preprocess > Filter. By default, this selection shows a

periodogram of the input and output spectra (see the etfe reference page).

Note To display smoothed spectral estimates instead of the periodogram,

select Options > Spectral analysis. This spectral estimate is computed
using spa and your previous settings in the Spectral Model dialog box.
To change these settings, select <--Estimate > Spectral model in the
System Identification Tool GUI, and specify new model settings.

2-110
 Filtering Data

4 If your data contains multiple input/output channels, in the Channel

menu, select the channel pair you want to view. Although you view only
one channel pair at a time, the filter applies to all input/output channels
in this data set.

5 Select the data of interest using one of the following ways:

• Graphically — Draw a rectangle with the mouse on either the

input-signal or the output-signal plot to select the desired frequency
interval. Your selection is displayed on both plots regardless of the plot
on which you draw the rectangle. The Range field is updated to match
the selected region. If you need to clear your selection, right-click the
plot.
• Specify the Range — Edit the beginning and the end frequency values.
For example:
8.5 20.0 (rad/s).

Tip To change the frequency units from rad/s to Hz, select

Style > Frequency (Hz). To change the frequency units from Hz to
rad/s, select Style > Frequency (rad/s).

6 In the Range is list, select one of the following:

• Pass band — Allows data in the selected frequency range.

• Stop band — Excludes data in the selected frequency range.

7 Click Filter to preview the filtered results. If you are satisfied, go to step 8.
Otherwise, return to step 5.

8 In the Data name field, enter the name of the data set containing the
selected data.

9 Click Insert to save the selection as a new data set and add it to the Data
Board.

10 To select another range, repeat steps 5 to 9.

2-111
 2 Data Import and Processing

Filtering Frequency-Domain or Frequency-Response Data in

the GUI
For frequency-domain and frequency-response data, filtering is equivalent to
selecting specific data ranges.

To select a range of data in frequency-domain or frequency-response data:

1 Import data into the System Identification Tool GUI, as described in

“Importing Data into the GUI” on page 2-14.

2 Drag the data set you want you want to filter to the Working Data area.

3 Select <--Preprocess > Select range. This selection displays one of the
following plots:
• Frequency-domain data — Plot shows the absolute of the squares of the
input and output spectra.
• Frequency-response data — Top axes show the frequency response
magnitude equivalent to the ratio of the output to the input, and the
bottom axes show the ratio of the input signal to itself, which has the
value of 1 at all frequencies.

4 If your data contains multiple input/output channels, in the Channel

menu, select the channel pair you want to view. Although you view only
one channel pair at a time, the filter applies to all input/output channels
in this data set.

5 Select the data of interest using one of the following ways:

• Graphically — Draw a rectangle with the mouse on either the

input-signal or the output-signal plot to select the desired frequency
interval. Your selection is displayed on both plots regardless of the plot
on which you draw the rectangle. The Range field is updated to match
the selected region.
If you need to clear your selection, right-click the plot.
• Specify the Range — Edit the beginning and the end frequency values.
For example:
8.5 20.0 (rad/s).

2-112
 Filtering Data

Tip If you need to change the frequency units from rad/s to Hz, select
Style > Frequency (Hz). To change the frequency units from Hz to
rad/s, select Style > Frequency (rad/s).

6 In the Range is list, select one of the following:

• Pass band — Allows data in the selected frequency range.

• Stop band — Excludes data in the selected frequency range.

7 In the Data name field, enter the name of the data set containing the
selected data.

8 Click Insert. This action saves the selection as a new data set and adds
it to the Data Board.

9 To select another range, repeat steps 5 to 8.

How to Filter Data at the Command Line

• “Simple Passband Filter” on page 2-113
• “Defining a Custom Filter” on page 2-114
• “Causal and Noncausal Filters” on page 2-115

Simple Passband Filter

Use idfilt to apply passband and other custom filters to a time-domain or
a frequency-domain iddata object.

In general, you can specify any custom filter. Use this syntax to filter an
iddata object data using the filter called filter:

fdata = idfilt(data,filter)

In the simplest case, you can specify a passband filter for time-domain data
using the following syntax:

fdata = idfilt(data,[wl wh])

2-113
 2 Data Import and Processing

In this case, w1 and wh represent the low and high frequencies of the passband,
respectively.

You can specify several passbands, as follows:

filter=[[w1l,w1h];[ w2l,w2h]; ....;[wnl,wnh]]

The filter is an n-by-2 matrix, where each row defines a passband in radians
per second.

To define a stopband between ws1 and ws2, use

filter = [0 ws1; ws2 Nyqf]

where, Nyqf is the Nyquist frequency.

For time-domain data, the passband filtering is cascaded Butterworth

filters of specified order. The default filter order is 5. The Butterworth
filter is the same as butter in the Signal Processing Toolbox product. For
frequency-domain data, select the indicated portions of the data to perform
passband filtering.

Defining a Custom Filter

Use idfilt to apply passband and other custom filters to a time-domain or
a frequency-domain iddata object.

In general, you can specify any custom filter. Use this syntax to filter an
iddata object data using the filter called filter:

fdata = idfilt(data,filter)

You can define a general single-input/single-output (SISO) system for filtering

time-domain or frequency-domain data. For frequency-domain only, you can
specify the (nonparametric) frequency response of the filter.

You use this syntax to filter an iddata object data using a custom filter
specified by filter:

fdata = idfilt(data,filter)

filter can be also any of the following:

2-114
 Filtering Data

filter = idm
filter = {num,den}
filter = {A,B,C,D}

idm is a SISO idmodel or LTI object. For more information about LTI objects,
see the Control System Toolbox documentation.

{num,den} defines the filter as a transfer function as a cell array of numerator

and denominator filter coefficients.

{A,B,C,D} is a cell array of SISO state-space matrices.

Specifically for frequency-domain data, you specify the frequency response

of the filter:

filter = Wf

Here, Wf is a vector of real or complex values that define the filter

frequency response, where the inputs and outputs of data at frequency
data.Frequency(kf) are multiplied by Wf(kf). Wf is a column vector with
the length equal to the number of frequencies in data.

When data contains several experiments, Wf is a cell array with the length
equal to the number of experiments in data.

Causal and Noncausal Filters

For time-domain data, the filtering is causal by default. Causal filters
typically introduce a phase shift in the results. To use a noncausal zero-phase
filter (corresponding to filtfilt in the Signal Processing Toolbox product),
specify a third argument in idfilt:

fdata = idfilt(data,filter,'noncausal')

For frequency-domain data, the signals are multiplied by the frequency

response of the filter. With the filters defined as passband filters, this
calculation gives ideal, zero-phase filtering (“brick wall filters”). Frequencies
that have been assigned zero weight by the filter (outside the passband or
via frequency response) are removed.

2-115
 2 Data Import and Processing

When you apply idfilt to an idfrd data object, the data is first converted
to a frequency-domain iddata object (see “Transforming Between
Frequency-Domain and Frequency-Response Data” on page 2-130). The result
is an iddata object.

See Also
To learn how to filter data during linear model estimation instead, you can
set the Focus property of the estimation algorithm to Filter and specify the
filter characteristics. For more information about model properties, see the
Algorithm Properties reference page.

For more information about prefiltering data, see the chapter on preprocessing
data in System Identification: Theory for the User, Second Edition, by Lennart
Ljung, Prentice Hall PTR, 1999.

For practical examples of prefiltering data, see the section on posttreatment

of data in Modeling of Dynamic Systems, by Lennart Ljung and Torkel Glad,
Prentice Hall PTR, 1994.

2-116
 Generating Data Using Simulation

Generating Data Using Simulation

In this section...
“Commands for Generating and Simulating Data” on page 2-117
“Example – Creating Data with Periodic Inputs” on page 2-118
“Example – Generating Data Using Simulation” on page 2-119
“Simulating Data Using Other MathWorks Products” on page 2-120

Commands for Generating and Simulating Data

You can generate input data and simulate output data using a specified
model structure.

Simulating output data requires that you have a parametric model. For more
information about commands for constructing models, see “Commands for
Constructing Model Structures” on page 1-16.

To generate input data, use idinput to construct a signal with the desired
characteristics, such as a random Gaussian or binary signal or a sinusoid.
idinput returns a matrix of input values.

The following table lists the commands you can use to simulate output data.
For more information about these commands, see the corresponding reference
pages.

Commands for Generating and Simulating Data

Command Description Example

iddata Constructs an iddata To construct input data data, use the

object with input following command:
channels only.
data = iddata([ ],[u v])

u is the input data, and v is white noise.

2-117
 2 Data Import and Processing

Commands for Generating and Simulating Data (Continued)

Command Description Example

idinput Constructs a signal

with the desired u = iddata([],...
characteristics, such idinput(400,'rbs',[0 0.3]));
as a random Gaussian
or binary signal or a
sinusoid, and returns a
matrix of input values.
sim Simulates response data To simulate the model output y for a given
based on existing linear input, use the following command:
or nonlinear parametric
model in the MATLAB y = sim(m,data)
workspace.
m is the model object name, and data is input
data matrix or iddata object.

Example – Creating Data with Periodic Inputs

1 Create a periodic input for two inputs and consisting of five periods, where
each period is 300 samples.

per_u = idinput([300 2 5])

2 Create an iddata object using the periodic input and leaving the output
empty.

u = iddata([],per_u,'Period',...
[300; 300]);

You can use the periodic input to simulate the output, and the use etfe to
compute the estimated response of the model.

% Construct polynomial model

m0 =idpoly([1 -1.5 0.7],[0 1 0.5]);
% Construct random binary input
u = idinput([10 1 150],'rbs');

2-118
 Generating Data Using Simulation

% Construct input data and noise

u = iddata([],u,'Period',10);
e = iddata([],randn(1500,1));
% Simulate model output with noise
y = sim(m0,[u e])
% Estimate frequency response
g = etfe([y u])
% Generate Bode plot
bode(g,'x',m0)

For periodic input, etfe honors the period and computes the frequency
response using an appropriate frequency grid. In this case, the Bode plot
shows a good fit at the five excited frequencies.

Example – Generating Data Using Simulation

This example demonstrates how you can create input data and a model, and
then use the data and the model to simulate output data. You create the
ARMAX model and simulate output data with random binary input u.

1 Load the three-input and one-output sample data.

load iddata8

2 Construct an ARMAX model, using the following commands:

A = [1 -1.2 0.7];
B(1,:) = [0 1 0.5 0.1]; % first input
B(2,:) = [0 1.5 -0.5 0]; % second input
B(3,:) = [0 -0.1 0.5 -0.1]; % third input
C = [1 0 0 0 0];
Ts = 1;
m = idpoly(A,B,C,'Ts',1);

In this example, the leading zeros in the B matrix indicate the input delay
(nk), which is 1 for each input channel. The trailing zero in B(2,:) makes
the number of coefficients equal for all channels.

3 Construct pseudorandom binary data for input to the simulation.

u = idinput([200,3],'prbs');

2-119
 2 Data Import and Processing

4 Simulate the model output.

sim(m,u)

5 Compare model output to measured data to see how well the models
captures the underlying dynamics.

compare(z8,m)

Simulating Data Using Other MathWorks Products

You can also simulate data using the Simulink® and Signal Processing
Toolbox software. Data simulated outside the System Identification Toolbox
product must be in the MATLAB workspace. For more information about
simulating models using the Simulink software, see “Simulating Model
Output” on page 10-6.

2-120
 Transforming Between Time- and Frequency-Domain Data

Transforming Between Time- and Frequency-Domain Data

In this section...
“Transforming Data Domain in the GUI” on page 2-121
“Transforming Data Domain at the Command Line” on page 2-128

Transforming Data Domain in the GUI

• “Transforming Time-Domain Data” on page 2-121
• “Transforming Frequency-Domain Data” on page 2-125
• “Transforming Frequency-Response Data” on page 2-126
• “See Also” on page 2-128

Transforming Time-Domain Data

In the System Identification Tool GUI, time-domain data has an icon with a
white background. You can transform time-domain data to frequency-domain
or frequency-response data. The frequency values of the resulting frequency
vector range from 0 to the Nyquist frequency fS = π Ts , where Ts is the
sampling interval.

Transforming from time-domain to frequency-response data is equivalent to

estimating a model from the data using the spafdr method.

2-121
 2 Data Import and Processing

1 In the System Identification Tool GUI, drag the icon of the data you want to
transform to the Working Data rectangle, as shown in the following figure.

2-122
 Transforming Between Time- and Frequency-Domain Data

2 In the Operations area, select <--Preprocess > Transform data in the

drop-down menu to open the Transform Data dialog box.

2-123
 2 Data Import and Processing

3 In the Transform to drop-down list, select one of the following:

• Frequency Function — Create a new idfrd object using the spafdr

method. Go to step 4.

• Frequency Domain Data — Create a new iddata object using the fft
method. Go to step 6.

4 In the Frequency Spacing list, select the spacing of the frequencies at

which the frequency function is estimated:
• linear — Uniform spacing of frequency values between the endpoints.
• logarithmic — Base-10 logarithmic spacing of frequency values
between the endpoints.

5 In the Number of Frequencies field, enter the number of frequency

values.

6 In the Name of new data field, type the name of the new data set. This
name must be unique in the Data Board.

7 Click Transform to add the new data set to the Data Board in the System
Identification Tool GUI.

8 Click Close to close the Transform Data dialog box.

2-124
 Transforming Between Time- and Frequency-Domain Data

Transforming Frequency-Domain Data

In the System Identification Tool GUI, frequency-domain data has an icon
with a green background. You can transform frequency-domain data to
time-domain or frequency-response (frequency-function) data.

Transforming from time-domain or frequency-domain data to

frequency-response data is equivalent to estimating a nonparametric model of
the data using the spafdr method.

1 In the System Identification Tool GUI, drag the icon of the data you want
to transform to the Working Data rectangle.

2 Select <--Preprocess > Transform data.

3 In the Transform to list, select one of the following:

• Frequency Function — Create a new idfrd object using the spafdr

method. Go to step 4.
• Time Domain Data — Create a new iddata object using the ifft
(inverse fast Fourier transform) method. Go to step 6.

4 In the Frequency Spacing list, select the spacing of the frequencies at

which the frequency function is estimated:
• linear — Uniform spacing of frequency values between the endpoints.
• logarithmic — Base-10 logarithmic spacing of frequency values
between the endpoints.

5 In the Number of Frequencies field, enter the number of frequency

values.

6 In the Name of new data field, type the name of the new data set. This
name must be unique in the Data Board.

7 Click Transform to add the new data set to the Data Board in the System
Identification Tool GUI.

8 Click Close to close the Transform Data dialog box.

2-125
 2 Data Import and Processing

Transforming Frequency-Response Data

In the System Identification Tool GUI, frequency-response data has an icon
with a yellow background. You can transform frequency-response data to
frequency-domain data (iddata object) or to frequency-response data with a
different frequency resolution.

When you select to transform single-input/single-output (SISO)

frequency-response data to frequency-domain data, the toolbox creates
outputs that equal the frequency responses, and inputs equal to 1. Therefore,
the ratio between the Fourier transform of the output and the Fourier
transform of the input is equal to the system frequency response.

For the multiple-input case, the toolbox transforms the frequency-response

data to frequency-domain data as if each input contributes independently to
the entire output of the system and then combines information. For example,
if a system has three inputs, u1, u2, and u3 and two frequency samples, the
input matrix is set to:

⎡1 0 0⎤
⎢1 0 0 ⎥⎥
⎢
⎢0 1 0⎥
⎢ ⎥
⎢0 1 0⎥
⎢0 0 1⎥
⎢ ⎥
⎢⎣0 0 1 ⎥⎦

In general, for nu inputs and ns samples (the number of frequencies), the

input matrix has nu columns and (ns ⋅ nu) rows.

Note To create a separate experiment for the response from each input,
see “Transforming Between Frequency-Domain and Frequency-Response
Data” on page 2-130.

When you transform frequency-response data by changing its frequency

resolution, you can modify the number of frequency values by changing
between linear or logarithmic spacing. You might specify variable frequency
spacing to increase the number of data points near the system resonance

2-126
 Transforming Between Time- and Frequency-Domain Data

frequencies, and also make the frequency vector coarser in the region outside
the system dynamics. Typically, high-frequency noise dominates away
from frequencies where interesting system dynamics occur. The System
Identification Tool GUI lets you specify logarithmic frequency spacing, which
results in a variable frequency resolution.

Note The spafdr command lets you lets you specify any variable frequency
resolution.

1 In the System Identification Tool GUI, drag the icon of the data you want
to transform to the Working Data rectangle.

2 Select <--Preprocess > Transform data.

3 In the Transform to list, select one of the following:

• Frequency Domain Data — Create a new iddata object. Go to step 6.

• Frequency Function — Create a new idfrd object with different
resolution (number and spacing of frequencies) using the spafdr method.
Go to step 4.

4 In the Frequency Spacing list, select the spacing of the frequencies at

which the frequency function is estimated:
• linear — Uniform spacing of frequency values between the endpoints.
• logarithmic — Base-10 logarithmic spacing of frequency values
between the endpoints.

5 In the Number of Frequencies field, enter the number of frequency

values.

6 In the Name of new data field, type the name of the new data set. This
name must be unique in the Data Board.

7 Click Transform to add the new data set to the Data Board in the System
Identification Tool GUI.

8 Click Close to close the Transform Data dialog box.

2-127
 2 Data Import and Processing

See Also
For a description of time-domain, frequency-domain, and frequency-response
data, see “Importing Data into the MATLAB Workspace” on page 2-7.

To learn how to transform data at the command line instead of the GUI, see
“Transforming Data Domain at the Command Line” on page 2-128.

Transforming Data Domain at the Command Line

• “Supported Data Transformations” on page 2-128
• “Transforming Between Time and Frequency Domain” on page 2-129
• “Transforming Between Frequency-Domain and Frequency-Response
Data” on page 2-130
• “See Also” on page 2-131

Supported Data Transformations

The following table shows the different ways you can transform data from one
data domain to another. If the transformation is supported for a given row
and column combination in the table, the method used by the software is
listed in the cell at their intersection.

Original Data To Time Domain To Frequency To Frequency

Format (iddata object) Domain Function
(iddata object) (idfrd object)
Time Domain No. Yes, using fft. Yes, using etfe,
(iddata object) spa, or spafdr.
Frequency Yes, using ifft. No. Yes, using etfe,
Domain spa, or spafdr.
(iddata object)

Frequency No. Yes. Calculation Yes. Calculates

Function creates a frequency
(idfrd object) frequency-domain function with
iddata object different

2-128
 Transforming Between Time- and Frequency-Domain Data

Original Data To Time Domain To Frequency To Frequency

Format (iddata object) Domain Function
(iddata object) (idfrd object)

that has the resolution

same ratio (number and
between output spacing of
and input as the frequencies)
original idfrd using spafdr.
object.

Transforming Between Time and Frequency Domain

The iddata object stores time-domain or frequency-domain data. The
following table summarizes the commands for transforming data between
time and frequency domains.

Command Description Syntax Example

fft Transforms time-domain To transform time-domain

data to the frequency iddata object t_data to
domain. frequency-domain iddata
object f_data with N
You can specify N, the
frequency points, use:
number of frequency
values. f_data =
fft(t_data,N)
ifft Transforms To transform
frequency-domain data frequency-domainiddata
to the time domain. object f_data to
Frequencies are linear and time-domain iddata
equally spaced. object t_data, use:

t_data =
ifft(f_data)

2-129
 2 Data Import and Processing

Transforming Between Frequency-Domain and

Frequency-Response Data
You can transform frequency-response data to frequency-domain data
(iddata object). The idfrd object represents complex frequency-response of
the system at different frequencies. For a description of this type of data, see
“Importing Frequency-Response Data into MATLAB” on page 2-11.

When you select to transform single-input/single-output (SISO)

frequency-response data to frequency-domain data, the toolbox creates
outputs that equal the frequency responses, and inputs equal to 1. Therefore,
the ratio between the Fourier transform of the output and the Fourier
transform of the input is equal to the system frequency response.

For information about changing the frequency resolution of frequency-response

data to a new constant or variable (frequency-dependent) resolution, see the
spafdr reference page. You might use this advanced feature to increase the
number of data points near the system resonance frequencies and make the
frequency vector coarser in the region outside the system dynamics. Typically,
high-frequency noise dominates away from frequencies where interesting
system dynamics occur.

Note You cannot transform an idfrd object to a time-domain iddata object.

To transform an idfrd object with the name idfrdobj to a frequency-domain

iddata object, use the following syntax:

dataf = iddata(idfrdobj)

The resulting frequency-domain iddata object contains values at the same

frequencies as the original idfrd object.

For the multiple-input case, the toolbox represents frequency-response data

as if each input contributes independently to the entire output of the system
and then combines information. For example, if a system has three inputs,
u1, u2, and u3 and two frequency samples, the input matrix is set to:

2-130
 Transforming Between Time- and Frequency-Domain Data

⎡1 0 0⎤
⎢1 0 0 ⎥⎥
⎢
⎢0 1 0⎥
⎢ ⎥
⎢0 1 0⎥
⎢0 0 1⎥
⎢ ⎥
⎢⎣0 0 1 ⎥⎦

In general, for nu inputs and ns samples, the input matrix has nu columns
and (ns ⋅ nu) rows.

If you have ny outputs, the transformation operation produces an output

matrix has ny columns and (ns ⋅ nu) rows using the values in the complex
frequency response G(iw) matrix (ny-by-nu-by-ns). In this example, y1 is
determined by unfolding G(1,1,:), G(1,2,:), and G(1,3,:) into three
column vectors and vertically concatenating these vectors into a single column.
Similarly, y2 is determined by unfolding G(2,1,:), G(2,2,:), and G(2,3,:)
into three column vectors and vertically concatenating these vectors.

If you are working with multiple inputs, you also have the option of
storing the contribution by each input as an independent experiment in a
multiexperiment data set. To transform an idfrd object with the name
idfrdobj to a multiexperiment data set datf, where each experiment
corresponds to each of the inputs in idfrdobj

datf = iddata(idfrdobj,'me')

In this example, the additional argument 'me' specifies that multiple

experiments are created.

By default, transformation from frequency-response to frequency-domain

data strips away frequencies where the response is inf or NaN. To preserve
the entire frequency vector, use datf = iddata(idfrdobj,'inf'). For more
information, type help idfrd/iddata.

See Also
Transforming from time-domain or frequency-domain data to
frequency-response data is equivalent to creating a frequency-response model

2-131
 2 Data Import and Processing

from the data. For more information, see “Identifying Frequency-Response

Models” on page 3-2.

2-132
 Manipulating Complex-Valued Data

Manipulating Complex-Valued Data

In this section...
“Supported Operations for Complex Data” on page 2-133
“Processing Complex iddata Signals at the Command Line” on page 2-133

Supported Operations for Complex Data

System Identification Toolbox estimation algorithms support complex data.
For example, the following estimation commands estimate complex models
from complex data: ar, armax, arx, bj, covf, ivar, iv4, oe, pem, spa, and
n4sid.

Model transformation routines, such as freqresp and zpkdata, work for

complex-valued models. However, they do not provide pole-zero confidence
regions. For complex models, the parameter variance-covariance information
refers to the complex-valued parameters and the accuracy of the real and
imaginary is not computed separately.

The display commands compare and plot also work with complex-valued
data and models, but only show the absolute values of the signals. To plot
the real and imaginary parts of the data separately, use plot(real(data))
and plot(imag(data)), respectively.

Processing Complex iddata Signals at the Command

Line
If the iddata object data contains complex values, you can use the following
commands to process the complex data and create a new iddata object.

Command Description
abs(data) Absolute value of complex signals in iddata object.
angle(data) Phase angle (in radians) of each complex signals in
iddata object.

2-133
 2 Data Import and Processing

Command Description
complex(data) For time-domain data, this command makes the iddata
object complex—even when the imaginary parts are
zero. For frequency-domain data that only stores
the values for nonnegative frequencies, such that
realdata(data)=1, it adds signal values for negative
frequencies using complex conjugation.
imag(data) Selects the imaginary parts of each signal in iddata
object.
isreal(data) 1 when data (time-domain or frequency-domain)
contains only real input and output signals, and returns
0 when data (time-domain or frequency-domain)
contains complex signals.
real(data) Real part of complex signals in iddata object.
realdata(data) Returns a value of 1 when data is a real-valued,
time-domain signal, and returns 0 otherwise.

For example, suppose that you create a frequency-domain iddata object Datf
by applying fft to a real-valued time-domain signal to take the Fourier
transform of the signal. The following is true for Datf:

isreal(Datf) = 0
realdata(Datf) = 1

2-134
 3

Linear Model Identification

• “Identifying Frequency-Response Models” on page 3-2

• “Identifying Impulse-Response Models” on page 3-11
• “Identifying Low-Order Transfer Functions (Process Models)” on page 3-20
• “Identifying Input-Output Polynomial Models” on page 3-39
• “Identifying State-Space Models” on page 3-73
• “Refining Linear Parametric Models” on page 3-104
• “Extracting Parameter Values from Linear Models” on page 3-109
• “Extracting Dynamic Model and Noise Model Separately” on page 3-111
• “Transforming Between Discrete-Time and Continuous-Time
Representations” on page 3-113
• “Transforming Between Linear Model Representations” on page 3-118
• “Subreferencing Model Objects” on page 3-120
• “Concatenating Model Objects” on page 3-125
• “Merging Model Objects” on page 3-129
 3 Linear Model Identification

Identifying Frequency-Response Models

In this section...
“What Is a Frequency-Response Model?” on page 3-2
“Data Supported by Frequency-Response Models” on page 3-3
“How to Estimate Frequency-Response Models in the GUI” on page 3-3
“How to Estimate Frequency-Response Models at the Command Line” on
page 3-5
“Options for Computing Spectral Models” on page 3-5
“Options for Frequency Resolution” on page 3-6
“Spectrum Normalization” on page 3-8

What Is a Frequency-Response Model?

You can estimate frequency-response models and visualize the responses on
a Bode plot, which shows the amplitude change and the phase shift as a
function of the sinusoid frequency.

The frequency-response function describes the steady-state response of a

system to sinusoidal inputs. For a linear system, a sinusoidal input of a
specific frequency results in an output that is also a sinusoid with the same
frequency, but with a different amplitude and phase. The frequency-response
function describes the amplitude change and phase shift as a function of
frequency.

For a discrete-time system sampled with a time interval T, the

frequency-response model G(z) relates the Z-transforms of the input U(z)
and output Y(z):

Y ( z) = G( z)U ( z)

In other words, the frequency-response function, G(eiwT), is the Laplace

transform of the impulse response that is evaluated on the imaginary axis.
The frequency-response function is the transfer function G(z) evaluated on
the unit circle.

3-2
 Identifying Frequency-Response Models

Data Supported by Frequency-Response Models

You can estimate spectral analysis models from data with the following
characteristics:

• Complex or real data.

• Time- or frequency-domain iddata or idfrd data object. To learn
more about estimating time-series models, see Chapter 6, “Time Series
Identification”.
• Single- or multiple-output data.

How to Estimate Frequency-Response Models in the

GUI
You must have already imported your data into the GUI and performed any
necessary preprocessing operations. For more information, see Chapter 2,
“Data Import and Processing”.

To estimate frequency-response models in the System Identification Tool GUI:

1 In the System Identification Tool GUI, select Estimate > Spectral

models to open the Spectral Model dialog box.

2 In the Method list, select the spectral analysis method you want to use.
For information about each method, see “Options for Computing Spectral
Models” on page 3-5.

3 Specify the frequencies at which to compute the spectral model in one of

the following ways:
• In the Frequencies field, enter either a vector of values, a MATLAB
expression that evaluates to a vector, or a variable name of a vector in
the MATLAB workspace. For example, logspace(-1,2,500).
• Use the combination of Frequency Spacing and Frequencies to
construct the frequency vector of values:
– In the Frequency Spacing list, select Linear or Logarithmic
frequency spacing.

3-3
 3 Linear Model Identification

Note For etfe, only the Linear option is available.

– In the Frequencies field, enter the number of frequency points.

For time-domain data, the frequency ranges from 0 to the Nyquist
frequency. For frequency-domain data, the frequency ranges from the
smallest to the largest frequency in the data set.

4 In the Frequency Resolution field, enter the frequency resolution, as

described in “Options for Frequency Resolution” on page 3-6. To use the
default value, enter default or, equivalently, the empty matrix [].

5 In the Model Name field, enter the name of the correlation analysis model.
The model name should be unique in the Model Board.

6 Click Estimate to add this model to the Model Board in the System
Identification Tool GUI.

7 In the Spectral Model dialog box, click Close.

8 To view the frequency-response plot, select the Frequency resp check box
in the System Identification Tool GUI. For more information about working
with this plot, see “Frequency Response Plots” on page 8-38.

9 To view the estimated disturbance spectrum, select the Noise spectrum

check box in the System Identification Tool GUI. For more information
about working with this plot, see “Noise Spectrum Plots” on page 8-46.

10 After estimating the model, see Chapter 8, “Model Analysis”, to validate

the model.

To export the model to the MATLAB workspace, drag it to the To Workspace

rectangle in the System Identification Tool GUI. You can retrieve the
responses from the resulting idfrd model object using the bode or nyquist
command.

3-4
 Identifying Frequency-Response Models

How to Estimate Frequency-Response Models at the

Command Line
You can use the etfe, spa, and spafdr commands to estimate spectral
models. The following table provides a brief description of each command
and usage examples.

The resulting models are stored as idfrd model objects. For detailed
information about the commands and their arguments, see the corresponding
reference page.

Commands for Frequency Response

Command Description Usage

etfe Estimates an empirical To estimate a model m, use the following syntax:
transfer function using
Fourier analysis. m=etfe(data)

spa Estimates a frequency To estimate a model m, use the following syntax:

response with a fixed
frequency resolution m=spa(data)
using spectral analysis.
spafdr Estimates a frequency To estimate a model m, use the following syntax:
response with a variable
frequency resolution m=spafdr(data,R,w)
using spectral analysis.
where R is the resolution vector and w is the frequency
vector.

After estimating the model, see Chapter 8, “Model Analysis” to validate the
model.

Options for Computing Spectral Models

This section describes how to select the method for computing spectral models
in the estimation procedures “How to Estimate Frequency-Response Models
in the GUI” on page 3-3 and “How to Estimate Frequency-Response Models
at the Command Line” on page 3-5.

3-5
 3 Linear Model Identification

You can choose from the following three spectral-analysis methods:

• etfe (Empirical Transfer Function Estimate)

For input-output data. This method computes the ratio of the Fourier
transform of the output to the Fourier transform of the input.
For time-series data. This method computes a periodogram as the
normalized absolute squares of the Fourier transform of the time series.
ETFE works well for highly resonant systems or narrowband systems.
The drawback of this method is that it requires linearly spaced frequency
values, does not estimate the disturbance spectrum, and does not provide
confidence intervals. ETFE also works well for periodic inputs and
computes exact estimates at multiples of the fundamental frequency of the
input and their ratio.
• spa (SPectral Analysis)
This method is the Blackman-Tukey spectral analysis method, where
windowed versions of the covariance functions are Fourier transformed.
• spafdr (SPectral Analysis with Frequency Dependent Resolution)
This method is a variant of the Blackman-Tukey spectral analysis method
with frequency-dependent resolution. First, the algorithm computes
Fourier transforms of the inputs and outputs. Next, the products of the
transformed inputs and outputs with the conjugate input transform are
smoothed over local frequency regions. The widths of the local frequency
regions can vary as a function of frequency. The ratio of these averages
computes the frequency-response estimate.

Options for Frequency Resolution

• “What Is Frequency Resolution?” on page 3-7
• “Frequency Resolution for etfe and spa” on page 3-7
• “Frequency Resolution for spafdr” on page 3-7
• “etfe Frequency Resolution for Periodic Input” on page 3-8

This section supports the estimation procedures “How to Estimate

Frequency-Response Models in the GUI” on page 3-3 and “How to Estimate
Frequency-Response Models at the Command Line” on page 3-5.

3-6
 Identifying Frequency-Response Models

What Is Frequency Resolution?

Frequency resolution is the size of the smallest frequency for which details
in the frequency response and the spectrum can be resolved by the estimate.
A resolution of 0.1 rad/s means that the frequency response variations at
frequency intervals at or below 0.1 rad/s are not resolved.

Note Finer resolution results in greater uncertainty in the model estimate.

Specifying the frequency resolution for etfe and spa is different than for
spafdr.

Frequency Resolution for etfe and spa

For etfe and spa, the frequency resolution is approximately equal to the
following value:

2π ⎛ radians ⎞
⎜ ⎟
M ⎝ sampling interval ⎠

M is a scalar integer that sets the size of the lag window. The value of M
controls the trade-off between bias and variance in the spectral estimate.

The default value of M for spa is good for systems without sharp resonances.
For etfe, the default value of M gives the maximum resolution.

A large value of M gives good resolution, but results in more uncertain

estimates. If a true frequency function has sharp peak, you should specify
higher M values.

Frequency Resolution for spafdr

In case of etfe and spa, the frequency response is defined over a uniform
frequency range, 0-Fs/2 radians per second, where Fs is the sampling
frequency—equal to twice the Nyquist frequency. In contrast, spafdr lets
you increase the resolution in a specific frequency range, such as near a
resonance frequency. Conversely, you can make the frequency grid coarser in
the region where the noise dominates—at higher frequencies, for example.

3-7
 3 Linear Model Identification

Such customizing of the frequency grid assists in the estimation process by

achieving high fidelity in the frequency range of interest.

For spafdr, the frequency resolution around the frequency k is the value R(k).
You can enter R(k) in any one of the following ways:

• Scalar value of the constant frequency resolution value in radians per

second.

Note The scalar R is inversely related to the M value used for etfe and
spa.

• Vector of frequency values the same size as the frequency vector.

• Expression using MATLAB workspace variables and evaluates to a
resolution vector that is the same size as the frequency vector.

The default value of the resolution for spafdr is twice the difference between
neighboring frequencies in the frequency vector.

etfe Frequency Resolution for Periodic Input

If the input data is marked as periodic and contains an integer number of
periods (data.Period is an integer), etfe computes the frequency response at

frequencies T
2πk
( Period
k
) where k = 1, 2,..., Period
.

For periodic data, the frequency resolution is ignored.

Spectrum Normalization
The spectrum of a signal is the square of the Fourier transform of the
signal. The spectral estimate using the commands spa, spafdr, and etfe is
normalized by the sampling interval T:

M
Φ y (ω) = T ∑ Ry (kT ) e−iwT WM (k)
k=− M

3-8
 Identifying Frequency-Response Models

where WM(k) is the lag window, and M is the width of the lag window. The
output covariance Ry(kT) is given by the following discrete representation:

N
ˆ (kT ) = 1
R y
N
∑ y(lT − kT ) y(lT )
l =1

Because there is no scaling in a discrete Fourier transform of a vector, the

purpose of T is to relate the discrete transform of a vector to the physically
meaningful transform of the measured signal. This normalization sets the
units of Φ y (ω) as power per radians per unit time, and makes the frequency
units radians per unit time.

The scaling factor of T is necessary to preserve the energy density of the

spectrum after interpolation or decimation.

By Parseval’s theorem, the average energy of the signal must equal the
average energy in the estimated spectrum, as follows:

1 π/T
2π ∫− π / T
Ey2 (t) = Φ y (ω) dω

S1 ≡ Ey2 (t)
1 π/T
2π ∫− π / T
S2 ≡ Φ y (ω) dω

To compare the left side of the equation (S1) to the right side (S2), enter the
following commands in the MATLAB Command Window:

load iddata1
% Create time-series iddata object
y = z1(:,1,[]);
% Define sample interval from the data
T = y.Ts;
% Estimate frequency response
sp = spa(y);
% Remove spurious dimensions
phiy = squeeze(sp.spec);
% Compute average energy from the estimated

3-9
 3 Linear Model Identification

% energy spectrum, where S1 is scaled by T

S1 = sum(phiy)/length(phiy)/T
% Compute average energy of the signal
S2 = sum(y.y.^2)/size(y,1)

In this code, phiy contains Φ y (ω) between ω = 0 and ω = π T with the

frequency step given as follows:

⎛ π ⎞
⎜ ⎟
⎝ T ⋅ length(phiy) ⎠

MATLAB computes the following values for S1 and S2:

S1 =

19.2076
S2 =

19.4646

Thus, the average energy of the signal approximately equals the average
energy in the estimated spectrum.

3-10
 Identifying Impulse-Response Models

Identifying Impulse-Response Models

In this section...
“What Is Time-Domain Correlation Analysis?” on page 3-11
“Data Supported by Correlation Analysis” on page 3-12
“How to Estimate Impulse and Step Response Models Using the GUI” on
page 3-12
“How to Estimate Impulse and Step Response Models at the Command
Line” on page 3-14
“How to Compute Response Values” on page 3-15
“How to Identify Delay Using Transient-Response Plots” on page 3-16
“Algorithm for Correlation Analysis” on page 3-18

What Is Time-Domain Correlation Analysis?

Time-domain correlation analysis is a nonparametric estimate of transient
response of dynamic systems, which computes a finite impulse response (FIR)
model from the data. Correlation analysis assumes a linear system and does
not require a specific model structure.

There are two types of transient response for a dynamic model:

• Impulse response
Impulse response is the output signal that results when the input is an
impulse and has the following definition for a discrete model:

u(t) = 0 t>0
u(t) = 1 t=0

• Step response

3-11
 3 Linear Model Identification

Step response is the output signal that results from a step input, defined as
follows:

u(t) = 0 t<0
u(t) = 1 t≥0

The response to an input u(t) is equal to the convolution of the impulse

response, as follows:

t
y(t) = ∫0 h ( t − z) ⋅ u( z)dz

Data Supported by Correlation Analysis

You can estimate correlation analysis models from data with the following
characteristics:

• Real or complex time-domain iddata object. To learn about estimating

time-series models, see Chapter 6, “Time Series Identification”.
• Frequency-domain iddata or idfrd object with the sampling interval T≠0.
• Single- or multiple-output data.

How to Estimate Impulse and Step Response Models

Using the GUI
Before you can perform this task, you must have

• Regularly sampled data imported into the System Identification Tool

GUI. See “Importing Time-Domain Data into the GUI” on page 2-16. For
supported data formats, see “Data Supported by Correlation Analysis”
on page 3-12.
• Performed any required data preprocessing operations. To improve the
accuracy of your model, you should detrend your data. See “Ways to
Process Data for System Identification” on page 2-4.

To estimate in the System Identification Tool GUI using time-domain

correlation analysis:

3-12
 Identifying Impulse-Response Models

1 In the System Identification Tool GUI, select Estimate > Correlation

models to open the Correlation Model dialog box.

2 In the Time span (s) field, specify a scalar value as the time interval over
which the impulse or step response is calculated. For a scalar time span T,
the resulting response is plotted from -T/4 to T.

Tip You can also enter a 2-D vector in the format [min_value max_value].

3 In the Order of whitening filter field, specify the filter order.

The prewhitening filter is determined by modeling the input as an

autoregressive process of order N. The algorithm applies a filter of the
form A(q)u(t)=u_F(t). That is, the input u(t) is subjected to an FIR filter A
to produce the filtered signal u_F(t). Prewhitening the input by applying
a whitening filter before estimation might improve the quality of the
estimated impulse response g.

The order of the prewhitening filter, N, is the order of the A filter. N equals
the number of lags. The default value of N is 10, which you can also specify
as [].

4 In the Model Name field, enter the name of the correlation analysis model.
The name of the model should be unique in the Model Board.

5 Click Estimate to add this model to the Model Board in the System
Identification Tool GUI.

6 In the Correlation Model dialog box, click Close.

Next Steps

• Export the model to the MATLAB workspace for further analysis by

dragging it to the To Workspace rectangle in the System Identification
Tool GUI.
• View the transient response plot by selecting the Transient resp check
box in the System Identification Tool GUI. For more information about

3-13
 3 Linear Model Identification

working with this plot and selecting to view impulse- versus step-response,
see “Impulse and Step Response Plots” on page 8-30.

How to Estimate Impulse and Step Response Models

at the Command Line
Before you can perform this task, you must have

• Regularly sampled data. See “Representing Time- and Frequency-Domain

Data Using iddata Objects” on page 2-49. For supported data formats, see
“Data Supported by Correlation Analysis” on page 3-12.
• Performed any required data preprocessing operations. Ro improve the
accuracy of your model, you should detrend your data. See “Ways to
Process Data for System Identification” on page 2-4.

The following tables summarize the commands for computing impulse- and
step-response models. Both impulse and step produce the same FIR model,
but generate different plots. The resulting models are stored as idarx model
objects and contain impulse-response coefficients in the model parameter
B. For detailed information about these commands, see the corresponding
reference page.

Note cra is an alternative method for computing impulse response from

time-domain data only.

Commands for Impulse and Step Response

Command Description Example

impulse Estimates a high-order, To estimate the model m and plot the impulse
noncausal FIR model response, use the following syntax:
using correlation analysis.
m=impulse(data,Time,'pw',N)

where data is a single- or multiple-output

time-domain iddata object, and Time is a
scalar value representing the time interval

3-14
 Identifying Impulse-Response Models

Commands for Impulse and Step Response (Continued)

Command Description Example

over which the impulse or step response is

calculated. For a scalar time span T, the
resulting response is plotted from -T/4 to T.
'pw' and N is an option property-value pair
that specifies the order N of the prewhitening
filter 'pw'.
step Estimates a high-order, To estimate the model m and plot the step
noncausal FIR model response, use the following syntax:
correlation analysis.
step(data,Time)

where data is a single- or multiple-output

time-domain iddata object, and Time is the
time span.

Next Steps

• Perform model analysis. See “Overview of Model Validation and Plots”

on page 8-2.

How to Compute Response Values

You can use impulse and step commands with output arguments to get
the numerical impulse- and step-response vectors as a function of time,
respectively.

To get the numerical response values:

1 Compute the FIR model by applying either impulse or step commands on

the data, as described in “How to Estimate Impulse and Step Response
Models at the Command Line” on page 3-14.

2 Apply the following syntax on the resulting model:

% To compute impulse-response data

[y,t,ysd] = impulse(model)

3-15
 3 Linear Model Identification

% To compute step-response data

[y,t,ysd] = step(model)

where y is the response data, t is the time vector, and ysd is the standard
deviations of the response.

How to Identify Delay Using Transient-Response Plots

You can use transient-response plots to estimate the input delay, or dead
time, of linear systems. Input delay represents the time it takes for the output
to respond to the input.

In the System Identification Tool GUI. To view the transient response

plot, select the Transient resp check box in the System Identification Tool
GUI. For example, the following step response plot shows a time delay of
about 0.25 s before the system responds to the input.

Step Response Plot

At the command line. You can use the impulse command to plot the
impulse response. The time delay is equal to the first positive peak in the

3-16
 Identifying Impulse-Response Models

transient response magnitude that is greater than the confidence region for
positive time values.

For example, the following commands create an impulse-response plot with a

1-standard-deviation confidence region:

% Load sample data

load dry2
% Split data into estimation and
% validation data sets
ze = dry2(1:500);
zr = dry2(501:1000);
impulse(ze,'sd',1,'fill')

The resulting figure shows that the first positive peak of the response
magnitude, which is greater than the confidence region for positive time
values, occurs at 0.24 s.

3-17
 3 Linear Model Identification

Algorithm for Correlation Analysis

To better understand the algorithm underlying correlation analysis, consider
the following description of a dynamic system:

y(t) = G(q)u(t) + v(t)

where u(t) and y(t) are the input and output signals, respectively. v(t) is the
additive noise term. G(q) is the transfer function of the system. The G(q)u(t)
notation represents the following operation:

∞
G(q)u(t) = ∑ g(k)u(t − k)
k=1

3-18
 Identifying Impulse-Response Models

q is the shift operator, defined by the following equation:

∞
G(q) = ∑ g(k)q−k q−1u(t) = u(t − 1)
k=1

For impulse response, the algorithm estimates impulse response coefficients

g for both the single- and multiple-output data. The impulse response is
estimated as a high-order, noncausal FIR model:

y(t) = g(−m)u(t + m) + … + g(−1)u(t + 1) + g(0)u(t)

+ g(1)u(t − 1) + … + g(n)u(t − n)

The estimation algorithm prefilters the data such that the input is as white
as possible. It then computes the correlations from the prefiltered data to
obtain the FIR coefficients.

g is also estimated for negative lags, which takes into account any noncausal
effects from input to output. Noncausal effects can result from feedback. The
coefficients are computed using the least-squares method.

For a multiple-input or multiple-output system, the impulse response gk is an

ny-by-nu matrix, where ny is the number of outputs and nu is the number
of inputs. The i-jth element of the impulse response matrix describes the
behavior of the ith output after an impulse in the jth input.

3-19
 3 Linear Model Identification

Identifying Low-Order Transfer Functions (Process Models)

In this section...
“What Is a Process Model?” on page 3-20
“Data Supported by a Process Model” on page 3-21
“How to Estimate Process Models Using the GUI” on page 3-21
“How to Estimate Process Models at the Command Line” on page 3-27
“Options for Specifying the Process-Model Structure” on page 3-33
“Options for Multiple-Input Models” on page 3-34
“Options for the Disturbance Model Structure” on page 3-35
“Options for Frequency-Weighing Focus” on page 3-36
“Options for Initial States” on page 3-37

What Is a Process Model?

The structure of a continuous-time process model is a simple transfer function
that describes linear system dynamics in terms of one or more of the following
elements:

• Static gain Kp.

• One or more time constants Tpk. For complex poles, the time constant is
called Tω —equal to the inverse of the natural frequency—and the damping
coefficient is ζ (zeta).
• Process zero Tz.
• Possible time delay Td before the system output responds to the input
(dead time).
• Possible enforced integration.

Process models are popular for describing system dynamics in many

industries and apply to various production environments. The primary
advantages of these models are that they provide delay estimation, and the
model coefficients have a physical interpretation.

3-20
 Identifying Low-Order Transfer Functions (Process Models)

You can create different model structures by varying the number of poles,
adding an integrator, or adding or removing a time delay or a zero. You can
specify a first-, second-, or third-order model, and the poles can be real or
complex (underdamped modes).

Note Continuous-time process models let you estimate the input delay.

For example, the following model structure is a first-order continuous-time

process model, where K is the static gain, Tp1 is a time constant, and Td is the
input-to-output delay:

K
G(s) = e− sTd
1 + sTp1

To learn more about estimating continuous-time process models in the GUI,

see “Tutorial – Identifying Low-Order Transfer Functions (Process Models)
Using the GUI” in System Identification Toolbox Getting Started Guide.

Data Supported by a Process Model

You can estimate low-order (up to third order), continuous-time transfer
functions from data with the following characteristics:

• Regularly sampled time- or frequency-domain iddata or idfrd data object

• Real data, or complex data in the time domain only
• Single-output data

You must import your data into the MATLAB workspace, as described in
Chapter 2, “Data Import and Processing”.

How to Estimate Process Models Using the GUI

Before you can perform this task, you must have

3-21
 3 Linear Model Identification

• Imported data into the System Identification Tool GUI. See “Importing
Time-Domain Data into the GUI” on page 2-16. For supported data
formats, see “Data Supported by a Process Model” on page 3-21.
• Performed any required data preprocessing operations. If you need to
model nonzero offsets, such as when model contains integration behavior,
do not detrend your data. In other cases, to improve the accuracy of your
model, you should detrend your data. See “Ways to Process Data for
System Identification” on page 2-4.

1 In the System Identification Tool GUI, select Estimate > Process models
to open the Process Models dialog box.

2 If your model contains multiple inputs, select the input channel in the
Input list. This list only appears when you have multiple inputs. For more
information, see “Options for Multiple-Input Models” on page 3-34.

3-22
 Identifying Low-Order Transfer Functions (Process Models)

3 In the Model Transfer Function area, specify the model structure using
the following options:
• Under Poles, select the number of poles, and then select All real or
Underdamped.

Note You need at least two poles to allow underdamped modes

(complex-conjugate pair).

• Select the Zero check box to include a zero, which is a numerator term
other than a constant, or clear the check box to exclude the zero.
• Select the Delay check box to include a delay, or clear the check box
to exclude the delay.
• Select the Integrator check box to include an integrator (self-regulating
process), or clear the check box to exclude the integrator.

The Parameter area shows as many active parameters as you included in

the model structure.

Note By default, the model Name is set to the acronym that reflects the
model structure, as described in “Options for Specifying the Process-Model
Structure” on page 3-33.

4 In the Initial Guess area, select Auto-selected to calculate the initial

parameter values for the estimation. The Initial Guess column in the

3-23
 3 Linear Model Identification

Parameter table displays Auto. If you do not have a good guess for the
parameter values, Auto works better than entering an ad hoc value.

5 (Optional) If you approximately know a parameter value, enter this value

in the Initial Guess column of the Parameter table. The estimation
algorithm uses this value as a starting point. If you know a parameter
value exactly, enter this value in the Initial Guess column, and also select
the corresponding Known check box in the table to fix its value.

If you know the range of possible values for a parameter, enter these values
into the corresponding Bounds field to help the estimation algorithm.

For example, the following figure shows that the delay value Td is fixed
at 2 s and is not estimated.

3-24
 Identifying Low-Order Transfer Functions (Process Models)

6 In the Disturbance Model list, select one of the available options. For
more information about each option, see “Options for the Disturbance
Model Structure” on page 3-35.

7 In the Focus list, select how to weigh the relative importance of the fit at
different frequencies. For more information about each option, see “Options
for Frequency-Weighing Focus” on page 3-36.

8 In the Initial state list, specify how you want the algorithm to treat initial
states. For more information about the available options, see “Options for
Initial States” on page 3-37.

Tip If you get a bad fit, you might try setting a specific method for handling
initial states, rather than choosing it automatically.

9 In the Covariance list, select Estimate if you want the algorithm to

compute parameter uncertainties. Effects of such uncertainties are
displayed on plots as model confidence regions.

3-25
 3 Linear Model Identification

To omit estimating uncertainty, select None. Skipping uncertainty

computation might reduce computation time for complex models and large
data sets.

10 In the Model Name field, edit the name of the model or keep the default.
The name of the model should be unique in the Model Board.

11 To view the estimation progress in the MATLAB Command Window, select

the Trace check box. During estimation, the following information is
displayed for each iteration:
• Loss function — Equals the determinant of the estimated covariance
matrix of the input noise.
• Parameter values — Values of the model structure coefficients you
specified.
• Search direction — Change in parameter values from the previous
iteration.
• Fit improvements — Shows the actual versus expected improvements in
the fit.

12 Click Estimate to add this model to the Model Board in the System
Identification Tool GUI.

13 To stop the search and save the results after the current iteration has been
completed, click Stop Iterations. To continue iterations from the current
model, click the Continue iter button to assign current parameter values
as initial guesses for the next search.

Next Steps

• Validate the model by selecting the appropriate check box in the Model
Views area of the System Identification Tool GUI. For more information
about validating models, see “Overview of Model Validation and Plots”
on page 8-2.
• Refine the model by clicking the Value —> Initial Guess button to assign
current parameter values as initial guesses for the next search, edit the
Model Name field, and click Estimate.

3-26
 Identifying Low-Order Transfer Functions (Process Models)

• Export the model to the MATLAB workspace for further analysis by

dragging it to the To Workspace rectangle in the System Identification
Tool GUI.

How to Estimate Process Models at the Command

Line
• “Prerequisites” on page 3-27
• “Using pem to Estimate Process Models” on page 3-27
• “Example – Estimating Process Models with Free Parameters at the
Command Line” on page 3-29
• “Example – Estimating Process Models with Fixed Parameters at the
Command Line” on page 3-30

Prerequisites
Before you can perform this task, you must have

• Regularly sampled data as an iddata object. See “Representing Time- and

Frequency-Domain Data Using iddata Objects” on page 2-49. For supported
data formats, see “Data Supported by a Process Model” on page 3-21.
• Performed any required data preprocessing operations. If you need to
model nonzero offsets, such as when model contains integration behavior,
do not detrend your data. In other cases, to improve the accuracy of your
model, you should detrend your data. See “Ways to Process Data for
System Identification” on page 2-4.

Using pem to Estimate Process Models

You can estimate process models using the iterative estimation method pem
that minimizes the prediction errors to obtain maximum likelihood estimates.
The resulting models are stored as idproc model objects.

You can use the following general syntax to both configure and estimate
process models:

m = pem(data,mod_struc,'Property1',Value1,...,
'PropertyN',ValueN)

3-27
 3 Linear Model Identification

To capture offsets that are essential to describe the dynamics of interest,

such as when the model contains integration behavior, set the InputLevel
property set to “estimate”.

data is the estimation data and mod_struc is a string that represents

the process model structure, as described in “Options for Specifying the
Process-Model Structure” on page 3-33.

Tip You do not need to construct the model object using idproc before
estimation unless you want to specify initial parameter guesses or fixed
parameter values, as described in “Example – Estimating Process Models
with Fixed Parameters at the Command Line” on page 3-30.

The property-value pairs specify any model properties that configure the
estimation algorithm and the initial conditions. For more information about
accessing and setting model properties, see “Model Properties” on page 1-17.

Note You can specify all property-value pairs in pem as a simple,

comma-separated list without worrying about the hierarchy of these
properties in the idproc model object.

For more information about validating a process model, see “Overview of

Model Validation and Plots” on page 8-2.

You can use pem to refine parameter estimates of an existing process model,
as described in “Refining Linear Parametric Models” on page 3-104.

For detailed information about pem and idproc, see the corresponding
reference page.

3-28
 Identifying Low-Order Transfer Functions (Process Models)

Example – Estimating Process Models with Free Parameters

at the Command Line
This example demonstrates how to estimate the parameters of a first-order
process model:

K
G(s) = e− sTd
1 + sTp1

This process has two inputs and the response from each input is estimated by
a first-order process model. All parameters are free to vary.

Use the following commands to estimate a model m from sample data:

% Load sample data

load co2data
% Sampling interval is 0.5 min (known)
Ts = 0.5;
% Split data set into estimation data ze
% and validation data zv
ze = iddata(Output_exp1,Input_exp1,Ts,...
'TimeUnit','min');
zv = iddata(Output_exp2,Input_exp2,Ts,...
'TimeUnit','min');
% Estimate model with one pole and a delay
m = pem(ze,'P1D')

3-29
 3 Linear Model Identification

MATLAB computes the following output:

Process model with 2 inputs:

y = G_1(s)u_1 + G_2(s)u_2
where
K
G_1(s) = ---------- * exp(-Td*s)
1+Tp1*s

with K = -3.2168
Tp1 = 23.033
Td = 10.101

K
G_2(s) = ---------- * exp(-Td*s)
1+Tp1*s

with K = 9.9877
Tp1 = 2.0314
Td = 4.8368

Use dot notation to get the value of any model parameter. For example, to get
the Value field in the K structure, type the following command:

m.K.value

Example – Estimating Process Models with Fixed Parameters

at the Command Line
When you know the values of certain parameters in the model and want
to estimate only the values you do not know, you must specify the fixed
parameters after creating the idproc model object.

3-30
 Identifying Low-Order Transfer Functions (Process Models)

Use the following commands to prepare the data and construct a process
model with one pole and a delay:

% Load sample data

load co2data
% Sampling interval is 0.5 min (known)
Ts = 0.5;
% Split data set into estimation data ze
% and validation data zv
ze = iddata(Output_exp1,Input_exp1,Ts,...
'TimeUnit','min');
zv = iddata(Output_exp2,Input_exp2,Ts,...
'TimeUnit','min');
mod=idproc('P1D')

MATLAB computes the following output:

Process model with transfer function

K
G(s) = ---------- * exp(-Td*s)
1+Tp1*s

with K = NaN
Tp1 = NaN
Td = NaN

This model was not estimated from data.

The model parameters K, Tp1, and Td are assigned NaN values, which means
that the parameters have not yet been estimated from the data.

All process-model parameters are structures with the following fields:

• status field specifies whether to estimate the parameter, or keep the

initial value fixed (do not estimate), or set the value to zero. This field can
have the values 'estimate', 'fixed', or 'zero'. For more information,
see “Options for Initial States” on page 3-37.
• min specifies the minimum bound on the parameter.
• max specifies the maximum bound on the parameter.

3-31
 3 Linear Model Identification

• value specifies the numerical value of the parameter, if known.

To set the value of K to 12 and keep it fixed, use the following commands:

mod.K.value=12;
mod.K.status='fixed';

Note mod is defined for one input. This model is automatically adjusted
to have a duplicate for each input.

To estimate Tp1 and Td only, use the following command:

mod_proc=pem(ze,mod)

MATLAB computes the following result:

Process model with 2 inputs:

y = G_1(s)u_1 + G_2(s)u_2
where
K
G_1(s) = ---------- * exp(-Td*s)
1+Tp1*s

with K = 12
Tp1 = 7.0998e+007
Td = 15

K
G_2(s) = ---------- * exp(-Td*s)
1+Tp1*s

with K = 12
Tp1 = 3.6962
Td = 3.817

In this case, the value of K is fixed at 12, but Tp1 and Td are estimated.

3-32
 Identifying Low-Order Transfer Functions (Process Models)

If you prefer to specify parameter constraints directly in the estimator syntax,

the following table provides examples of pem commands.

Action Example
Fix the value of K to 12.
m=pem(ze,'p1d','k','fix','k',12)

Initialize K for the iterative search

without fixing this value. m=pem(ze,'p1d','k',12)

Constrain the value of K between

3 and 4. m=pem(ze,'p1d','k',...
{'min',3},'k',{'max',4})

Options for Specifying the Process-Model Structure

This section describes how to specify the model structure in the estimation
procedures “How to Estimate Process Models Using the GUI” on page 3-21
and “How to Estimate Process Models at the Command Line” on page 3-27.

In the System Identification Tool GUI. Specify the model structure by

selecting the number of real or complex poles, and whether to include a zero,
delay, and integrator. The resulting transfer function is displayed in the
Process Models dialog box.

At the command line. Specify the model structure using an acronym that
includes the following letters and numbers:

• (Required) P for a process model

• (Required) 0, 1, 2 or 3 for the number of poles

• (Optional) D to include a time-delay term e− sTd

• (Optional) Z to include a process zero (numerator term)
• (Optional) U to indicate possible complex-valued (underdamped) poles
• (Optional) I to indicate enforced integration

3-33
 3 Linear Model Identification

Typically, you specify the model-structure acronym as a string argument in

the estimation command pem:

• pem(data,'P1D') to estimate the following structure:

K
G(s) = e− sTd
1 + sTp1

• pem(data,'P2ZU') to estimate the following structure:

K p (1 + sTz )
G(s) =
1 + 2sζTw + s2 Tw2

• pem(data,'P0ID') to estimate the following structure:

Kp
G(s) = e− sTd
s

• pem(data,'P3Z') to estimate the following structure:

K p (1 + sTz )
G(s) =
(1 + sTp1 ) (1 + sTp2 ) (1 + sTp3 )
For more information about estimating models , see “How to Estimate Process
Models at the Command Line” on page 3-27.

Options for Multiple-Input Models

If your model contains multiple inputs, you can specify whether to estimate
the same transfer function for all inputs, or a different transfer function
for each input. The information in this section supports the estimation
procedures “How to Estimate Process Models Using the GUI” on page 3-21
and “How to Estimate Process Models at the Command Line” on page 3-27.

In the System Identification Tool GUI. To fit a data set with multiple
inputs in the Process Models dialog box, configure the process model settings
for one input at a time. When you finish configuring the model and the

3-34
 Identifying Low-Order Transfer Functions (Process Models)

estimation settings for one input, select a different input in the Input
Number list.

If you want the same transfer function to apply to all inputs, select the Same
structure for all channels check box. To apply a different structure to each
channel, leave this check box clear, and create a different transfer function for
each input.

At the command line. Specify the model structure as a cell array of acronym
strings in the estimation command pem. For example, use this command to
specify the first-order transfer function for the first input, and a second-order
model with a zero and an integrator for the second input:

m = idproc({'P1','P2ZI'})
m = pem(data,m)

To apply the same structure to all inputs, define a single structure in idproc.

Options for the Disturbance Model Structure

This section describes how to specify a noise model in the estimation
procedures “How to Estimate Process Models Using the GUI” on page 3-21
and “How to Estimate Process Models at the Command Line” on page 3-27.

In addition to the transfer function G, a linear system can include an additive

noise term He, as follows:

y = Gu + He

where e is white noise.

You can estimate only the dynamic model G, or estimate both the dynamic
model and the disturbance model H. For process models, H is a rational
transfer function C/D, where the C and D polynomials for a first- or
second-order ARMA model.

In the GUI. To specify whether to include or exclude a noise model in the

Process Models dialog box, select one of the following options from the
Disturbance Model list:

3-35
 3 Linear Model Identification

• None — The algorithm does not estimate a noise model (C=D=1). This
option also sets Focus to Simulation.
• Order 1 — Estimates a noise model as a continuous-time, first-order
ARMA model.
• Order 2 — Estimates a noise model as a continuous-time, second-order
ARMA model.

At the command line. Specify the disturbance model as an argument in

the estimation command pem. For example, use this command to estimate a
first-order transfer function and a first-order noise model:

pem(data,'P1D','DisturbanceModel','ARMA1')

Tip You can type 'dis' instead of 'DisturbanceModel'.

For a complete list of values for the DisturbanceModel model property, see
the idproc reference page.

Options for Frequency-Weighing Focus

You can specify how the estimation algorithm weighs the fit at various
frequencies. This information supports the estimation procedures “How to
Estimate Process Models Using the GUI” on page 3-21 and “How to Estimate
Process Models at the Command Line” on page 3-27.

In the System Identification Tool GUI. Set Focus to one of the following
options:

• Prediction — Uses the inverse of the noise model H to weigh the relative
importance of how closely to fit the data in various frequency ranges.
Corresponds to minimizing one-step-ahead prediction, which typically
favors the fit over a short time interval. Optimized for output prediction
applications.
• Simulation — Uses the input spectrum to weigh the relative importance of
the fit in a specific frequency range. Does not use the noise model to weigh
the relative importance of how closely to fit the data in various frequency
ranges. Optimized for output simulation applications.

3-36
 Identifying Low-Order Transfer Functions (Process Models)

• Stability — Behaves the same way as the Prediction option, but also
forces the model to be stable. For more information about model stability,
see “Unstable Models” on page 8-66.
• Filter — Specify a custom filter to open the Estimation Focus dialog box,
where you can enter a filter, as described in “Simple Passband Filter” on
page 2-113 or “Defining a Custom Filter” on page 2-114. This prefiltering
applies only for estimating the dynamics from input to output. The
disturbance model is determined from the estimation data.

At the command line. Specify the focus as an argument in the estimation

command pem using the same options as in the GUI. For example, use this
command to optimize the fit for simulation and estimate a disturbance model:

pem(data,'P1D','dist','arma2','Focus','Simulation')

Options for Initial States

Because the process models are dynamic, you need initial states that capture
past input properties. Thus, you must specify how the iterative algorithm
treats initial states. This information supports the estimation procedures
“How to Estimate Process Models Using the GUI” on page 3-21 and “How to
Estimate Process Models at the Command Line” on page 3-27.

In the System Identification Tool GUI. Set Initial state to one of the
following options:

• Zero — Sets all initial states to zero.

• Estimate — Treats the initial states as an unknown vector of parameters
and estimates these states from the data.
• Backcast — Estimates initial states using a backward filtering method
(least-squares fit).
• U-level est — Estimates both the initial states and the InputLevel
model property that represents the input offset level. For multiple inputs,
the input level for each input is estimated individually. Use if you included
an integrator in the transfer function.
• Auto — Automatically chooses one of the preceding options based on the
estimation data. If initial states have negligible effect on the prediction
errors, the initial states are set to zero to optimize algorithm performance.

3-37
 3 Linear Model Identification

At the command line. Specify the initial states as an argument in the

estimation command pem using the same options as in the GUI. For example,
use this command to estimate a first-order transfer function and set the
initial states to zero:

m=pem(data,'P1D','InitialState','zero')

For a complete list of values for the InitialState model property, see the
idproc reference page.

3-38
 Identifying Input-Output Polynomial Models

Identifying Input-Output Polynomial Models

In this section...
“What Are Black-Box Polynomial Models?” on page 3-39
“Data Supported by Polynomial Models” on page 3-46
“Preliminary Step – Estimating Model Orders and Input Delays” on page
3-48
“How to Estimate Polynomial Models in the GUI” on page 3-56
“How to Estimate Polynomial Models at the Command Line” on page 3-59
“Options for Multiple-Input and Multiple-Output ARX Orders” on page 3-64
“Option for Frequency-Weighing Focus” on page 3-65
“Options for Initial States” on page 3-66
“Algorithms for Estimating Polynomial Models” on page 3-66
“Example – Estimating Models Using armax” on page 3-67

What Are Black-Box Polynomial Models?

• “Polynomial Model Structure” on page 3-40
• “Understanding the Time-Shift Operator q” on page 3-41
• “Definition of a Discrete-Time Polynomial Model” on page 3-41
• “Definition of a Continuous-Time Polynomial Model” on page 3-44
• “Definition of Multiple-Output ARX Models” on page 3-44

3-39
 3 Linear Model Identification

Polynomial Model Structure

You can estimate the following types of linear polynomial model structures:

nu
Bi (q) C(q)
A(q) y(t) = ∑ ui ( t − nki ) + e(t)
F
i=1 i
( q) D(q)

The polynomials A, Bi, C, D, and F i contain the time-shift operator q. ui is the

ith input, nu is the total number of inputs, and nki is the ith input delay that
characterizes the delay response time. The variance of the white noise e(t)
is assumed to be λ . For more information about the time-shift operator, see
“Understanding the Time-Shift Operator q” on page 3-41.

Note This form is completely equivalent to the Z-transform form: q

corresponds to z.

To estimate polynomial models, you must specify the model order as a set of
integers that represent the number of coefficients for each polynomial you
include in your selected structure—na for A, nb for B, nc for C, nd for D,
and nf for F. You must also specify the number of samples nk corresponding
to the input delay—dead time—given by the number of samples before the
output responds to the input.

The number of coefficients in denominator polynomials is equal to the number

of poles, and the number of coefficients in the numerator polynomials is equal
to the number of zeros plus 1. When the dynamics from u(t) to y(t) contain a
delay of nk samples, then the first nk coefficients of B are zero.

For more information about the family of transfer-function models, see the
corresponding section in System Identification: Theory for the User, Second
Edition, by Lennart Ljung, Prentice Hall PTR, 1999.

3-40
 Identifying Input-Output Polynomial Models

Understanding the Time-Shift Operator q

The general polynomial equation is written in terms of the time-shift
operator q–1. To understand this time-shift operator, consider the following
discrete-time difference equation:

y(t) + a1 y(t − T ) + a2 y(t − 2T ) =

b1u(t − T ) + b2u(t − 2T )

where y(t) is the output, u(t) is the input, and T is the sampling interval. q-1
is a time-shift operator that compactly represents such difference equations

using q−1 u(t) = u(t − T ) :

y(t) + a1 q−1 y(t) + a2 q−2 y(t) =

b1 q−1u(t) + b2 q−2u(t)
or
A(q) y(t) = B(q)u(t)

In this case, A(q) = 1 + a1 q−1 + a2 q−2 and B(q) = b1 q−1 + b2 q−2 .

Note This q description is completely equivalent to the Z-transform form: q

corresponds to z.

Definition of a Discrete-Time Polynomial Model

These model structures are subsets of the following general polynomial
equation:

nu
Bi (q) C(q)
A(q) y(t) = ∑ ui ( t − nki ) + e(t)
F (q)
i=1 i
D(q)

The model structures differ by how many of these polynomials are included
in the structure. Thus, different model structures provide varying levels of
flexibility for modeling the dynamics and noise characteristics. For more

3-41
 3 Linear Model Identification

information about the time-shift operator, see “Understanding the Time-Shift

Operator q” on page 3-41.

The following table summarizes common linear polynomial model structures

supported by the System Identification Toolbox product. If you have a specific
structure in mind for your application, you can decide whether the dynamics
and the noise have common or different poles. A(q) corresponds to poles that
are common for the dynamic model and the noise model. Using common poles
for dynamics and noise is useful when the disturbances enter the system at
the input. F i determines the poles unique to the system dynamics, and D
determines the poles unique to the disturbances.

Model Discrete-Time Form Noise Model

Structure
ARX
nu The noise model is A 1 and the

A(q) y(t) = ∑ Bi (q)ui ( t − nki ) + e(t) noise is coupled to the dynamics

i=1 model. ARX does not let you
model noise and dynamics
independently. Estimate an
ARX model to obtain a simple
model at good signal-to-noise
ratios.
ARMAX Extends the ARX structure by
nu providing more flexibility for
A(q) y(t) = ∑ Bi (q)ui ( t − nki ) + C(q) e(t) modeling noise using the C
i=1 parameters (a moving average of
white noise). Use ARMAX when
the dominating disturbances
enter at the input. Such
disturbances are called load
disturbances.

3-42
 Identifying Input-Output Polynomial Models

Model Discrete-Time Form Noise Model

Structure
Box-Jenkins Provides completely independent
(BJ) nu parameterization for the
Bi (q) C(q)
y(t) = ∑ ui ( t − nki ) + e(t) dynamics and the noise using
F (q)
i=1 i
D(q) rational polynomial functions.
Use BJ models when the noise
does not enter at the input,
but is primary a measurement
disturbance, This structure
provides additional flexibility for
modeling noise.
Output-Error Use when you want to
(OE) nu parameterize dynamics, but
Bi (q)
y(t) = ∑ ui ( t − nki ) + e(t) do not want to estimate a noise
F (q)
i=1 i model.

Note In this case, the noise

models is H = 1 in the general
equation and the white noise
source e(t) affects only the
output.

The input-output polynomial models for single output systems are represented
by the idpoly object. Multi-output polynomial ARX models are represented
by the idarx object.

The System Identification Tool GUI supports only the polynomial models
listed in the table. However, you can use pem to estimate all five polynomial
or any subset of polynomials in the general equation. For more information
about working with pem, see “Using pem to Estimate Polynomial Models”
on page 3-61.

3-43
 3 Linear Model Identification

Definition of a Continuous-Time Polynomial Model

In continuous time, the general frequency-domain equation is written in terms
of the Laplace transform variable s, which corresponds to a differentiation
operation:

B(s) C(s)
A(s)Y (s) = U (s) + E(s)
F (s) D(s)

In the continuous-time case, the underlying time-domain model is a

differential equation and the model order integers represent the number of
estimated numerator and denominator coefficients. For example, na=3 and
nb=2 correspond to the following model:

A(s) = s4 + a1 s3 + a2 s2 + a3
B(s) = b1 s + b2

The simplest way to estimate continuous-time polynomial models of

arbitrary structure is to first estimate a discrete-time model of arbitrary
order and then use d2c to convert this model to continuous time. For more
information, see “Transforming Between Discrete-Time and Continuous-Time
Representations” on page 3-113.

You can also estimate continuous-time polynomial models directly using

continuous-time frequency-domain data. In this case, you must set the Ts data
property to 0 to indicate that you have continuous-time frequency-domain
data, and use the oe command to estimate an Output-Error polynomial model.

Definition of Multiple-Output ARX Models

You can use a multiple-output ARX model to model a multiple-output dynamic
system. The ARX model structure is represented by the idarx object, and
given by the following equation:

A(q) y(t) = B(q)u ( t − nk ) + e(t)

3-44
 Identifying Input-Output Polynomial Models

For a system with nu inputs and ny outputs, A(q) is an ny-by-ny matrix. A(q)
can be represented as a polynomial in the shift operator q-1:

A(q) = Iny + A1 q−1 + … + Ana q− na

For more information about the time-shift operator, see “Understanding the
Time-Shift Operator q” on page 3-41.

A(q) can also be represented as a matrix:

⎛ a11 (q) a12 (q) … a1ny (q) ⎞

⎜ ⎟
a21 (q) a22 (q) … a2ny (q) ⎟
A(q) = ⎜
⎜ … … … … ⎟
⎜⎜ ⎟
a
⎝ ny1 ( q) any2 ( q) … anyny( q) ⎟⎠

where the matrix element akj is a polynomial in the shift operator q-1:

− nakj
akj (q) = δkj + a1kj q−1 + … + akj kj q
na

δkj represents the Kronecker delta, which equals 1 for k=j and equals 0
for k≠j. This polynomial describes how the old values of the jth output are
affected by the kth output. The ith row of A(q) represents the contribution of
the past output values for predict the current value of the ith output.

B(q) is an ny-by-ny matrix. B(q) can be represented as a polynomial in the

shift operator q-1:

B(q) = B0 + B1 q−1 + … + Bnb q− nb

3-45
 3 Linear Model Identification

B(q) can also be represented as a matrix:

⎛ b11 (q) b12 (q) … b1nu (q) ⎞

⎜ ⎟
b21 (q) b22 (q) … b2nu (q) ⎟
B(q) = ⎜
⎜ … … … … ⎟
⎜⎜ ⎟
⎝ bny1 (q) bny2 (q) … bnynu( q) ⎟⎠

where the matrix element bkj is a polynomial in the shift operator q-1:

− nbkj nk − nbkj − nbkj +1

bkj (q) = a1kj q + … + akj kj q

nkkj is the delay from the jth input to the kth output. B(q) represents the
contributions of inputs to predicting all output values.

Data Supported by Polynomial Models

• “Types of Supported Data” on page 3-46
• “Designating Data for Estimating Continuous-Time Models” on page 3-47
• “Designating Data for Estimating Discrete-Time Models” on page 3-47

Types of Supported Data

You can estimate linear, black-box polynomial models from data with the
following characteristics:

• Time- or frequency-domain data (iddata or idfrd data objects).

Note For frequency-domain data, you can only estimate ARX and OE
models.

To estimate black-box polynomial models for time-series data, see Chapter

6, “Time Series Identification”.
• Real data or complex data in any domain.

3-46
 Identifying Input-Output Polynomial Models

• Single-output and multiple-output (ARX structure only).

You must import your data into the MATLAB workspace, as described in
Chapter 2, “Data Import and Processing”.

Designating Data for Estimating Continuous-Time Models

To get a linear, continuous-time model of arbitrary structure for time-domain
data, you can estimate a discrete-time model, and then use d2c to transform
it to a continuous-time model.

For continuous-time frequency-domain data, you can estimate directly only

the ARX and Output-Error (OE) continuous-time models. Other structures
include noise models, which is not supported for frequency-domain data.

Tip To denote continuous-time frequency-domain data, set the data sampling

interval to 0. You can set the sampling interval when you import data into
the GUI or set the Ts property of the data object at the command line.

Designating Data for Estimating Discrete-Time Models

You can estimate arbitrary-order, linear state-space models for both time- or
frequency-domain data.

Set the data property Ts to:

• 0, for frequency response data that is measured directly from an

experiment.
• Equal to the Ts of the original data, for frequency response data obtained
by transforming time-domain iddata (using spa and etfe).

Tip You can set the sampling interval when you import data into the GUI or
set the Ts property of the data object at the command line.

3-47
 3 Linear Model Identification

Preliminary Step – Estimating Model Orders and

Input Delays
• “Why Estimate Model Orders and Delays?” on page 3-48
• “Estimating Orders and Delays in the GUI” on page 3-48
• “Estimating Model Orders at the Command Line” on page 3-52
• “Estimating Delays at the Command Line” on page 3-53
• “Selecting Model Orders from the Best ARX Structure” on page 3-54

Why Estimate Model Orders and Delays?

To estimate polynomial models, you must provide input delays and model
orders. If you already have insight into the physics of your system, you can
specify the number of poles and zeros.

In most cases, you do not know the model orders in advance. To get initial
model orders and delays for your system, you can estimate several ARX
models with a range of orders and delays and compare the performance of
these models. You choose the model orders that correspond to the best model
performance and use these orders as an initial guess for further modeling.

Because this estimation procedure uses the ARX model structure, which
includes the A and B polynomials, you only get estimates for the na, nb, and
nk parameters. However, you can use these results as initial guesses for the
corresponding polynomial orders and input delays in other model structures,
such as ARMAX, OE, and BJ.

If the estimated nk is too small, the leading nb coefficients are much smaller
than their standard deviations. Conversely, if the estimated nk is too large,
there is a significant correlation between the residuals and the input for lags
that correspond to the missing B terms. For information about residual
analysis plots, see “Residual Analysis” on page 8-21.

Estimating Orders and Delays in the GUI

The following procedure assumes that you have already imported your data
into the GUI and performed any necessary preprocessing operations. For
more information, see Chapter 2, “Data Import and Processing”.

3-48
 Identifying Input-Output Polynomial Models

To estimate model orders and input delays in the System Identification Tool
GUI:

1 In the System Identification Tool GUI, select Estimate > Linear

parametric models to open the Linear Parametric Models dialog box.

The ARX model is already selected by default in the Structure list.

Note For time-series models, select the AR model structure.

2 Edit the Orders field to specify a range of poles, zeros, and delays. For
example, enter the following values for na, nb, and nk:

[1:10 1:10 1:10]

Tip As a shortcut for entering 1:10 for each required model order, click
Order Selection.

3-49
 3 Linear Model Identification

3-50
 Identifying Input-Output Polynomial Models

3 Click Estimate to open the ARX Model Structure Selection window, which
displays the model performance for each combination of model parameters.
The following figure shows an example plot.

4 Select a rectangle that represents the optimum parameter combination and

click Insert to estimates a model with these parameters. For information
about using this plot, see “Selecting Model Orders from the Best ARX
Structure” on page 3-54.

This action adds a new model to the Model Board in the System
Identification Tool GUI. The default name of the parametric model contains
the model type and the number of poles, zeros, and delays. For example,
arx692 is an ARX model with na=6, nb=9, and a delay of two samples.

5 Click Close to close the ARX Model Structure Selection window.

After estimating model orders and delays, use these values as initial guesses
for estimating other model structures, as described in “How to Estimate
Polynomial Models in the GUI” on page 3-56.

3-51
 3 Linear Model Identification

Estimating Model Orders at the Command Line

You can estimate model orders using the struc, arxstruc, and selstruc
commands in combination.

If you are working with a multiple-output system, you must use struc,
arxstruc, and selstruc commands for each output. In this case, you must
subreference the correct output channel in your estimation and validation
data sets.

For each estimation, you use two independent data sets—an estimation
data set and a validation data set. These independent data set can be from
different experiments, or data subsets from a single experiment. For more
information about subreferencing data, see “Subreferencing iddata Objects”
on page 2-57 and “Subreferencing idfrd Objects” on page 2-72.

For an example of estimating model orders for a multiple-input system, see

“Estimating Delays in the Multiple-Input System” in System Identification
Toolbox Getting Started Guide.

struc. The struc command creates a matrix of possible model-order

combinations for a specified range of na, nb, and nk values.

For example, the following command defines the range of model orders and
delays na=2:5, nb=1:5, and nk=1:5:

NN = struc(2:5,1:5,1:5))

arxstruc. The arxstruc command takes the output from struc, estimates
an ARX model for each model order, and compares the model output to the
measured output. arxstruc returns the loss for each model, which is the
normalized sum of squared prediction errors.

For example, the following command uses the range of specified orders NN
to compute the loss function for single-input/single-output estimation data
data_e and validation data data_v:

V = arxstruc(data_e,data_v,NN)

Each row in NN corresponds to one set of orders:

[na nb nk]

3-52
 Identifying Input-Output Polynomial Models

selstruc. The selstruc command takes the output from arxstruc and opens
the ARX Model Structure Selection window to guide your choice of the model
order with the best performance.

For example, to open the ARX Model Structure Selection window and
interactively choose the optimum parameter combination, use the following
command:

selstruc(V)

For more information about working with the ARX Model Structure Selection
window, see “Selecting Model Orders from the Best ARX Structure” on page
3-54.

To find the structure that minimizes Akaike’s Information Criterion, use

the following command:

nn = selstruc(V,'AIC')

where nn contains the corresponding na, nb, and nk orders.

Similarly, to find the structure that minimizes the Rissanen’s Minimum

Description Length (MDL), use the following command:

nn = selstruc(V,'MDL')

To select the structure with the smallest loss function, use the following
command:

nn = selstruc(V,0)

After estimating model orders and delays, use these values as initial guesses
for estimating other model structures, as described in “Using pem to Estimate
Polynomial Models” on page 3-61.

Estimating Delays at the Command Line

The delayest command estimates the time delay in a dynamic system by
estimating a low-order, discrete-time ARX model and treating the delay as an
unknown parameter.

3-53
 3 Linear Model Identification

By default, delayest assumes that na=nb=2 and that there is a good

signal-to-noise ratio, and uses this information to estimate nk.

To estimate the delay for a data set data, type the following at the prompt:

delayest(data)

If your data has a single input, MATLAB computes a scalar value for the
input delay—equal to the number of data samples. If your data has multiple
inputs, MATLAB returns a vector, where each value is the delay for the
corresponding input signal.

To compute the actual delay time, you must multiply the input delay by the
sampling interval of the data.

You can also use the ARX Model Structure Selection window to estimate input
delays and model order together, as described in “Estimating Model Orders at
the Command Line” on page 3-52.

Selecting Model Orders from the Best ARX Structure

You generate the ARX Model Structure Selection window for your data to
select the best-fit model.

For a procedure on generating this plot in the System Identification Tool GUI,
see “Estimating Orders and Delays in the GUI” on page 3-48. To open this
plot at the command line, see “Estimating Model Orders at the Command
Line” on page 3-52.

The following figure shows a sample plot in the ARX Model Structure
Selection window.

3-54
 Identifying Input-Output Polynomial Models

The horizontal axis in the ARX Model Structure Selection window is the total
number of ARX parameters:

Number of parameters = na + nb

The vertical axis, called Unexplained output variance (in %), is the ARX
model prediction error for a specific number of parameters. The prediction
error is the sum of the squares of the differences between the validation
data output and the model output. In other words, Unexplained output
variance (in %) is the portion of the output not explained by the model.

Three rectangles are highlighted on the plot—green, blue, and red. Each color
indicates a type of best-fit criterion, as follows:

• Red minimizes the sum of the squares of the difference between the
validation data output and the model output. This option is considered the
overall best fit.
• Green minimizes Rissanen MDL criterion.

3-55
 3 Linear Model Identification

• Blue minimizes Akaike AIC criterion.

In the ARX Model Structure Selection window, click any bar to view the
orders that give the best fit. The area on the right is dynamically updated to
show the orders and delays that give the best fit.

For more information about the AIC criterion, see “Akaike’s Criteria for
Model Validation” on page 8-58.

How to Estimate Polynomial Models in the GUI

Prerequisites
Before you can perform this task, you must have:

• Imported data into the System Identification Tool GUI. See “Importing
Time-Domain Data into the GUI” on page 2-16. For supported data
formats, see “Data Supported by Polynomial Models” on page 3-46.
• Performed any required data preprocessing operations. To improve the
accuracy of your model, you should detrend your data. Removing offsets
and trends is especially important for Output-Error (OE) models and has
less impact on the accuracy of models that include a flexible noise model
structure, such as ARMAX and Box-Jenkins. See “Ways to Process Data for
System Identification” on page 2-4.
• Select a model structure, model orders, and delays. For a list of available
structures, see “What Are Black-Box Polynomial Models?” on page
3-39 For more information about how to estimate model orders and
delays, see “Estimating Orders and Delays in the GUI” on page 3-48.
For multiple-output ARX model, you must specify order matrices in the
MATLAB workspace, as described in “Options for Multiple-Input and
Multiple-Output ARX Orders” on page 3-64.

1 In the System Identification Tool GUI, select Estimate > Linear

parametric models to open the Linear Parametric Models dialog box.

3-56
 Identifying Input-Output Polynomial Models

2 In the Structure list, select the polynomial model structure you want to
estimate from the following options:
• ARX:[na nb nk]
• ARMAX:[na nb nc nk]
• OE:[nb nf nk]
• BJ:[nb nc nd nf nk]

This action updates the options in the Linear Parametric Models dialog box
to correspond with this model structure. For information about each model
structure, see “What Are Black-Box Polynomial Models?” on page 3-39.

Note For time-series data, only AR and ARMA models are available. For
more information about estimating time-series models, see Chapter 6,
“Time Series Identification”.

3 In the Orders field, specify the model orders and delays, as follows:

• For single-output polynomial models. Enter the model orders and

delays according to the sequence displayed in the Structure field. For
multiple-input models, specify nb and nk as row vectors with as many
elements as there are inputs. If you are estimating BJ and OE models,
you must also specify nf as a vector.
For example, for a three-input system, nb can be [1 2 4], where each
element corresponds to an input.
• For multiple-output ARX models. Enter the model orders, as
described in “Options for Multiple-Input and Multiple-Output ARX
Orders” on page 3-64.

Tip To enter model orders and delays using the Order Editor dialog box,
click Order Editor.

4 (ARX models only) Select the estimation Method as ARX or IV

(instrumental variable method). For information about the algorithms, see
“Algorithms for Estimating Polynomial Models” on page 3-66.

3-57
 3 Linear Model Identification

5 In the Name field, edit the name of the model or keep the default.

6 In the Focus list, select how to weigh the relative importance of the fit at
different frequencies. For more information about each option, see “Option
for Frequency-Weighing Focus” on page 3-65.

7 In the Initial state list, specify how you want the algorithm to treat initial
states. For more information about the available options, see “Options for
Initial States” on page 3-37.

Tip If you get an inaccurate fit, try setting a specific method for handling
initial states rather than choosing it automatically.

8 In the Covariance list, select Estimate if you want the algorithm to

compute parameter uncertainties. Effects of such uncertainties are
displayed on plots as model confidence regions.

To omit estimating uncertainty, select None. Skipping uncertainty

computation for large, multiple-output ARX models might reduce
computation time.

9 (ARMAX, OE, and BJ models only) To view the estimation progress in

the MATLAB Command Window, select the Trace check box. During
estimation, the following information is displayed for each iteration:
• Loss function — Equals the determinant of the estimated covariance
matrix of the input noise.
• Parameter values — Values of the model structure coefficients you
specified.
• Search direction — Change in parameter values from the previous
iteration.
• Fit improvements — Shows the actual versus expected improvements in
the fit.

10 Click Estimate to add this model to the Model Board in the System
Identification Tool GUI.

3-58
 Identifying Input-Output Polynomial Models

11 (Prediction-error method only) To stop the search and save the results
after the current iteration has been completed, click Stop Iterations. To
continue iterations from the current model, click the Continue iter button
to assign current parameter values as initial guesses for the next search.

Next Steps

• Validate the model by selecting the appropriate check box in the Model
Views area of the System Identification Tool GUI. For more information
about validating models, see “Overview of Model Validation and Plots”
on page 8-2.
• Export the model to the MATLAB workspace for further analysis by
dragging it to the To Workspace rectangle in the System Identification
Tool GUI.

Tip For ARX and OE models, you can use the exported model for
initializing a nonlinear estimation at the command line. This initialization
may improve the fit of the model. See “Using Linear Model for
Nonlinear ARX Estimation” on page 4-28, and “Using Linear Model for
Hammerstein-Wiener Estimation” on page 4-64.

How to Estimate Polynomial Models at the Command

Line
• “Using arx and iv4 to Estimate ARX Models” on page 3-60
• “Using pem to Estimate Polynomial Models” on page 3-61

Prerequisites

Before you can perform this task, you must have

• Regularly sampled data as an iddata object. See “Representing Time- and

Frequency-Domain Data Using iddata Objects” on page 2-49. For supported
data formats, see “Data Supported by Polynomial Models” on page 3-46.

3-59
 3 Linear Model Identification

• Performed any required data preprocessing operations. To improve the

accuracy of your model, you should detrend your data. Removing offsets
and trends is especially important for Output-Error (OE) models and has
less impact on the accuracy of models that include a flexible noise model
structure, such as ARMAX and Box-Jenkins. See “Ways to Process Data for
System Identification” on page 2-4.
• Select a model structure, model orders, and delays. For a list of available
structures, see “What Are Black-Box Polynomial Models?” on page
3-39 For more information about how to estimate model orders and
delays, see “Estimating Model Orders at the Command Line” on page
3-52 and “Estimating Delays at the Command Line” on page 3-53. For
multiple-output ARX model, you must specify order matrices in the
MATLAB workspace, as described in “Options for Multiple-Input and
Multiple-Output ARX Orders” on page 3-64.

Using arx and iv4 to Estimate ARX Models

You can estimate single-output and multiple-output ARX models using
the arx and iv4 commands. For information about the algorithms, see
“Algorithms for Estimating Polynomial Models” on page 3-66.

If you are estimating a multiple-output ARX model, you must specify order
matrices in the MATLAB workspace before estimation, as described in
“Options for Multiple-Input and Multiple-Output ARX Orders” on page 3-64.

For single-output data, the arx and iv4 commands produce an idpoly model
object, and for multiple-output data these commands produce an idarx model
object.

You can use the following general syntax to both configure and estimate ARX
models:

% Using ARX method

m = arx(data,[na nb nk],'Property1',Value1,...,
'PropertyN',ValueN)
% Using IV method
m = iv4(data,[na nb nk],'Property1',Value1,...,
'PropertyN',ValueN)

3-60
 Identifying Input-Output Polynomial Models

data is the estimation data and [na nb nk] specifies the model orders, as
discussed in “What Are Black-Box Polynomial Models?” on page 3-39.

The property-value pairs specify any model properties that configure the
estimation algorithm and the initial conditions. For more information about
accessing and setting model properties, see “Model Properties” on page 1-17.

Note You can specify all property-value pairs as a comma-separated list.

To get discrete-time models, use the time-domain data (iddata object). To get
a single-output continuous-time model, apply d2c to a discrete-time model
or use continuous-time frequency-domain data—either idfrd object, or
frequency-domain iddata with Ts=0.

Note The System Identification Toolbox product does not support

multiple-output continuous-time idarx models.

For more information about validating you model, see “Overview of Model
Validation and Plots” on page 8-2.

You can use pem to refine parameter estimates of an existing polynomial

model, as described in “Refining Linear Parametric Models” on page 3-104.

For detailed information about these commands, see the corresponding

reference page.

Tip You can use the estimated ARX model for initializing a nonlinear
estimation at the command line, which improves the fit of the model. See
“Using Linear Model for Nonlinear ARX Estimation” on page 4-28.

Using pem to Estimate Polynomial Models

You can estimate any single-output polynomial model using the iterative
prediction-error estimation method pem. For Gaussian disturbances, this

3-61
 3 Linear Model Identification

method gives the maximum likelihood estimate. that minimizes the prediction
errors to obtain maximum-likelihood values. The resulting models are stored
as idpoly model objects.

Use the following general syntax to both configure and estimate polynomial
models:

m = pem(data,'na',na,
'nb',nb,
'nc',nc,
'nd',nb,
'nf',nc,
'nk',nk,
'Property1',Value1,...,
'PropertyN',ValueN)

where data is the estimation data. na, nb, nc, nd, nf are integers that specify
the model orders, and nk specifies the input delays for each input. If you
skip any property-value pair, the corresponding parameter value is set to
zero—except nk, which has the default value 1. For more information about
model orders, see “What Are Black-Box Polynomial Models?” on page 3-39.

Tip You do not need to construct the model object using idoly before
estimation.

If you want to estimate the coefficients of all five polynomials, A, B, C, D, and

F, you must specify an integer order for each polynomial. However, if you
want to specify an ARMAX model for example, which includes only the A, B,
and C polynomials, you must set nd and nf to 0.

Note To get faster estimation of ARX models, use arx or iv4 instead of pem.

In addition to the polynomial models listed in “What Are Black-Box

Polynomial Models?” on page 3-39, you can use pem to model the ARARX
structure—called the generalized least-squares model—by setting nc=nf=0.

3-62
 Identifying Input-Output Polynomial Models

You can also model the ARARMAX structure—called the extended matrix
model—by setting nf=0.

The property-value pairs specify any model properties that configure

the estimation algorithm and the initial conditions. You can enter all
property-value pairs in pem as a comma-separated list without worrying
about the hierarchy of these properties in the idpoly model object. For
more information about accessing and setting model properties, see “Model
Properties” on page 1-17.

For multiple inputs, nb, nf, and nk are row vectors of the same lengths as the
number of input channels:

nb = [nb1 ... nbnu];

nf = [nf1 ... nfnu];
nk = [nk1 ... nknu];

For ARMAX, Box-Jenkins, and Output-Error models—which can only be

estimated using the iterative prediction-error method—use the armax, bj, and
oe estimation commands, respectively. These commands are versions of pem
with simplified syntax for these specific model structures, as follows:

m = armax(Data,[na nb nc nk])
m = oe(Data,[nb nf nk])
m = bj(Data,[nb nc nd nf nk])

Tip If your data is sampled fast, it might help to apply a lowpass filter to the
data before estimating the model, or specify a frequency range for the Focus
property during estimation. For example, to model only data in the frequency
range 0-10 rad/s, use the Focus property, as follows:

m = oe(Data,[nb nf nk],'Focus',[0 10])

For more information about validating your model, see “Overview of Model
Validation and Plots” on page 8-2.

You can use pem to refine parameter estimates of an existing polynomial

model, as described in “Refining Linear Parametric Models” on page 3-104.

3-63
 3 Linear Model Identification

Tip For ARX and OE models, you can use the model for initializing a
nonlinear estimation at the command line, which may improve the fit of the
model. See “Using Linear Model for Nonlinear ARX Estimation” on page 4-28,
and “Using Linear Model for Hammerstein-Wiener Estimation” on page 4-64.

For detailed information about pem and idpoly, see the corresponding
reference page.

Options for Multiple-Input and Multiple-Output ARX

Orders
To estimate a multiple-input and multiple-output (MIMO) ARX model, you
must first specify the model order matrices, as follows:

• NA — An ny-by-ny matrix whose i-jth entry is the order of the polynomial

that relates the jth output to the ith output.
• NB — An ny-by-nu matrix whose i-jth entry is the order of the polynomial
that relates the jth input to the ith output.
• NK — An ny-by-nu matrix whose i-jth entry is the delay from the jth input
to the ith output.
• For ny outputs and nu inputs, the A coefficients are ny-by-ny matrices and
the B coefficients are ny-by-nu matrices. For more information about
MIMO ARX structure, see “Definition of Multiple-Output ARX Models”
on page 3-44.

Note For multiple-output time-series models, only AR models are supported.

AR models require only the NA matrix.

In the System Identification Tool GUI. You can enter the matrices
directly in the Orders field.

At the command line. Define variables that store the model order matrices
and specify these variables in the model-estimation command. You can use
the following syntax to estimate a model with these orders:

3-64
 Identifying Input-Output Polynomial Models

arx(data,'na',NA,'nb',NB,'nk',NK)

Tip To simplify entering large matrices orders in the System Identification

Tool GUI, define the variable NN=[NA NB NK] at the command line. You can
specify this variable in the Orders field.

Option for Frequency-Weighing Focus

You can specify how the estimation algorithm weighs the fit at various
frequencies. This information supports the estimation procedures “How to
Estimate Polynomial Models in the GUI” on page 3-56 and “Using pem to
Estimate Polynomial Models” on page 3-61.

In the System Identification Tool GUI. Set Focus to one of the following
options:

• Prediction — Uses the inverse of the noise model H to weigh the relative
importance of how closely to fit the data in various frequency ranges.
Corresponds to minimizing one-step-ahead prediction, which typically
favors the fit over a short time interval. Optimized for output prediction
applications.
• Simulation — Uses the input spectrum to weigh the relative importance of
the fit in a specific frequency range. Does not use the noise model to weigh
the relative importance of how closely to fit the data in various frequency
ranges. Optimized for output simulation applications.
• Stability — Estimates the best stable model. For more information about
model stability, see “Unstable Models” on page 8-66.
• Filter — Specify a custom filter to open the Estimation Focus dialog box,
where you can enter a filter, as described in “Simple Passband Filter” on
page 2-113 or “Defining a Custom Filter” on page 2-114. This prefiltering
applies only for estimating the dynamics from input to output. The
disturbance model is determined from the unfiltered estimation data.

At the command line. Specify the focus as an argument in the

model-estimation command using the same options as in the GUI. For
example, use this command to estimate an ARX model and emphasize the
frequency content related to the input spectrum only:

3-65
 3 Linear Model Identification

m=arx(data,[2 2 3],'Focus','Simulation')

This Focus setting might produce more accurate simulation results.

Options for Initial States

When you use the iterative estimation algorithm PEM to estimate ARMAX,
Box-Jenkins (BJ), Output-Error (OE), you must specify how the algorithm
treats initial states.

This information supports the estimation procedures “How to Estimate

Polynomial Models in the GUI” on page 3-56 and “Using pem to Estimate
Polynomial Models” on page 3-61.

In the System Identification Tool GUI. For ARMAX, OE, and BJ models,
set Initial state to one of the following options:

• Auto — Automatically chooses Zero, Estimate, or Backcast based on the

estimation data. If initial states have negligible effect on the prediction
errors, the initial states are set to zero to optimize algorithm performance.
• Zero — Sets all initial states to zero.
• Estimate — Treats the initial states as an unknown vector of parameters
and estimates these states from the data.
• Backcast — Estimates initial states using a smoothing filter.

At the command line. Specify the initial states as an argument in the

model-estimation command. For example, use this command to estimate an
ARMAX model and set the initial states to zero:

m=armax(data,[2 2 2 3],'InitialState','zero')

For a complete list of values for the InitialState model property, see the
idpoly reference page.

Algorithms for Estimating Polynomial Models

For linear ARX and AR models, you can choose between the ARX and IV
algorithms. ARX implements the least-squares estimation method that uses
QR-factorization for overdetermined linear equations. IV is the instrumental

3-66
 Identifying Input-Output Polynomial Models

variable method. For more information about IV, see the section on
variance-optimal instruments in System Identification: Theory for the User,
Second Edition, by Lennart Ljung, Prentice Hall PTR, 1999.

The ARX and IV algorithms treat noise differently. ARX assumes white noise.
However, the instrumental variable algorithm, IV, is not sensitive to noise
color. Thus, use IV when the noise in your system is not completely white and
it is incorrect to assume white noise. If the models you obtained using ARX
are inaccurate, try using IV.

Note AR models apply to time-series data, which has no input. For

more information, see Chapter 6, “Time Series Identification”. For more
information about working with AR and ARX models, see “Identifying
Input-Output Polynomial Models” on page 3-39.

Example – Estimating Models Using armax

You can use estimation commands to both construct a model object and
estimate the model parameters. In this example, you estimate a linear,
polynomial model with an ARMAX structure for a three-input and
single-output (MISO) system using the iterative estimation method armax.
For a summary of all available estimation commands in the toolbox, see
“Commands for Model Estimation” on page 1-12.

1 Load a sample data set z8 with three inputs and one output, measured at
1-second intervals and containing 500 data samples:

load iddata8

2 Use armax to both construct the idpoly model object, and estimate the
parameters:

nu
A(q) y(t) = ∑ Bi (q)ui ( t − nki ) + C(q) e(t)
i=1

3-67
 3 Linear Model Identification

Typically you try different model orders and compare results, ultimately
choosing the simplest model that best describes the system dynamics. The
following command specifies the estimation data set, z8, and the orders of
the A, B, and C polynomials as na, nb, and nc, respectively. nk of [0 0 0]
specifies that there is no input delay for all three input channels.

m_armax=armax(z8,'na',4,...
'nb',[3 2 3],...
'nc',4,...
'nk',[0 0 0],...
'focus', 'simulation',...
'tolerance',1e-5,...
'maxiter',50);

covariance, focus, tolerance, and maxiter are optional arguments

specify additional information about the computation. focus specifies
whether the model is optimized for simulation or prediction applications,
tolerance and maxiter specify when to stop estimation. For more
information about these properties, see the algorithm properties
reference page.

armax is a version of pem with simplified syntax for the ARMAX model
structure. The armax method both constructs the idpoly model object
and estimates its parameters.

Tip Instead of specifying model orders and delays as individual

property-value pairs, you can use the equivalent shorthand notation that
includes all of the order integers in a single vector, as follows:

m_armax=armax(z8,[4 3 2 3 4 0 0 0],...
'focus', 'simulation',...
'tolerance',1e-5,...
'maxiter',50);

3-68
 Identifying Input-Output Polynomial Models

3 To view information about the resulting model object, type the following at
the prompt:

m_armax

MATLAB returns the following information about this model object:

Discrete-time IDPOLY model: A(q)y(t) = B(q)u(t) + C(q)e(t)

A(q) = 1 - 1.284 q^-1 + 0.3048 q^-2 + 0.2648 q^-3 - 0.05708 q^-4

B1(q) = -0.07547 + 1.087 q^-1 + 0.7166 q^-2

B2(q) = 1.019 + 0.1142 q^-1

B3(q) = -0.06739 + 0.06829 q^-1 + 0.5509 q^-2

C(q) = 1 - 0.06097 q^-1 - 0.1296 q^-2 + 0.02488 q^-3 - 0.04698 q^-4

Estimated using ARMAX from data set z8

Loss function 0.957009 and FPE 1.02068
Sampling interval: 1

m_armax is an idpoly model object. The coefficients represent estimated

parameters of this polynomial model.

Tip You can use present(m_armax) to show additional information about

the model, including parameter uncertainties.

4 To view all property values for this model, type the following command:

get(m_armax)

MATLAB returns the following information:

ans =

a: [1 -1.2836 0.3048 0.2648 -0.0571]

b: [3x3 double]
c: [1 -0.0610 -0.1296 0.0249 -0.0470]

3-69
 3 Linear Model Identification

d: 1
f: [3x1 double]
da: [0 0.1012 0.1804 0.1210 0.0303]
db: [3x3 double]
dc: [0 0.1111 0.0767 0.0484 0.0460]
dd: 0
df: [3x1 double]
na: 4
nb: [3 2 3]
nc: 4
nd: 0
nf: [0 0 0]
nk: [0 0 0]
InitialState: 'Auto'
Name: ''
Ts: 1
InputName: {3x1 cell}
InputUnit: {3x1 cell}
OutputName: {'y1'}
OutputUnit: {''}
TimeUnit: ''
ParameterVector: [16x1 double]
PName: {}
CovarianceMatrix: [16x16 double]
NoiseVariance: 0.9899
InputDelay: [3x1 double]
Algorithm: [1x1 struct]
EstimationInfo: [1x1 struct]
Notes: {}
UserData: []

3-70
 Identifying Input-Output Polynomial Models

5 The Algorithm and EstimationInfo model properties are structures. To

view the properties and values inside these structure, use dot notation.
For example:

m_armax.Algorithm

This action displays the complete list of Algorithm properties and values
that specify the iterative computational algorithm:

ans =

Focus: 'Simulation'
MaxIter: 50
Tolerance: 1.0000e-005
LimitError: 0
MaxSize: 'Auto'
SearchMethod: 'Auto'
Criterion: 'det'
Weighting: 1
FixedParameter: []
Display: 'Off'
N4Weight: 'Auto'
N4Horizon: 'Auto'
Advanced: [1x1 struct]

3-71
 3 Linear Model Identification

Similarly, to view the properties and values of the EstimationInfo

structure, type the following command:

m_armax.EstimationInfo

This action displays the complete list of read-only EstimationInfo

properties and values that describe the estimation data set, quantitative
measures of model quality (loss function and FPE), the number of iterations
actually used, and the behavior of the iterative model estimation.

ans =

Status: 'Estimated model (PEM with focus)'

Method: 'ARMAX'
LossFcn: 0.9570
FPE: 1.0207
DataName: 'z'
DataLength: 500
DataTs: 1
DataDomain: 'time'
DataInterSample: {3x1 cell}
WhyStop: 'Near (local) minimum, (norm(g) < tol).'
UpdateNorm: 4.7296e-006
LastImprovement: 4.7296e-006
Iterations: 4
InitialState: 'Zero'
Warning: ''

6 If you want to repeat the model estimation using different model orders,
but keep the algorithm properties the same, you can store the model
properties used for m_armax in a variable, as follows:

myAlg=m_armax.Algorithm;

This action stores the specified focus, tolerance, and maxiter, and the
default algorithm.

7 To reuse the algorithm properties in estimating the ARMAX model with

different orders, use the following command:

m_armax2=armax(z8,[4 3 2 3 3 1 1 1],...
'algorithm',myAlg);

3-72
 Identifying State-Space Models

Identifying State-Space Models

In this section...
“What Are State-Space Models?” on page 3-73
“Data Supported by State-Space Models” on page 3-77
“Supported State-Space Parameterizations” on page 3-78
“Preliminary Step – Estimating State-Space Model Orders” on page 3-79
“How to Estimate State-Space Models in the GUI” on page 3-84
“How to Estimate State-Space Models at the Command Line” on page 3-87
“How to Estimate Free-Parameterization State-Space Models” on page 3-91
“How to Estimate State-Space Models with Canonical Parameterization”
on page 3-92
“How to Estimate State-Space Models with Structured Parameterization”
on page 3-94
“How to Estimate the State-Space Equivalent of ARMAX and OE Models”
on page 3-100
“Options for Frequency-Weighing Focus” on page 3-101
“Options for Initial States” on page 3-102
“Algorithms for Estimating State-Space Models” on page 3-103

What Are State-Space Models?

• “Definition of State-Space Models” on page 3-74
• “Continuous-Time Representation” on page 3-74
• “Discrete-Time Representation” on page 3-75
• “Relationship Between Continuous-Time and Discrete-Time State
Matrices” on page 3-75
• “State-Space Representation of Transfer Functions” on page 3-76

3-73
 3 Linear Model Identification

Definition of State-Space Models

State-space models are models that use state variables to describe a system by
a set of first-order differential or difference equations, rather than by one or
more nth-order differential or difference equations. State variables x(t) can be
reconstructed from the measured input-output data, but are not themselves
measured during an experiment.

The state-space model structure is a good choice for quick estimation because
it requires only two parameters:

• n — Model order or the number of poles (size of the A matrix).

• nk — One or more input delays.

The model order for state-space models is an integer equal to the dimension
of x(t) and relates to the number of delayed inputs and outputs used in the
corresponding linear difference equation.

Continuous-Time Representation
In continuous-time, the state-space description has the following form:

 (t)
x (t) = Fx(t) + Gu(t) + Kw
y(t) = Hx(t) + Du(t) + w(t)
x(0) = x0

It is often easier to define a parameterized state-space model in continuous

time because physical laws are most often described in terms of differential
equations. In this case, the matrices F, G, H, and D contain elements with
physical significance—for example, material constants. x0 specifies the initial
states.

Note K=0 gives the state-space representation of an Output-Error model.

For more information about Output-Error models, see “What Are Black-Box
Polynomial Models?” on page 3-39.

3-74
 Identifying State-Space Models

Discrete-Time Representation
Discrete-time state-space models provide the same type of linear difference
relationship between the inputs and the outputs as the linear ARX model, but
are rearranged such that there is only one delay in the expressions. The
discrete-time state-space model structure is often written in the innovations
form that describes noise:

x(kT + T ) = Ax(kT ) + Bu(kT ) + Ke(kT )

y(kT ) = Cx(kT ) + Du(kT ) + e(kT )
x(0) = x0

where T is the sampling interval, u(kT) is the input at time instant kT, and
y(kT) is the output at time instant kT.

Note K=0 gives the state-space representation of an Output-Error model.

For more information about Output-Error models, see “What Are Black-Box
Polynomial Models?” on page 3-39.

Relationship Between Continuous-Time and Discrete-Time

State Matrices
The relationships between the discrete state-space matrices A, B, C, D, and K
and the continuous-time state-space matrices F, G, H, D, and K are given for
piece-wise-constant input, as follows:

A = e FT
T
Fτ
B= ∫e Gdτ
0
C=H

These relationships assume that the input is piece-wise-constant over time

intervals kT ≤ t < (k + 1)T .

3-75
 3 Linear Model Identification

The exact relationship between K and K is complicated. However, for short

sampling intervals T, the following approximation works well:

T
Fτ  τ
K= ∫e Kd
0

State-Space Representation of Transfer Functions

For linear models, the general symbolic model description is given by:

y = Gu + He

G is a transfer function that takes the input u to the output y. H is a transfer

function that describes the properties of the additive output noise model.

The discrete-time state-space representation is given by the following

equation:

x(kT + T ) = Ax(kT ) + Bu(kT ) + Ke(kT )

y(kT ) = Cx(kT ) + Du(kT ) + e(kT )
x(0) = x0

where T is the sampling interval, u(kT) is the input at time instant kT, and
y(kT) is the output at time instant kT.

The relationships between the transfer functions and the discrete-time

state-space matrices are given by the following equations:

G(q) = C(qInx − A)−1 B + D

H (q) = C(qInx − A)−1 K + Iny

where Inx is the nx-by-nx identity matrix, Iny is the nx-by-nx identity matrix,
and ny is the dimension of y and e.

3-76
 Identifying State-Space Models

Data Supported by State-Space Models

• “Types of Supported Data” on page 3-77
• “Estimating Continuous-Time Models” on page 3-77
• “Designating Data for Estimating Discrete-Time Models” on page 3-78

Types of Supported Data

You can estimate linear state-space models from data with the following
characteristics:

• Real data or complex data in any domain

• Single-output and multiple-output
• Time- or frequency-domain data
To estimate state-space models for time-series data, see Chapter 6, “Time
Series Identification”.

You must first import your data into the MATLAB workspace, as described in
Chapter 2, “Data Import and Processing”.

Estimating Continuous-Time Models

Use either of the following ways to estimate continuous-time, state-space
models:

• To get a linear, continuous-time model of arbitrary structure for

time-domain data, you can estimate a discrete-time model, and then use
d2c to transform it to a continuous-time model.
• Use continuous-time frequency-domain data.
To denote continuous-time frequency-domain data, set the data sampling
interval to 0. You can set the sampling interval when you import data into
the GUI or set the Ts property of the data object at the command line.

3-77
 3 Linear Model Identification

Tip Continuous state-space models are available for canonical and

structured parameterizations and grey-box models. In this case, no
disturbance model can be estimated.

Designating Data for Estimating Discrete-Time Models

You can estimate arbitrary-order, linear state-space models for both time- or
frequency-domain data.

You must specify your data to have the sampling interval equal to the
experimental data sampling interval.

You can set the sampling interval when you import data into the GUI or set
the Ts property of the data object at the command line.

Supported State-Space Parameterizations

The System Identification Toolbox product supports the following
parameterizations that indicate which parameters are estimated and which
remain fixed at specific values:

• Free parameterization results in the estimation of all system matrix

elements A, B, C, D, and K.
• Canonical forms of A, B, C, D, and K matrices.
Canonical parameterization represents a state-space system in its minimal
form, using the minimum number of free parameters to capture the
dynamics. Thus, free parameters appear in only a few of the rows and
columns in system matrices A, B, C, and D, and the remaining matrix
elements are fixed to zeros and ones.
• Structured parameterization lets you specify the values of specific
parameters and exclude these parameters from estimation.
• Completely arbitrary mapping of parameters to state-space matrices. For
more information, see “Estimating Linear Grey-Box Models” on page 5-6.

3-78
 Identifying State-Space Models

You can only estimate free state-space models in discrete time.

Continuous state-space models are available for canonical and structured
parameterizations and grey-box models.

Note To estimate canonical and structured state-space models in the System

Identification Tool GUI, define the corresponding model structures at the
command line and import them into the System Identification Tool GUI.

Preliminary Step – Estimating State-Space Model

Orders
• “Why Estimate Model Orders?” on page 3-79
• “Estimating Model Order in the GUI” on page 3-79
• “Estimating the Model Order at the Command Line” on page 3-82
• “Using the Model Order Selection Window” on page 3-83

Why Estimate Model Orders?

To estimate a state-space model, you must provide a model order and one
or more input delays.

To get an initial model order for your system, you can estimate a group of
state-space models with a range of orders for a specific delay and compare the
performance of these models. You choose the model order that include states
with the highest contribution to the input-output behavior of the model and
use this order as an initial guess for further modeling.

The model order is always a single integer—regardless of the number of

inputs and outputs. However, the number of input delays must correspond to
the number of input channels.

Estimating Model Order in the GUI

You must have already imported your data into the GUI, as described in
“Importing Data into the GUI” on page 2-14.

3-79
 3 Linear Model Identification

To estimate model orders for a specific input delay:

1 In the System Identification Tool GUI, select Estimate > Linear

parametric models to open the Linear Parametric Models dialog box.

2 In the Structure list, select State Space: n [nk].

3 Edit the Orders field to specify a range of orders for a specific delay. For
example, enter the following values for n and nk:

1:10 [1]

Tip As a shortcut for entering 1:10 [1], click Order Selection.

4 Verify that the Method is set to N4SID.

3-80
 Identifying State-Space Models

5 Click Estimate to open the Model Order Selection window, which

displays the relative measure of how much each state contributes to the
input-output behavior of the model (log of singular values of the covariance
matrix). The following figure shows an example plot.

6 Select the rectangle that represents the cutoff for the states on the left that
provide a significant contribution to the input-output behavior, and click
Insert to estimate a model with this order. Red indicates the recommended
choice. States 1 and 2 provide the most significant contribution. The
contributions to the right of state 2 drop significantly. For information
about using the Model Order Selection window, see “Using the Model Order
Selection Window” on page 3-83.

This action adds a new model to the Model Board in the System
Identification Tool GUI. The default name of the parametric model
combines the string n4s and the selected model order.

7 Click Close to close the Model Order Selection window.

3-81
 3 Linear Model Identification

After estimating model orders, use this value as an initial guess for estimating
other state-space models, as described in “How to Estimate State-Space
Models in the GUI” on page 3-84.

Estimating the Model Order at the Command Line

You can estimate the state-space model order using the n4sid command.

Use following syntax to specify the range of model orders to try for a specific
input delay.

m = n4sid(data,n1:n2,'nk',nk);

where data is the estimation data set, n1 and n2 specify the range of orders,
and nk specifies the input delay. For multiple-input systems, nk is a vector
of input delays.

This command opens the Model Order Selection window. For information
about using this plot, see “Using the Model Order Selection Window” on page
3-83.

Alternatively, you can use the pem command to open the Model Order
Selection window, as follows:

m = pem(Data,'nx',nn)

where nn = [n1,n2,...,nN] specifies the vector or range of orders you

want to try.

To omit opening the Model Order Selection window and automatically select
the best order, use the following syntax:

m = pem(Data,'best')

For a tutorial on estimating model orders for a multiple-input system, see

“Estimating a State-Space Model” in System Identification Toolbox Getting
Started Guide.

3-82
 Identifying State-Space Models

Using the Model Order Selection Window

You can generate the Model Order Selection window for your data to select
the number of states that provide the highest relative contribution to the
input-output behavior of the model (log of singular values of the covariance
matrix).

For a procedure on generating this plot in the System Identification Tool

GUI, see “Estimating Model Order in the GUI” on page 3-79. To open this
plot at the command line, see “Estimating the Model Order at the Command
Line” on page 3-82.

The following figure shows a sample Model Order Selection window.

The horizontal axis corresponds to the model order n. The vertical axis, called
Log of Singular values, shows the singular values of a covariance matrix
constructed from the observed data.

You use this plot to decide which states provide a significant relative
contribution to the input-output behavior, and which states provide the

3-83
 3 Linear Model Identification

smallest contribution. Based on this plot, select the rectangle that represents
the cutoff for the states on the left that provide a significant contribution to
the input-output behavior. The recommended choice is red.

For example, in the previous figure, states 1 and 2 provide the most significant
contribution. However, the contributions of the states to the right of state 2
drop significantly. This sharp decrease in the log of the singular values after
n=2 indicates that using two states is sufficient to get an accurate model.

How to Estimate State-Space Models in the GUI

• “Supported State-Space Models in the GUI” on page 3-84
• “Prerequisites” on page 3-84
• “Estimating State-Space Models in the GUI” on page 3-85
• “Next Steps” on page 3-87

Supported State-Space Models in the GUI

Only free parameterization is directly supported in the System Identification
Tool GUI. You can also estimate canonical and structured parameterizations
at the command line and import them into the System Identification Tool
GUI for parameter estimation. For more information about state-space
parameterization, see “Supported State-Space Parameterizations” on page
3-78.

Prerequisites
Before you can perform this task, you must have

• Imported data into the System Identification Tool GUI. See “Importing
Time-Domain Data into the GUI” on page 2-16. For supported data
formats, see “Data Supported by State-Space Models” on page 3-77.
• Performed any required data preprocessing operations. To improve the
accuracy of your model, you should detrend your data. See “Ways to
Process Data for System Identification” on page 2-4.

3-84
 Identifying State-Space Models

• Select a model order. For more information about how to estimate model
orders, see “Preliminary Step – Estimating State-Space Model Orders”
on page 3-79.

Estimating State-Space Models in the GUI

To estimate a state-space model with free parameterization in the System
Identification Tool GUI:

1 In the System Identification Tool GUI, select Estimate > Linear

parametric models to open the Linear Parametric Models dialog box.

2 In the Structure list, select State Space: n [nk].

This action updates the options in the Linear Parametric Models dialog box
to correspond with this model structure. For information about each model
structure, see “What Are State-Space Models?” on page 3-73.

3 In the Orders field, specify the model order and delay, as follows:

• For single-input models. Enter the model order integer and the input
delay in terms of the number of samples. Omitting nk uses the default
value nk=1.
For example, enter 4 [2] for a fourth-order model and nk=2.
• For multiple-input models. Enter the model order integer and the
input delay vector—which is a 1-by-nu vector whose ith entry is the
delay for the ith input.
For example, for a two-input system, enter 4 [1 1] for a fourth-order
model and a delay of 1 for each input.
• For multiple-output models. Enter the model order integer the same
way as for single-input models.

Tip To enter model order and any delays using the Order Editor dialog
box, click Order Editor.

3-85
 3 Linear Model Identification

4 Select the estimation Method as N4SID or PEM. For more information

about these methods, “Algorithms for Estimating State-Space Models” on
page 3-103.

5 In the Name field, edit the name of the model or keep the default. The
name of the model should be unique in the Model Board.

6 In the Focus list, select how to weigh the relative importance of the fit at
different frequencies. For more information about each option, see “Options
for Frequency-Weighing Focus” on page 3-101.

7 (PEM only) In the Initial state list, specify how you want the algorithm to
treat initial states. For more information about the available options, see
“Options for Initial States” on page 3-102.

Tip If you get an inaccurate fit, try setting a specific method for handling
initial states rather than choosing it automatically.

8 In the Covariance list, select Estimate if you want the algorithm to

compute parameter uncertainties. Effects of such uncertainties are
displayed on plots as model confidence regions.

To omit estimating uncertainty, select None. Skipping uncertainty

computation reduces computation time for complex models and large data
sets.

9 (PEM only) To view the estimation progress in the MATLAB Command

Window, select the Trace check box. During estimation, the following
information is displayed for each iteration:
• Loss function — Equals the determinant of the estimated covariance
matrix of the input noise.
• Parameter values — Values of the model structure coefficients you
specified.
• Search direction — Change in parameter values from the previous
iteration.
• Fit improvements — Shows the actual versus expected improvements in
the fit.

3-86
 Identifying State-Space Models

10 Click Estimate to add this model to the System Identification Tool GUI.

11 (PEM only) To stop the search and save the results after the current
iteration has been completed, click Stop Iterations. To continue iterations
from the current model, click the Continue iter button to assign current
parameter values as initial guesses for the next search.

Next Steps

• Validate the model by selecting the appropriate check box in the Model
Views area of the System Identification Tool GUI. For more information
about validating models, see “Overview of Model Validation and Plots”
on page 8-2.
• Export the model to the MATLAB workspace for further analysis by
dragging it to the To Workspace rectangle in the System Identification
Tool GUI.

How to Estimate State-Space Models at the Command

Line
• “Supported State-Space Models” on page 3-87
• “Prerequisites” on page 3-87
• “Estimating State-Space Models Using pem and n4sid” on page 3-88
• “Common Properties to Specify Model Estimation” on page 3-89
• “Choosing to Estimate D, K, and X0 Matrices” on page 3-90

Supported State-Space Models

You can only estimate discrete-time state-space models with free
parameterization. Continuous state-space models are available for canonical
and structured parameterizations.

Prerequisites
Before you can perform this task, you must have

3-87
 3 Linear Model Identification

• Regularly sampled data as an iddata object. See “Representing Time- and

Frequency-Domain Data Using iddata Objects” on page 2-49. For supported
data formats, see “Data Supported by State-Space Models” on page 3-77.
• Performed any required data preprocessing operations. To improve the
accuracy of your model, you should detrend your data. See “Ways to
Process Data for System Identification” on page 2-4.
• Select a model order. For more information about how to estimate model
orders, see “Preliminary Step – Estimating State-Space Model Orders”
on page 3-79.

Estimating State-Space Models Using pem and n4sid

You can estimate continuous-time and discrete-time polynomial model using
the iterative estimation command pem that minimizes the prediction errors
to obtain maximum-likelihood values. You can also use the noniterative
subspace command n4sid.

Use the following general syntax to both configure and estimate state-space
models:

m = pem(data,n,
'nk',nk,
'Property1',Value1,...,
'PropertyN',ValueN)

where data is the estimation data, n is the model order, and nk specifies the
input delays for each input.

As an alternative to pem, you can use n4sid:

m = n4sid(data,n,
'nk',nk,
'Property1',Value1,...,
'PropertyN',ValueN)

Note pem uses n4sid to initialize the state-space matrices, and takes longer
than n4sid to estimate a model but typically provides better fit to data.

3-88
 Identifying State-Space Models

For more information about the most common property-value pairs you can
specify, see “Common Properties to Specify Model Estimation” on page 3-89.

For detailed information about the syntax, see the corresponding reference
page.

For more information about estimating model order, see “Estimating the
Model Order at the Command Line” on page 3-82.

For information about validating your model, see “Overview of Model

Validation and Plots” on page 8-2

Common Properties to Specify Model Estimation

The following properties are common to specify in the estimation syntax:

• SSparameterization — Specifies the state-space parameterization

form. For more information about estimating a specific state-space
parameterization, see the following topics:
- “How to Estimate Free-Parameterization State-Space Models” on page
3-91
- “How to Estimate State-Space Models with Canonical Parameterization”
on page 3-92
- “How to Estimate State-Space Models with Structured Parameterization”
on page 3-94
• Focus — Specifies the frequency-weighing of the noise model during
estimation. See “Options for Frequency-Weighing Focus” on page 3-101.
• DisturbanceModel — Specifies to estimate or omit the noise model for
time-domain data. See “K Matrix” on page 3-90.
• InitialStates — Specifies to set or estimate the initial states. See
“Options for Initial States” on page 3-102.

For more information about these properties, see the idss reference page.

3-89
 3 Linear Model Identification

Choosing to Estimate D, K, and X0 Matrices

For state-space models with any parameterization, you can specify whether
to estimate the K and X0 matrices, which represent the noise model and the
initial states, respectively.

For state-space models with structured parameterization, you can also specify
to estimate the D matrix. However, for free and canonical forms, the structure
of the D matrix is set based on your choice of nk.

For more information about state-space structure, see “What Are State-Space
Models?” on page 3-73.

D Matrix. By default, the D matrix is not estimated. Set the model property
nk to estimate the D matrix, as follows:

• To estimate the kth column of D (corresponding to the kth input), set nk to

0. For nu inputs, nk is a 1-by-nu vector.
• To estimate the full D matrix, set all nk values to 0. For example, for two
inputs:

m = pem(Data,n,'nk',[0 0])

To omit estimating the D matrix, set the nk value or values to 1, which is

the default.

K Matrix. K represents the noise model.

For frequency-domain data, no noise model is estimated and K is set to 0. For

time-domain data, K is estimated by default.

To modify whether K is estimated for time-domain data, you can specify the
DisturbanceModel property in the estimator syntax.

3-90
 Identifying State-Space Models

Initially, you can omit estimating the noise parameters in K to focus on

achieving a reasonable model for the system dynamics. After estimating the
dynamic model, you can use pem to refine the model and set the K parameters
to be estimated. For example:

m = pem(Data,md,'DisturbanceModel','Estimate')

where md is the dynamic model without noise.

To set K to zero, set the value of the DisturbanceModel property to 'None'.

For example:

m = pem(Data,n,'DisturbanceModel','None')

XO Matrices. X0 stores the estimated or specified initial states of the model.

To specify how to handle the initial states, set the value of the InitialStates
model property. For example, to set the initial states to zero, set the
InitialStates property to 'zero', as follows:

m = pem(Data,n,'InitialStates','zero')

When you estimate models using multiexperiment data and InitialStates

is set to 'Estimate', X0 stores the estimated initial states corresponding to
the last experiment in the data set.

For a complete list of values for the InitialStates property, see “Options for
Initial States” on page 3-102.

How to Estimate Free-Parameterization State-Space

Models
The default parameterization of the state-space matrices A, B, C, D, and K is
free; that is, any elements in the matrices are adjustable by the estimation
routines. Because the parameterization of A, B, and C is free, a basis for
the state-space realization is automatically selected to give well-conditioned
calculations.

3-91
 3 Linear Model Identification

You can only estimate discrete-time state-space models with any

parameterization. Continuous state-space models are available for canonical
and structured parameterizations only.

To estimate the disturbance model K, you must use time domain data.

Suppose that you have no knowledge about the internal structure of the
discrete-time state-space model. To quickly get started, use the following
syntax:

m = pem(data)

where data is your estimation data. This command estimates a state-space

model for an automatically selected order between 1 and 10.

To find a black-box model of a specific order n, use the following syntax:

m = pem(Data,n)

The iterative algorithm pem is initialized by the subspace method n4sid. You
can use n4sid directly, as an alternative to pem:

m = n4sid(Data,n)

How to Estimate State-Space Models with Canonical

Parameterization
• “What Is Canonical Parameterization?” on page 3-92
• “Estimating Canonical State-Space Models” on page 3-93

What Is Canonical Parameterization?

Canonical parameterization represents a state-space system in its minimal
form, using the minimum number of free parameters to capture the dynamics.
Thus, free parameters appear in only a few of the rows and columns in system
matrices A, B, C, and D, and the remaining matrix elements are fixed to
zeros and ones.

Of the two popular canonical forms, which include controllable canonical

form and observable canonical form, the toolbox supports only controllable

3-92
 Identifying State-Space Models

forms. Controllable canonical structures include free parameters in output

rows of the A matrix, free B and K matrices, and the fixed C matrix. The
representation within controllable canonical forms is not unique and the
exact form depends on the actual choices of canonical indices. For more
information about the distribution of free parameters in canonical forms, see
the appendix on identifiability of black-box multivariable model structures in
System Identification: Theory for the User, Second Edition, by Lennart Ljung,
Prentice Hall PTR, 1999 (equation 4A.16).

Estimating Canonical State-Space Models

You can estimate state-space models with canonical parameterization at the
command line.

To specify a canonical form for A, B, C, and D, set the SSparameterization

model property directly in the estimator syntax, as follows:

m = pem(data,n,'SSparameterization','canonical')

If you have time-domain data, the preceding command estimates a

discrete-time model.

Note When you estimate the D matrix in canonical form, you must set the nk
property. See “Choosing to Estimate D, K, and X0 Matrices” on page 3-90.

If you have continuous-time frequency-domain data, the preceding syntax

estimates an nth order continuous-time state-space model with no direct
contribution from the input to the output (D=0). To include a D matrix, set
the nk property to 0 in the estimation, as follows:

m = pem(data,n,'SSparameterization','canonical',
'nk',0)

You can specify additional property-value pairs similar to the

free-parameterization case, as described in “How to Estimate
Free-Parameterization State-Space Models” on page 3-91.

For information about validating your model, see “Overview of Model

Validation and Plots” on page 8-2.

3-93
 3 Linear Model Identification

How to Estimate State-Space Models with Structured

Parameterization
• “What Is Structured Parameterization?” on page 3-94
• “Specifying the State-Space Structure” on page 3-95
• “Are Grey-Box Models Similar to State-Space Models with Structured
Parameterization?” on page 3-96
• “Example – Estimating Structured Discrete-Time State-Space Models”
on page 3-97
• “Example – Estimating Structured Continuous-Time State-Space Models”
on page 3-98

What Is Structured Parameterization?

Structured parameterization lets you exclude specific parameters from
estimation by setting these parameters to specific values. This approach is
useful when you can derive state-space matrices from physical principles and
provide initial parameter values based on physical insight. You can use this
approach to discover what happens if you fix specific parameter values or if
you free certain parameters.

In the case of structured parameterization, there are two stages to the

estimation procedure:

1 Using the idss command to specify the structure of the state-space

matrices and the initial values of the free parameters

2 Using the pem estimation command to estimate the free model parameters

This approach differs from estimating models with free and canonical
parameterizations, where it is not necessary to specify initial parameter
values before the estimation. For free parameterization, there is no
structure to specify because it is assumed to be unknown. For canonical
parameterization, the structure is fixed to a specific form.

For information about validating your model, see “Overview of Model

Validation and Plots” on page 8-2.

3-94
 Identifying State-Space Models

Specifying the State-Space Structure

To specify the state-space model structure, first define the A, B, C, D, K and X0
matrices in the MATLAB workspace.

To define a discrete-time state-space structure, use the following syntax:

m = idss(A,B,C,D,K,X0,...
'Ts',T,...
'SSparameterization','structured')

where A, B, C, D, and K specify both the fixed parameter values and the
initial values for the free parameters. T is the sampling interval. Setting
SSparameterization to 'structured' flags that you want to estimate a
partial structure for this state-space model.

Similarly, to define a continuous-time state-space structure, use the following

syntax:

m = idss(A,B,C,D,K,X0,...
'Ts',0,...
'SSparameterization','structured')

In the continuous-time case, you must set the sampling interval property Ts
to zero.

After you create the nominal model structure, you must specify which
parameters to estimate and which to set to specific values. To accomplish this,
you must edit the structures of the following model properties: As, Bs, Cs, Ds,
Ks, and x0s. These structure matrices are properties of the nominal model
you constructed and have the same sizes as A, B, C, D, K, and x0, respectively.
Initially, the structure matrices contain NaN values.

Specify the structure matrix values, as follows:

• Set a NaN value to flag free parameters at the corresponding locations in

A, B, C, D, K, and x0.
• Specify the values of fixed parameters at the corresponding locations in
A, B, C, D, K, and x0.

3-95
 3 Linear Model Identification

For example, suppose that you constructed a nominal state-space model m

with the following A matrix:

A = [2 0; 0 3]

Suppose you want to fix A(1,2)=A(2,1)=0. To specify the parameters you

want to fix, enter their values at the corresponding locations in the structure
matrix As:

m.As = [NaN 0; 0 NaN]

The estimation algorithm only estimates the parameters in A that have a

NaN value in As.

Finally, use pem to estimate the model, as described in “How to Estimate

State-Space Models at the Command Line” on page 3-87.

Use physical insight, whenever possible, to initialize the parameters for

the iterative search algorithm. Because it is possible that the numerical
minimization gets stuck in a local minimum, try several different initialization
values for the parameters. For random initialization, use the init command.
When the model structure contains parameters with different orders of
magnitude, try to scale the variables so that the parameters are all roughly
the same magnitude.

The iterative search computes gradients of the prediction errors with respect
to the parameters using numerical differentiation. The step size is specified
by the nuderst command. The default step size is equal to 10–4 times the
absolute value of a parameter or equal to 10–7, whichever is larger. To specify
a different step size, edit the nuderst code file.

Are Grey-Box Models Similar to State-Space Models with

Structured Parameterization?
Structured parameterization state-space models are similar to grey-box
modeling. However, the state-space models are simpler to estimate than
grey-box models. To learn more about grey-box models, see Chapter 5, “ODE
Parameter Estimation (Grey-Box Modeling)”.

3-96
 Identifying State-Space Models

Example – Estimating Structured Discrete-Time State-Space

Models
In this example, you estimate the unknown parameters ( θ1 , θ2 , θ3 , θ4 , θ5 )
in the following discrete-time model:

⎡1 θ1 ⎤ ⎡ θ2 ⎤ ⎡ θ4 ⎤
x(t + 1) = ⎢ ⎥ x(t) + ⎢ ⎥ u(t) + ⎢ ⎥ e(t)
⎣0 1 ⎦ ⎣ θ3 ⎦ ⎣ θ5 ⎦
y(t) = [1 0] x(t) + e(t)
⎡0 ⎤
x(0) = ⎢ ⎥
⎣0 ⎦

Suppose that the nominal values of the unknown parameters ( θ1 , θ2 , θ3 , θ4 , θ5 )

are -1, 2, 3, 4, and 5, respectively.

The discrete-time state-space model structure is defined by the following

equation:

x(kT + T ) = Ax(kT ) + Bu(kT ) + Ke(kT )

y(kT ) = Cx(kT ) + Du(kT ) + e(kT )
x(0) = x0

To construct and estimate the parameters of this discrete-time state-space

model:

1 Construct the parameter matrices and initialize the parameter values

using the nominal parameter values:

A = [1,-1;0,1];
B = [2;3];
C = [1,0];
D = 0;
K = [4;5];

2 Construct the state-space model object:

m = idss(A,B,C,D,K);

3-97
 3 Linear Model Identification

3 Specify the parameter values in the structure matrices that you do not
want to estimate:

m.As = [1, NaN; 0 ,1];

m.Bs = [NaN;NaN];
m.Cs = [1, 0];
m.Ds = 0;
m.Ks = [NaN;NaN];
m.x0s = [0;0];

4 Estimate the model structure:

m = pem(data,m)

where data is name of the iddata object containing time-domain or

frequency-domain data. The iterative search starts with the nominal
values in the A, B, C, D, K, and x0 matrices.

Example – Estimating Structured Continuous-Time State-Space

Models
In this example, you estimate the unknown parameters ( θ1 , θ2 , θ3 ) in the
following continuous-time model:

⎡0 1 ⎤ ⎡0⎤
x (t) = ⎢ ⎥ x(t) + ⎢θ ⎥ u(t)
⎣ 0 θ 1⎦ ⎣ 2⎦
⎡1 0 ⎤
y(t) = ⎢ ⎥ x(t) + e(t)
⎣0 1 ⎦
⎡ θ3 ⎤
x(0) = ⎢ ⎥
⎣0⎦

This equation represents an electrical motor, where y1 (t) = x1 (t) is the

angular position of the motor shaft, and y2 (t) = x2 (t) is the angular velocity.

The parameter −θ1 is the inverse time constant of the motor, and − θ2 θ is the
1
static gain from the input to the angular velocity.

3-98
 Identifying State-Space Models

The motor is at rest at t=0, but its angular position θ3 is unknown. Suppose
that the approximate nominal values of the unknown parameters are θ1 = −1
and θ2 = 0.25 . The variance of the errors in the position measurement is
0.01, and the variance in the angular velocity measurements is 0.1. For
more information about this example, see the section on state-space models
in System Identification: Theory for the User, Second Edition, by Lennart
Ljung, Prentice Hall PTR, 1999.

The continuous-time state-space model structure is defined by the following

equation:

 (t)
x (t) = Fx(t) + Gu(t) + Kw
y(t) = Hx(t) + Du(t) + w(t)
x(0) = x0

To construct and estimate the parameters of this continuous-time state-space

model:

1 Construct the parameter matrices and initialize the parameter values

using the nominal parameter values:

Note The following matrices correspond to continuous-time

representation. However, to be consistent with the idss object property
name, this example uses A, B, and C instead of F, G, and H.

A = [0 1;0 -1];
B = [0;0.25];
C = eye(2);
D = [0;0];
K = zeros(2,2);
x0 = [0;0];

2 Construct the continuous-time state-space model object:

m = idss(A,B,C,D,K,x0,'Ts',0);

3-99
 3 Linear Model Identification

3 Specify the parameter values in the structure matrices that you do not
want to estimate:

m.As = [0 1;0 NaN];

m.Bs = [0;NaN];
m.Cs = m.c;
m.Ds = m.d;
m.Ks = m.k;
m.x0s = [NaN;0]
m.NoiseVariance = [0.01 0; 0 0.1];

4 Estimate the model structure:

m = pem(data,m)

where data is name of the iddata object containing time-domain or

frequency-domain data. The iterative search for a minimum is initialized
by the parameters in the nominal model m. The continuous-time model is
sampled using the same sampling interval as the data.

5 To simulate this system using the sampling interval T = 0.1 for input u
and the noise realization e, use the following commands:

e = randn(300,2);
u = idinput(300);
simdat = iddata([],[u e],'Ts',0.1);
y = sim(m,simdat)

The continuous system is automatically sampled using Ts=0.1. The noise

sequence is scaled according to the matrix m.noisevar.

If you discover that the motor was not initially at rest, you can estimate x2(0)
by setting the second element of the x0s structure matrix to NaN, as follows:

m_new = pem(data,m,'x0s',[NaN;NaN])

How to Estimate the State-Space Equivalent of

ARMAX and OE Models
You can estimate the equivalent of ARMAX and output-error (OE)
multiple-output models using state-space model structures. For the ARMAX

3-100
 Identifying State-Space Models

case, specify to estimate the K matrix for the state-space model. For the OE
case, set K = 0.

Tip You can use a state-space model with K = 0 (Output-Error (OE) form)
for initializing a Hammerstein-Wiener estimation at the command line. This
initialization may improve the fit of the model. See “Using Linear Model for
Hammerstein-Wiener Estimation” on page 4-64.

For more information about ARMAX and OE models, see “Identifying

Input-Output Polynomial Models” on page 3-39.

Options for Frequency-Weighing Focus

You can specify how the estimation algorithm weighs the fit at various
frequencies. This information supports the estimation procedures “How to
Estimate State-Space Models in the GUI” on page 3-84 and “How to Estimate
State-Space Models at the Command Line” on page 3-87.

In the System Identification Tool GUI. Set Focus to one of the following
options:

• Prediction — Uses the inverse of the noise model H to weigh the relative
importance of how closely to fit the data in various frequency ranges.
Corresponds to minimizing one-step-ahead prediction, which typically
favors the fit over a short time interval. Optimized for output prediction
applications.
• Simulation — Uses the input spectrum to weigh the relative importance of
the fit in a specific frequency range. Does not use the noise model to weigh
the relative importance of how closely to fit the data in various frequency
ranges. Optimized for output simulation applications.
• Stability — Estimates the best stable model. For more information about
model stability, see “Unstable Models” on page 8-66.
• Filter — Specify a custom filter to open the Estimation Focus dialog box,
where you can enter a filter, as described in “Simple Passband Filter” on
page 2-113 or “Defining a Custom Filter” on page 2-114. This prefiltering

3-101
 3 Linear Model Identification

applies only for estimating the dynamics from input to output. The
disturbance model is determined from the estimation data.

At the command line. Specify the focus as an argument in the

model-estimation command using the same options as in the GUI. For
example, use this command to emphasize the fit between the 5 and 8 rad/s:

pem(data,4,'Focus',[5 8])

Options for Initial States

If you estimate state-space models using the iterative estimation algorithm
pem, you must specify how the algorithm treats initial states. This information
supports the estimation procedures “How to Estimate State-Space Models
in the GUI” on page 3-84 and “How to Estimate State-Space Models at the
Command Line” on page 3-87.

In the System Identification Tool GUI. Set Initial state to one of the
following options:

• Auto — Automatically chooses Zero, Estimate, or Backcast based on the

estimation data. If initial states have negligible effect on the prediction
errors, the initial states are set to zero to optimize algorithm performance.
• Zero — Sets all initial states to zero.
• Estimate — Treats the initial states as an unknown vector of parameters
and estimates these states from the data.
• Backcast — Estimates initial states using a backward filtering method
(least-squares fit).

At the command line. Specify the initial states as an argument in the

estimation command pem. For example, use this command to estimate a
fourth-order state-space model and set the initial states to be estimated from
the data:

m=pem(data,4,'InitialState','estimate')

For a complete list of values for the InitialState model property, see the
idss reference page.

3-102
 Identifying State-Space Models

Algorithms for Estimating State-Space Models

For linear state-space models, you can use the subspace method, called
N4SID. You can use the subspace method N4SID to get an initial model (see
the n4sid reference page), and then try to refine the initial estimate using the
iterative prediction-error method PEM (see the pem reference page).

N4SID is faster than PEM, but is typically less accurate and robust, and
requires additional arguments that might be difficult to specify.

You can use the iterative prediction-error minimization (PEM) (maximum

likelihood) algorithm for all linear and nonlinear model types.

3-103
 3 Linear Model Identification

Refining Linear Parametric Models

In this section...
“When to Refine Models” on page 3-104
“What You Specify to Refine a Model” on page 3-104
“How to Refine Linear Parametric Models in the GUI” on page 3-105
“How to Refine Linear Parametric Models at the Command Line” on page
3-106

When to Refine Models

There are two situations where you can refine estimates of linear parametric
models.

In the first situation, you have already estimated a parametric model and
wish to refine the model. However, if your model captures the essential
dynamics, it is usually not necessary to continue improving the fit—especially
when the improvement is a fraction of a percent.

In the second situation, you might have constructed a model using one of
the model constructors described in “Commands for Constructing Model
Structures” on page 1-16. In this case, you built initial parameter guesses into
the model structure and wish to refine these parameter values.

Note Because it is difficult to specify nonlinear model parameters in

advance, you typically only estimate nonlinear models.

What You Specify to Refine a Model

When you refine a model, you must provide two inputs:

• Parametric model
• Data — You can either use the same data set for refining the model as the
one you originally used to estimate the model, or you can use a different
data set.

3-104
 Refining Linear Parametric Models

How to Refine Linear Parametric Models in the GUI

The following procedure assumes that the model you want to refine is already
in the System Identification Tool GUI. You might have estimated this model
in the current session or imported the model from the MATLAB workspace.
For information about importing models into the GUI, see “Importing Models
into the GUI” on page 11-9.

To refine your model:

1 In the System Identification Tool GUI, verify that you have the correct data
set in the Working Data area for refining your model.

If you are using a different data set than the one you used to estimate the
model, drag the correct data set into the Working Data area. For more
information about specifying estimation data, see “Specifying Estimation
and Validation Data” on page 2-31.

2 Select Estimate > Linear parametric models to open the Linear

Parametric Models dialog box, if this dialog box is not already open.

3 In the Linear Parametric Models dialog box, select By Initial Model from
the Structure list.

4 Enter the model name into the Initial model field, and press Enter.

The model name must be in the Model Board of the System Identification
Tool GUI or a variable in the MATLAB workspace.

Tip As a shortcut for specifying a model in the Model Board, you can drag
the model icon from the System Identification Tool GUI into the Initial
model field.

When you enter the model name, algorithm settings in the Linear
Parametric Models dialog box override the initial model settings.

5 Modify the algorithm settings, displayed in the Linear Parametric Models

dialog box, if necessary.

6 Click Estimate to refine the model.

3-105
 3 Linear Model Identification

7 Validate the new model, as described in Chapter 8, “Model Analysis”.

Tip To continue refining the model using additional iterations, click

Continue iter. This action continues parameter estimation using the most
recent model.

How to Refine Linear Parametric Models at the

Command Line
If you are working at the command line, you can use pem to refine parametric
model estimates.

The general syntax for refining initial models is as follows:

m = pem(data,init_model)

pem uses the properties of the initial model unless you specify different
properties. For more information about specifying model properties directly in
the estimator, see “Specifying Model Properties for Estimation” on page 1-19.

3-106
 Refining Linear Parametric Models

Example – Refining an Initial ARMAX Model at the Command

Line
The following example shows to estimate an initial model and try to refine
this model using pem:

load iddata8

% Split the data z8 into two parts.

% Create new data object with first hundred samples
z8a = z8(1:100);

% Create new data object with remaining samples

z8b = z8(101:end);

% Estimate ARMAX model with default Algorithm

% properties, na=4, nb=[3 2 3], nc=2, and nk=[0 0 0]
m1 = armax(z8a,[4 3 2 3 2 0 0 0]);

% Refine the initial model m1 using the data set z8b,

% and stricter algorithm settings with increased number
% of maximum iterations (MaxIter) and smaller tolerance
m2 = pem(z8b,m1,'tol',1e-5,'maxiter',50);

For more information about estimating polynomial models, see “Identifying

Input-Output Polynomial Models” on page 3-39.

3-107
 3 Linear Model Identification

Example – Refining an ARMAX Model with Initial Parameter

Guesses at the Command Line
The following example shows how to refine models for which you have initial
parameter guesses. This example estimates an ARMAX model for the data
and requires you to initialize the A, B, and C polynomials.

In this case, you must first create a model object and set the initial parameter
values in the model properties. Next, you provide this initial model as input
to pem, which refines the initial parameter guesses using the data.

load iddata8
% Define model parameters.
% Leading zeros in B indicate input delay (nk),
% which is 1 for each input channel.
A = [1 -1.2 0.7];
B{1} = [0 1 0.5 0.1]; % first input
B{2} = [0 1.5 -0.5]; % second input
B{3} = [0 -0.1 0.5 -0.1]; % third input
C = [1 0 0 0 0];
Ts = 1;

% Create model object.

init_model = idpoly(A,B,C,'Ts',1);

% Use pem to refine initial model.

model = pem(z8,init_model)

% Compare the two models.

compare(z8,init_model,model)

For more information about estimating polynomial models, see “Identifying

Input-Output Polynomial Models” on page 3-39.

3-108
 Extracting Parameter Values from Linear Models

Extracting Parameter Values from Linear Models

You can extract the numerical parameter values and uncertainties of
linear model objects. For example, you can extract state-space matrices for
state-space models, and polynomials for polynomial models.

Tip To view a list of model properties, type get(model).

You can operate on extracted model data as you would on any other MATLAB
vectors, matrices and cell arrays. You can also pass these numerical values to
Control System Toolbox commands, for example, or Simulink blocks.

If you specified to estimate model uncertainty data, this information is stored

in the property Model.CovarianceMatrix in the estimated model. The
covariance matrix is used to compute uncertainties in parameter estimates,
model output plots, Bode plots, residual plots, and pole-zero plots.

You can use dot notation to access model properties. For example, m.A
displays the A polynomial coefficients from model m. Alternatively, you can
use the get command, as follows: get(m,'A').

The following table summarizes commands for extracting numerical data

from models. All of these commands have the following syntax form:

[G,dG] = command(model)

where G stores model parameters and dG stores standard deviation of

parameters or covariance.

3-109
 3 Linear Model Identification

Commands for Extracting Numerical Model Data

Command Description Syntax

arxdata Extracts ARX
parameters from [A,B,dA,dB] = arxdata(m)
multiple-output idarx
or single-output idpoly
objects that represent
ARX models.
freqresp Extracts
frequency-response [H,w,CovH] = freqresp(m)
data from any idmodel
or idfrd object.
polydata Extracts polynomials
from any single-output [A,B,C,D,F,dA,dB,dC,dD,dF] = ...
idmodel object. polydata(m)

ssdata Extracts state-space

matrices from any [A,B,C,D,K,X0,...
idmodel object. dA,dB,dC,dD,dK,dX0] = ...
ssdata(Model)
tfdata Extracts numerator
and denominator [Num,Den,dNum,dDen] = ...
polynomials from any tfdata(Model)
idmodel object.
zpkdata Extracts zeros, poles,
and transfer function [Z,P,K,covZ,covP,covK] = ...
gains from any idmodel zpkdata(m)
object.

3-110
 Extracting Dynamic Model and Noise Model Separately

Extracting Dynamic Model and Noise Model Separately

You can extract the numerical data associated with a dynamic model and
the noise model separately.

For linear models, the general symbolic model description is given by:

y = Gu + He

G is an operator that takes the measured inputs u to the outputs and captures
the system dynamics. H is an operator that describes the properties of the
additive output disturbance and takes the hypothetical (unmeasured) noise
source inputs e to the outputs, also called the noise model. When you estimate
a noise model, the toolbox includes one noise channel e at the input for each
output in your system.

The following table summarizes the results of ssdata, tfdata, and zpkdata
commands for extracting the numerical values of the dynamic model and
noise model separately. fcn represents ssdata, tfdata, and zpkdata, and
m is a model object. L represents the covariance matrix e, as defined in
“Subreferencing Measured and Noise Models” on page 3-121.

For information about subreferencing noise channels or treating noise

channels as measured input, see “Subreferencing Model Objects” on page
3-120.

Note The syntax fcn(m('noise')) is equivalent to fcn(m('n')).

Syntax for Extracting Transfer-Function Data

Command Syntax
fcn(m) Returns the properties of G for ny outputs and nu inputs.
fcn(m('noise')) Returns the properties of H for ny outputs and ny inputs.
fcn(noisecnv(m)) Returns the properties of [G H] ny outputs and ny+nu
inputs.

3-111
 3 Linear Model Identification

Syntax for Extracting Transfer-Function Data (Continued)

Command Syntax
fcn(noisecnv(m,'Norm')) Returns the properties of [G HL] ny outputs and ny+nu
inputs.
fcn(noisecnv(m('noise'),'Norm')) Returns the properties of HL ny outputs and ny inputs.
fcn(m) If m is a time-series model, returns the properties of H.
fcn(noisecnv(m,'Norm')) If m is a time-series model, returns the properties of HL.

Note The estimated covariance matrix NoiseVariance is uncertain. Thus,

the uncertainty of H differs from the uncertainty of HL.

3-112
 Transforming Between Discrete-Time and Continuous-Time Representations

Transforming Between Discrete-Time and Continuous-Time

Representations
In this section...
“Why Transform Between Continuous and Discrete Time?” on page 3-113
“Using the c2d, d2c, and d2d Commands” on page 3-113
“Specifying Intersample Behavior” on page 3-115
“How d2c Handles Input Delays” on page 3-115
“Effects on the Noise Model” on page 3-116

Why Transform Between Continuous and Discrete

Time?
Transforming between continuous-time and discrete-time representations is
useful, for example, if you have estimated a discrete-time linear model and
require a continuous-time model instead.

d2d is useful is you want to change the sampling interval of a discrete

model. All of these operations change the sampling interval, which is called
resampling the model.

Using the c2d, d2c, and d2d Commands

You can use c2d and d2c to transform any idmodel object between
continuous-time and discrete-time representations.

The following table summarizes the commands for transforming between

continuous-time and discrete-time model representations. These commands
also transform the estimated model uncertainty, which corresponds to the
estimated covariance matrix of the parameters. For detailed information
about these commands, see the corresponding reference page.

3-113
 3 Linear Model Identification

Note c2d and d2d correctly approximate the transformation of the noise
model when the sampling interval T is small compared to the bandwidth
of the noise.

Command Description Usage Example

c2d Converts To transform a continuous-time
continuous-time model mod_c to a discrete-time
models to discrete-time form, use the following
models. command:

mod_d = c2d(mod_c,T)

where T is the sampling interval

of the discrete-time model.
d2c Converts parametric To transform a discrete-time
discrete-time models model mod_d to a
to continuous-time continuous-time form, use
models. the following command:

mod_c = d2c(mod_d)
d2d Resample a linear To resample a discrete-time
discrete-time model mod_d1 to a discrete-time
model and produce form with a new sampling
an equivalent interval Ts, use the following
discrete-time model command:
with a new sampling
interval. mod_d2 = d2d(mod_d1,Ts)
You can use the
resampled model to
simulate or predict
output with a specified
time interval.

3-114
 Transforming Between Discrete-Time and Continuous-Time Representations

The following commands compare estimated model m and its continuous-time

counterpart mc on a Bode plot:

% Estimate discrete-time ARMAX model

% from the data
m = armax(data,[2 3 1 2]);
% Convert to continuous-time form
mc = d2c(m);
% Plot bode plot for both models
bode(m,mc)

Specifying Intersample Behavior

A sampled signal is characterized only by its values at the sampling instants.
However, when you apply a continuous-time input to a continuous-time
system, the output values at the sampling instants depend on the inputs at
the sampling instants and on the inputs between these points. Thus, the
InterSample data property describes how the algorithms should handle the
input between samples. For example, you can specify the behavior between
the samples to be piece-wise constant (zero-order hold, zoh) or linearly
interpolated between the samples (first order hold, foh). The transformation
formulas for c2d and d2c are affected by the intersample behavior of the input.

By default, c2d and d2c use the intersample behavior you assigned to the
estimation data. To override this setting during transformation, add an extra
argument in the syntax. For example:

% Set first-order hold intersample behavior

mod_d = c2d(mod_c,T,'foh')

How d2c Handles Input Delays

The discrete-to-continuous-time conversion d2c properly handles any
input delays in the discrete-time model, and stores this information in the
continuous-time model. An input delay is the delay in the response of the
output to the input signal.

The relationship between discrete-time and continuous-time delays depends

on the input intersample behavior. For example, a continuous-time system
without a delay shows a delay when sampled with a zero-order-hold input.

3-115
 3 Linear Model Identification

A delay in the discrete-time model that corresponds to an actual delay in the

continuous-time model is stored in the in the InputDelay property of the
resulting continuous-time model. Typically, this InputDelay is (nk-1)/Ts,
where nk is the delay of the discrete-time system and Ts is the sampling
interval.

Note Unlike for discrete-time models, the nk property of continuous-time

model is only used to flag when immediate response to step changes is
present; nk is not used to store input delays greater than or equal to 1. When
nk(i)=0, then there is an immediate response to a step change in the input
ith. When nk(i)=1, then there is no immediate response to the input.

Effects on the Noise Model

c2d, d2c, and d2d change the sampling interval of both the dynamic model and
the noise model. Resampling a model affects the variance of its noise model.

A parametric noise model is a time-series model with the following

mathematical description:

y(t) = H (q) e(t)

Ee2 = λ

The noise spectrum is computed by the following discrete-time equation:

( )
2
Φ v (ω) = λT H eiωT

where λ is the variance of the white noise e(t), and λT represents the spectral
density of e(t). Resampling the noise model preserves the spectral density λ T
. The spectral density λ T is invariant up to the Nyquist frequency. For more
information about spectrum normalization, see “Spectrum Normalization”
on page 3-8.

d2d resampling of the noise model affects simulations with noise using
sim. If you resample a model to a faster sampling rate, simulating this
model results in higher noise level. This higher noise level results from the

3-116
 Transforming Between Discrete-Time and Continuous-Time Representations

underlying continuous-time model being subject to continuous-time white

noise disturbances, which have infinite, instantaneous variance. In this
case, the underlying continuous-time model is the unique representation for
discrete-time models. To maintain the same level of noise after interpolating

the noise signal, scale the noise spectrum by TNew T , where Tnew is the new
Old
sampling interval and Told is the original sampling interval. before applying
sim.

c2d and d2c transformations produce warnings when the continuous-time

disturbance model does not have the required white-noise component.
These warnings occur because the underlying state-space model, which
is formed and used by these transformations, is ill-defined. In this case,
modify the C-polynomial such that the degree of the monic C-polynomial
in continuous-time equals the sum of the degrees of the monic A- and
D-polynomials in continuous-time. For example:

length(C)-1 = (length(A)-1)+(length(D)-1)

3-117
 3 Linear Model Identification

Transforming Between Linear Model Representations

You can transform linear models between state-space and polynomial forms.
You can also transform between frequency-response, state-space, and
polynomial forms.

If you used the System Identification Tool GUI to estimate models, you must
export the models to the MATLAB workspace before converting models.

For detailed information about each command in the following table, see the
corresponding reference page.

Commands for Transforming Model Representations

Command Model Type to Convert Usage Example

idfrd Converts any single- or To get frequency response of m at default
multiple-output idmodel frequencies, use the following command:
object to idfrd model.
If you have the Control m_f = idfrd(m)
System Toolbox product, this
command converts any LTI To get frequency response at specific
object. frequencies, use the following command:

m_f = idfrd(m,f)

To get frequency response for a submodel

from input 2 to output 3, use the
following command:

m_f = idfrd(m(2,3))
idpoly Converts single-output To get an ARMAX model from state-space
idmodel object to ARMAX model m_ss, use the following command:
representation.
If you have the Control m_p = idpoly(m_ss)
System Toolbox product,
this command converts any
single-output LTI object
except frd.

3-118
 Transforming Between Linear Model Representations

Commands for Transforming Model Representations (Continued)

Command Model Type to Convert Usage Example

idss Converts any single- or To get a state-space model from an ARX
multiple-output idmodel model m_arx, use the following command:
object to state-space
representation. m_ss = idss(m_arx)
If you have the Control
System Toolbox product, this
command converts any LTI
object except frd.

Note The idss conversion produces warnings when the continuous-time

disturbance model does not have the required white-noise component.
These warnings occur because the underlying state-space model, which
is formed and used by these transformations, is ill defined. In this case,
modify the C-polynomial such that the degree of the monic C-polynomial
in continuous-time equals the sum of the degrees of the monic A- and
D-polynomials in continuous-time. For example:

length(C)-1 = (length(A)-1)+(length(D)-1)

3-119
 3 Linear Model Identification

Subreferencing Model Objects

In this section...
“What Is Subreferencing?” on page 3-120
“Limitation on Supported Models” on page 3-120
“Subreferencing Specific Measured Channels” on page 3-120
“Subreferencing Measured and Noise Models” on page 3-121
“Treating Noise Channels as Measured Inputs” on page 3-123

What Is Subreferencing?
You can use subreferencing to create models with subsets of inputs and
outputs from existing multivariable models. Subreferencing is also useful
when you want to generate model plots for only certain channels, such as
when you are exploring multiple-output models for input channels that have
minimal effect on the output.

The toolbox supports subreferencing operations for idarx, idgrey, idpoly,

idproc, idss, and idfrd model objects.

In addition to subreferencing the model for specific combinations of measured

inputs and output, you can subreference dynamic and noise models
individually.

Limitation on Supported Models

Subreferencing nonlinear models is not supported.

Subreferencing Specific Measured Channels

Use the following general syntax to subreference specific input and output
channels in models:

model(outputs,inputs)

In this syntax, outputs and inputs specify channel indexes or channel names.

3-120
 Subreferencing Model Objects

To select all output or all input channels, use a colon (:). To select no
channels, specify an empty matrix ([]). If you need to reference several
channel names, use a cell array of strings.

For example, to create a new model m2 from m from inputs 1 ('power') and
4 ('speed') to output number 3 ('position'), use either of the following
equivalent commands:

m2 = m('position',{'power','speed'})

or

m2 = m(3,[1 4])

For a single-output model, you can use the following syntax to subreference
specific input channels without ambiguity:

m3 = m(inputs)

Similarly, for a single-input model, you can use the following syntax to
subreference specific output channels:

m4 = m(outputs)

Subreferencing Measured and Noise Models

For linear models, the general symbolic model description is given by:

y = Gu + He

G is an operator that takes the measured inputs u to the outputs and captures
the system dynamics.

H is an operator that describes the properties of the additive output

disturbance and takes the hypothetical (unmeasured) noise source inputs to
the outputs. H represents the noise model. When you specify to estimate a
noise model, the resulting model include one noise channel e at the input for
each output in your system.

3-121
 3 Linear Model Identification

Thus, linear, parametric models represent input-output relationships for two

kinds of input channels: measured inputs and (unmeasured) noise inputs. For
example, consider the ARX model given by one of the following equations:

A(q) y(t) = B(q)u ( t − nk ) + e(t)

or

B(q) 1
y(t) = u(t) + e(t)
A(q) A(q)

In this case, the dynamic model is the relationship between the measured

input u and output y, G = B( q) A( q) . The noise model is the contribution of the

input noise e to the output y, given by H = 1 A( q) .

Suppose that the model m contains both a dynamic model G and a noise model
H. To create a new model by subreferencing G due to measured inputs, use
the following syntax:

m_G = m('measured')

Tip Alternatively, you can use the following shorthand syntax: m_G = m('m')

3-122
 Subreferencing Model Objects

To create a new model by subreferencing H due to unmeasured inputs, use

the following syntax:

m_H = m('noise')

Tip Alternatively, you can use the following shorthand syntax: m_H = m('n')

This operation creates a time-series model from m by ignoring the measured

input.

The covariance matrix of e is given by the idmodel property NoiseVariance,

which is the matrix Λ :

Λ = LLT

The covariance matrix of e is related to v, as follows:

e = Lv
where v is white noise with an identity covariance matrix representing
independent noise sources with unit variances.

Treating Noise Channels as Measured Inputs

To study noise contributions in more detail, it might be useful to convert the
noise channels to measured channels using noisecnv:

m_GH = noisecnv(m)

This operation creates a model m_GH that represents both measured inputs u
and noise inputs e, treating both sources as measured signals. m_GH is a model
from u and e to y, describing the transfer functions G and H.

3-123
 3 Linear Model Identification

Converting noise channels to measured inputs loses information about

the variance of the innovations e. For example, step response due to the
noise channels does not take into consideration the magnitude of the noise
contributions. To include this variance information, normalize e such that v
becomes white noise with an identity covariance matrix, where

e = Lv

To normalize e, use the following command:

m_GH = noisecnv(m,'Norm')

This command creates a model where u and v are treated as measured

signals, as follows:

⎡u ⎤
y(t) = Gu(t) + HLv = [ G HL] ⎢ ⎥
⎣v⎦

For example, the scaling by L causes the step responses from v to y to reflect
the size of the disturbance influence.

The converted noise sources are named in a way that relates the noise channel
to the corresponding output. Unnormalized noise sources e are assigned
names such as 'e@y1', 'e@y2', ..., 'e@yn', where 'e@yn' refers to the noise
input associated with the output yn. Similarly, normalized noise sources v,
are named 'v@y1', 'v@y2', ..., 'v@yn'.

Note When you plot models in the GUI that include noise sources, you
can select to view the response of the noise model corresponding to specific
outputs. For more information, see “Selecting Measured and Noise Channels
in Plots” on page 11-18.

3-124
 Concatenating Model Objects

Concatenating Model Objects

In this section...
“About Concatenating Models” on page 3-125
“Limitation on Supported Models” on page 3-125
“Horizontal Concatenation of Model Objects” on page 3-126
“Vertical Concatenation of Model Objects” on page 3-126
“Concatenating Noise Spectrum Data of idfrd Objects” on page 3-127
“See Also” on page 3-128

About Concatenating Models

You can perform horizontal and vertical concatenation of linear model objects
to grow the number of inputs or outputs in the model.

When you concatenate parametric models, such as idarx, idgrey, idpoly,

idproc, and idss model objects, the resulting model combines the parameters
of the individual models.

You can also concatenate nonparametric models, which contain the estimated
impulse-response (idarx object) and frequency-response (idfrd object) of a
system.

In case of idfrd models, concatenation combines information in the

ResponseData properties of the individual model objects. ResponseData is
an ny-by-nu-by-nf array that stores the response of the system, where ny
is the number of output channels, nu is the number of input channels, and
nf is the number of frequency values. The (j,i,:) vector of the resulting
response data represents the frequency response from the ith input to the
jth output at all frequencies.

Limitation on Supported Models

Concatenation is supported for linear models only.

3-125
 3 Linear Model Identification

Horizontal Concatenation of Model Objects

Horizontal concatenation of model objects requires that they have the same
outputs. If the output channel names are different and their dimensions are
the same, the concatenation operation uses the names of output channels in
the first model object you listed. Input channels must have unique names.

The following syntax creates a new model object m that contains the horizontal
concatenation of m1,m2,...,mN:

m = [m1,m2,...,mN]

m takes all of the inputs of m1,m2,...,mN to the same outputs as in the

original models. The following diagram is a graphical representation of
horizontal concatenation of the models.

u1 y1 y1
Model 1 u3 Model 2
u2 y2 y2

u1 y1
Combined Horizonal Concatenation Same
Inputs u2 of Model 1 and Model 2 Outputs
y2
u3

Note Horizontal concatenation of idarx objects creates an idss object.

Vertical Concatenation of Model Objects

Vertical concatenation combines output channels of specified models. Vertical
concatenation of model objects requires that they have the same inputs
and frequency vectors. If the input channel names are different and their
dimensions are the same, the concatenation operation uses the names of
input channels in the first model object you listed. Output channels must
have unique names.

3-126
 Concatenating Model Objects

Note You cannot concatenate the single-output idproc and idpoly model
objects.

The following syntax creates a new model object m that contains the vertical
concatenation of m1,m2,...,mN:

m = [m1;m2;... ;mN]

m takes the same inputs in the original models to all of the output of
m1,m2,...,mN. The following diagram is a graphical representation of vertical
concatenation of frequency-response data.

u1 y1 u1
Model 1 Model 2 y3
u2 y2 u2

u1 y1
Same Vertical Concatenation y2 Combined
Inputs u2 of Model 1 and Model 2 Outputs
y3

Concatenating Noise Spectrum Data of idfrd Objects

When the idfrd objects contain the frequency-response data you measured
or constructed manually, the concatenation operation combines only the
ResponseData properties. Because noise spectrum data does not exist (unless
you also entered it manually), SpectrumData is empty in both the individual
idfrd objects and the concatenated idfrd object.

However, when the idfrd objects are spectral models that you estimated, the
SpectrumData property is not empty and contains the power spectra and
cross spectra of the output noise in the system. For each output channel, this
toolbox estimates one noise channel to explain the difference between the
output of the model and the measured output.

3-127
 3 Linear Model Identification

When the SpectrumData property of individual idfrd objects is not empty,

horizontal and vertical concatenation handle SpectrumData, as follows.

In case of horizontal concatenation, there is no meaningful way to combine the

SpectrumData of individual idfrd objects, and the resulting SpectrumData
property is empty. An empty property results because each idfrd object has
its own set of noise channels, where the number of noise channels equals the
number of outputs. When the resulting idfrd object contains the same output
channels as each of the individual idfrd objects, it cannot accommodate the
noise data from all the idfrd objects.

In case of vertical concatenation, this toolbox concatenates individual noise

models diagonally. The following shows that m.SpectrumData is a block
diagonal matrix of the power spectra and cross spectra of the output noise in
the system:

⎛ m1.s
⎜ 0 ⎞⎟
m.s = ⎜  ⎟
⎜⎜ ⎟
⎝ 0 mN .s ⎟⎠

s in m.s is the abbreviation for the SpectrumData property name.

See Also
If you have the Control System Toolbox product, see “Combining Model
Objects” on page 9-6 about additional functionality for combining models.

3-128
 Merging Model Objects

Merging Model Objects

You can merge models of the same structure to obtain a single model with
parameters that are statistically weighed means of the parameters of the
individual models. When computing the merged model, the covariance
matrices of the individual models determine the weights of the parameters.

You can perform the merge operation for the idarx, idgrey, idpoly, idproc,
and idss model objects.

Note Each merge operation merges the same type of model object.

Merging models is an alternative to merging data sets into a single

multiexperiment data set, and then estimating a model for the merged data.
Whereas merging data sets assumes that the signal-to-noise ratios are about
the same in the two experiments, merging models allows greater variations
in model uncertainty, which might result from greater disturbances in an
experiment.

When the experimental conditions are about the same, merge the data
instead of models. This approach is more efficient and typically involves
better-conditioned calculations. For more information about merging data
sets into a multiexperiment data set, see “Creating Multiexperiment Data at
the Command Line” on page 2-55.

For more information about merging models, see the merge reference page.

3-129
 3 Linear Model Identification

3-130
 4

Nonlinear Black-Box Model

Identification

• “About Nonlinear Model Identification” on page 4-2

• “Preparing Data for Nonlinear Identification” on page 4-7
• “Identifying Nonlinear ARX Models” on page 4-8
• “Identifying Hammerstein-Wiener Models” on page 4-49
• “Linear Approximation of Nonlinear Black-Box Models” on page 4-81
 4 Nonlinear Black-Box Model Identification

About Nonlinear Model Identification

In this section...
“What Are Nonlinear Models?” on page 4-2
“When to Fit Nonlinear Models” on page 4-2
“Available Nonlinear Models” on page 4-4

What Are Nonlinear Models?

Dynamic models in System Identification Toolbox software are mathematical
relationships between the system’s inputs u(t) and outputs y(t). You can use
these relationships to compute the current output from previous inputs and
outputs. The general form of a model in discrete time is:

y(t) = f(u(t - 1), y(t - 1), u(t - 2), y(t - 2), . . .)

Such a model is nonlinear if the function f is a nonlinear function, which

might include nonlinear components representing arbitrary nonlinearities,
such as switches and saturations.

When to Fit Nonlinear Models

In practice, all systems are nonlinear and the output is a nonlinear function of
the input variables. However, a linear model is often sufficient to accurately
describe the system dynamics. In most cases, you should first try to fit linear
models.

Here are some scenarios when you might need the additional flexibility of
nonlinear models:

• “Linear Model Is Not Good Enough” on page 4-3

• “Physical System Is Weakly Nonlinear” on page 4-3
• “Physical System Is Inherently Nonlinear” on page 4-3
• “Linear and Nonlinear Dynamics Are Captured Separately” on page 4-4

4-2
 About Nonlinear Model Identification

Linear Model Is Not Good Enough

You might need nonlinear models when a linear model provides a poor fit
to the measured output signals and cannot be improved by changing the
model structure or order. Nonlinear models have more flexibility in capturing
complex phenomena than the linear models of similar orders.

Physical System Is Weakly Nonlinear

From physical insight or data analysis, you might know that a system
is weakly nonlinear. In such cases, you can estimate a linear model and
then use this model an initial model for nonlinear estimation. Nonlinear
estimation can improve the fit by using nonlinear components of the model
structure to capture the dynamics not explained by the linear model. For
more information, see “Using Linear Model for Nonlinear ARX Estimation”
on page 4-28 and “Using Linear Model for Hammerstein-Wiener Estimation”
on page 4-64.

Physical System Is Inherently Nonlinear

You might have physical insight that your system is nonlinear. Certain
phenomena are inherently nonlinear in nature, including dry friction in
mechanical systems, actuator power saturation, and sensor nonlinearities in
electro-mechanical systems. You can try modeling such systems using the
Hammerstein-Wiener model structure, which lets you interconnect linear
models with static nonlinearities. For more information, see “Identifying
Hammerstein-Wiener Models” on page 4-49.

Nonlinear models might be necessary to represent systems that operate

over a range of operating points. In some cases, you might fit several linear
models, where each model is accurate at specific operating conditions. You
can also try using the nonlinear ARX model structure with tree partitions to
model such systems. For more information, see “Identifying Nonlinear ARX
Models” on page 4-8.

If you know the nonlinear equations describing a system, you can represent
this system as a nonlinear grey-box model and estimate the coefficients from
experimental data. In this case, the coefficients are the parameters of the
model. For more information, see Chapter 5, “ODE Parameter Estimation
(Grey-Box Modeling)”.

4-3
 4 Nonlinear Black-Box Model Identification

Before fitting a nonlinear model, try transforming your input and output
variables such that the relationship between the transformed variables is
linear. For example, you might be dealing with a system that has current and
voltage as inputs to an immersion heater, and the temperature of the heated
liquid as an output. In this case, the output depends on the inputs via the
power of the heater, which is equal to the product of current and voltage.
Instead of fitting a nonlinear model to two-input and one-output data, you can
create a new input variable by taking the product of current and voltage and
then fitting a linear model to the single-input/single-output data.

Linear and Nonlinear Dynamics Are Captured Separately

You might have multiple data sets that capture the linear and nonlinear
dynamics separately. For example, one data set with low amplitude input
(excites the linear dynamics only) and another data set with high amplitude
input (excites the nonlinear dynamics). In such cases, first estimate a linear
model using the first data set. Next, use the model as an initial model to
estimate a nonlinear model using the second data set. For more information,
see “Using Linear Model for Nonlinear ARX Estimation” on page 4-28 and
“Using Linear Model for Hammerstein-Wiener Estimation” on page 4-64.

Available Nonlinear Models

System Identification Toolbox supports these nonlinear models:

• “Nonlinear ARX Models” on page 4-4

• “Hammerstein-Wiener Models” on page 4-5
• “Nonlinear State-Space Models” on page 4-5

Nonlinear ARX Models

Nonlinear ARX models extend the linear ARX models to the nonlinear case
and have this structure:

y(t) = f(y(t - 1), ..., y(t - na), u(t - nk), ..., u(t -nk -nb + 1))

where the function f depends on a finite number of previous inputs u and

outputs y. na is the number of past output terms used to predict the current
output. nb is the number of past input terms used to predict the current

4-4
 About Nonlinear Model Identification

output. nk is the delay from the input to the output, specified as the number
of samples.

Typically, you use nonlinear ARX models as black-box structures. The

nonlinear function of the nonlinear ARX model is a flexible nonlinearity
estimator with parameters that need not have physical significance.

System Identification Toolbox software uses idnlarx objects to represent

nonlinear ARX models. For more information about estimation, see:

• “Tutorial – Identifying Nonlinear Black-Box Models Using the GUI”

• “Identifying Nonlinear ARX Models” on page 4-8

Hammerstein-Wiener Models
Hammerstein-Wiener models describe dynamic systems using one or two
static nonlinear blocks in series with a linear block. The linear block is a
discrete transfer function and represents the dynamic component of the model.

You can use the Hammerstein-Wiener structure to capture physical

nonlinear effects in sensors and actuators that affect the input and output
of a linear system, such as dead zones and saturation. Alternatively, use
Hammerstein-Wiener structures as black box structures that do not represent
physical insight into system processes.

System Identification Toolbox software uses idnlhw objects to represent

Hammerstein-Wiener models. For more information about estimation, see:

• “Tutorial – Identifying Nonlinear Black-Box Models Using the GUI”

• “Identifying Hammerstein-Wiener Models” on page 4-49

Nonlinear State-Space Models

Nonlinear state-space models have this representation:

x ( t ) = F ( x ( t ) , u ( t ) )
y ( t ) = H ( x ( t ) , u ( t ))

4-5
 4 Nonlinear Black-Box Model Identification

where F and H can have any parameterization. You use the idnlgrey object
to specify the structures of such models as nonlinear ODEs, based on physical
insight about your system. The parameters of such models typically have
physical interpretations.

For more information about estimating nonlinear state-space models, see

Chapter 5, “ODE Parameter Estimation (Grey-Box Modeling)”.

4-6
 Preparing Data for Nonlinear Identification

Preparing Data for Nonlinear Identification

Estimating nonlinear ARX and Hammerstein-Wiener models requires
uniformly sampled time-domain data. Your data can have one or more input
and output channels.

For time-series data, you can only fit nonlinear ARX models and nonlinear
state-space models.

Tip Whenever possible, use different data sets for model estimation and
validation.

Before estimating models, import your data into the MATLAB workspace
and do one of the following:

• In the System Identification Tool GUI. Import data into the GUI, as
described in “Importing Data into the GUI” on page 2-14.
• At the command line. Represent your data as an iddata object, as
described in the corresponding reference page.

You can analyze data quality and preprocess data by interpolating missing
values, filtering to emphasize a specific frequency range, or resampling using
a different time interval (see “Ways to Process Data for System Identification”
on page 2-4).

Data detrending can be useful in certain cases, such as before modeling the
relationship between the change in input and the change in output about an
operating point. However, most applications do not require you to remove
offsets and linear trends from the data before nonlinear modeling.

4-7
 4 Nonlinear Black-Box Model Identification

Identifying Nonlinear ARX Models

In this section...
“Nonlinear ARX Model Extends the Linear ARX Structure” on page 4-8
“Structure of Nonlinear ARX Models” on page 4-9
“Nonlinearity Estimators for Nonlinear ARX Models” on page 4-10
“Ways to Configure Nonlinear ARX Estimation” on page 4-12
“How to Estimate Nonlinear ARX Models in the GUI” on page 4-16
“How to Estimate Nonlinear ARX Models at the Command Line” on page
4-19
“Using Linear Model for Nonlinear ARX Estimation” on page 4-28
“Validating Nonlinear ARX Models” on page 4-35
“Using Nonlinear ARX Models” on page 4-40
“Low-Level Manipulation of Nonlinear ARX Components” on page 4-41

Nonlinear ARX Model Extends the Linear ARX

Structure
A linear SISO ARX model has this structure:

y(t) + a1 y(t − 1) + a2 y(t − 2) + ... + ana y(t − na) =

b1u(t) + b2u(t − 1) + ... + bnbu(t − nb − 1) + e ( t )

where the input delay nk is zero to simplify the notation.

This structure implies that the current output y(t) is predicted as a weighted
sum of past output values and current and past input values. Rewriting the
equation as a product:

y p (t) = [ − a1 , − a2 ,..., − ana , b1 , b2 ,.., bnb ] ∗

[ y(t − 1), y(t − 2),..., y(t − na), u(t), u(t − 1),..., u(t − nb − 1)]T

4-8
 Identifying Nonlinear ARX Models

where y(t − 1), y(t − 2),..., y(t − na), u(t), u(t − 1),..., u(t − nb − 1) are delayed input
and output variables, called regressors. The linear ARX model predicts the
current output yp as a weighted sum of its regressors.

The nonlinear ARX structure is an extension of the linear ARX structure:

• Instead of the weighted sum that represents a linear mapping, the

nonlinear ARX model has a more flexible nonlinear mapping function:

y p (t) = f ( y(t − 1), y(t − 2), y(t − 3),..., u(t), u(t − 1), u(t − 2),..)

where f is a nonlinear function. Inputs to f are model regressors. When you

specify the nonlinear ARX model structure, you can choose one of several
available nonlinear mapping functions in this toolbox (see “Nonlinearity
Estimators for Nonlinear ARX Models” on page 4-10).
• Nonlinear ARX regressors can be both delayed input-output variables and
more complex, nonlinear expressions of delayed input and output variables.
Examples of such nonlinear regressors are y(t-1)2, u(t-1)*y(t-2), tan(u(t-1)),
and u(t-1)*y(t-3).

Structure of Nonlinear ARX Models

This block diagram represents the structure of a nonlinear ARX model:

Nonlinearity Estimator

u Nonlinear
Regressors Function y
u(t),u(t-1),y(t-1), ...
Linear
Function

The nonlinear ARX model computes the output y in two stages:

1 Computes regressors from the current and past input values and past
output data.

4-9
 4 Nonlinear Black-Box Model Identification

In the simplest case, regressors are delayed inputs and outputs, such as
u(t-1) and y(t-3)—called standard regressors. You can also specify custom
regressors, which are nonlinear functions of delayed inputs and outputs.
For example, tan(u(t-1)) or u(t-1)*y(t-3).

By default, all regressors are inputs to both the linear and the nonlinear
function blocks of the nonlinearity estimator. You can choose a subset of
regressors as inputs to the nonlinear function block.

2 The nonlinearity estimator block maps the regressors to the model output
using a combination of nonlinear and linear functions. You can select from
available nonlinearity estimators, such as tree-partition networks, wavelet
networks, and multi-layer neural networks. You can also exclude either
the linear or the nonlinear function block from the nonlinearity estimator.

The nonlinearity estimator block can include linear and nonlinear blocks in
parallel. For example:

F ( x) = LT ( x − r) + d + g ( Q( x − r) )

x is a vector of the regressors. LT ( x) + d is the output of the linear function

block and is affine when d≠0. d is a scalar offset. g ( Q( x − r) ) represents the
output of the nonlinear function block. r is the mean of the regressors x. Q is
a projection matrix that makes the calculations well conditioned. The exact
form of F(x) depends on your choice of the nonlinearity estimator.

Estimating a nonlinear ARX model computes the model parameter values,

such as L, r, d, Q, and other parameters specifying g. Resulting models are
idnlarx objects that store all model data, including model regressors and
parameters of the nonlinearity estimator. See the idnlarx reference page
for more information.

Nonlinearity Estimators for Nonlinear ARX Models

System Identification Toolbox software provides several nonlinearity
estimators F(x) for nonlinear ARX models. For more information about F(x),
see “Structure of Nonlinear ARX Models” on page 4-9.

4-10
 Identifying Nonlinear ARX Models

Each nonlinearity estimator corresponds to an object class in this toolbox.

When you estimate nonlinear ARX models in the GUI, System Identification
Toolbox creates and configures objects based on these classes. You can also
create and configure nonlinearity estimators at the command line.

Most nonlinearity estimators represent the nonlinear function as a summed

series of nonlinear units, such as wavelet networks or sigmoid functions. You
can configure the number of nonlinear units n for estimation. For a detailed
description of each estimator, see the references page of the corresponding
nonlinearity class.

Nonlinearity Class Structure Comments

Wavelet wavenet By default,
network the estimation
n
(default) algorithm
g( x) = ∑ α k κ ( βk ( x − γ k ) ) determines the
k=1
number of units n
where κ(s) is the wavelet function. automatically.
One layer sigmoidnet Default number of
sigmoid units n is 10.
n
network
g( x) = ∑ α k κ ( βk ( x − γ k ) )
k=1

( )
−1
where  (s) = es + 1 is the sigmoid
function. βk is a row vector such that
βk ( x − γ k ) is a scalar.

4-11
 4 Nonlinear Black-Box Model Identification

Nonlinearity Class Structure Comments

Tree partition treepartition Piecewise linear function over partitions The estimation
of the regressor space defined by a algorithm
binary tree. determines the
number of units
automatically.
Try using tree
partitions for
modeling data
collected over a
range of operating
conditions.
F is linear in linear This estimator produces a model that Use to specify
x is similar to the linear ARX model, custom regressors
but offers the additional flexibility of as the nonlinearity
specifying custom regressors. estimator and
exclude a
nonlinearity
mapping function.
Multilayered neuralnet Uses as a network object created using
neural the Neural Network Toolbox™ software.
network
Custom customnet Similar to sigmoid network but you (For advanced use)
network specify κ(s) . Uses the unit
(user-defined) function that you
specify.

Ways to Configure Nonlinear ARX Estimation

• “Configurable Elements of Nonlinear ARX Structure” on page 4-13
• “Default Nonlinear ARX Structure” on page 4-14
• “Nonlinear ARX Order and Delay” on page 4-14
• “Estimation Algorithm for Nonlinear ARX Models” on page 4-15

4-12
 Identifying Nonlinear ARX Models

Configurable Elements of Nonlinear ARX Structure

You can adjust various elements of the nonlinear ARX model structure and fit
different models to your data.

Configure model regressors by:

• Specifying model order and delay, which creates the set of standard
regressors.
For a definition, see “Nonlinear ARX Order and Delay” on page 4-14.
• Creating custom regressors.
Custom regressors are arbitrary functions of past inputs and outputs, such
as products, powers, and other MATLAB expressions of input and output
variables. You can specify custom regressors in addition to or instead of
standard regressors for greater flexibility in modeling your data.
• Including a subset of regressors in the nonlinear function of the nonlinear
estimator block.
Selecting which regressors are inputs to the nonlinear function reduces
model complexity and keeps the estimation well-conditioned.
• Initializing using a linear ARX model.
You can perform this operation only at the command line. The initialization
configures the nonlinear ARX model to use standard regressors, which the
toolbox computes using the orders and delays of the linear model. See
“Using Linear Model for Nonlinear ARX Estimation” on page 4-28.

Configure the nonlinearity estimator block by:

• Specifying and configuring the nonlinear function, including the number

of units.
• Excluding the nonlinear function from the nonlinear estimator such that
F(x)= LT ( x) + d .
• Excluding the linear function from the nonlinear estimator such that
F(x)= g ( Q( x − r) ) .

4-13
 4 Nonlinear Black-Box Model Identification

Note You cannot exclude the linear function from tree partitions and
neural networks.

See these topics for detailed steps to change the model structure:

• “Tutorial – Identifying Nonlinear Black-Box Models Using the GUI”

• “How to Estimate Nonlinear ARX Models in the GUI” on page 4-16
• “How to Estimate Nonlinear ARX Models at the Command Line” on page
4-19

Default Nonlinear ARX Structure

Estimate a nonlinear ARX model with default configuration by:

• Specifying only model order and input delay. Specifying the order
automatically creates standard regressors.
• Specifying a linear ARX model. The linear model sets the model orders and
linear function of the nonlinear model. You can perform this operation
only at the command line.

By default:

• The nonlinearity estimator is a wavelet network (see the wavenet reference

page).
This nonlinearity often provides satisfactory results and uses a fast
estimation method.
• All of the standard regressors are inputs to the linear and nonlinear
functions of the wavelet network.

Nonlinear ARX Order and Delay

The order and delay of nonlinear ARX models are positive integers:

• na — Number of past output terms used to predict the current output.

• nb — Number of past input terms used to predict the current output.

4-14
 Identifying Nonlinear ARX Models

• nk — Delay from input to the output in terms of the number of samples.

The meaning of na, nb, and nk is similar to linear ARX model parameters.
Orders are scalars for SISO data, and matrices for MIMO data. If you are not
sure how to specify the order and delay, you can estimate them as described
in “Preliminary Step – Estimating Model Orders and Input Delays” on page
3-48. Such an estimate is based on linear ARX models and only provides
initial guidance—the best orders for a linear ARX model might not be the best
orders for a nonlinear ARX model.

System Identification Toolbox software computes standard regressors using

model orders.

For example, if you specify this order and delay for a SISO model with input
u and output y:

na=2, nb=3, and nk=5

the toolbox computes standard regressors y(t-2), y(t-1), u(t-5), u(t-6), and u(t-7).

You can specify custom regressors in addition to standard regressors, as

described in “How to Estimate Nonlinear ARX Models in the GUI” on page
4-16 and “How to Estimate Nonlinear ARX Models at the Command Line”
on page 4-19.

Estimation Algorithm for Nonlinear ARX Models

The estimation algorithm depends on your choice of nonlinearity estimator
and other properties of the idnlarx class. You can set algorithm properties
both in the GUI and at the command line.

Focus property of idnlarx class. By default, estimating nonlinear ARX

models minimizes one-step prediction errors, which corresponds to Focus
value of Prediction.

If you want a model that is optimized for reproducing simulation behavior,

try setting the Focus value to Simulation. In this case, you cannot use
treepartition and neuralnet because these nonlinearity estimators are
not differentiable. Minimization of simulation error requires differentiable

4-15
 4 Nonlinear Black-Box Model Identification

nonlinear functions. Simulation error minimization takes more time than

one-step-ahead prediction error minimization.

Common algorithm properties of idnlarx class.

• MaxIter — Maximum number of iterations.

• SearchMethod — Search method for minimization of prediction or
simulation errors, such as Gauss-Newton and Levenburg-Marquardt line
search, and Trust-region reflective Newton approach. By default, the
algorithm uses a combination of these methods.
• Tolerance — Condition for terminating iterative search when the expected
improvement of the parameter values is less than a specified value.
• Display — Shows progress of iterative minimization in the MATLAB
Command Window.

How to Estimate Nonlinear ARX Models in the GUI

Prerequisites

• Learn about the nonlinear ARX model structure (see “Structure of

Nonlinear ARX Models” on page 4-9).
• Import data into the System Identification Tool GUI (see “Preparing Data
for Nonlinear Identification” on page 4-7).
• (Optional) Choose a nonlinearity estimator in “Nonlinearity Estimators for
Nonlinear ARX Models” on page 4-10.

1 In the System Identification Tool GUI, select Estimate > Nonlinear

models to open the Nonlinear Models dialog box.

2 In the Configure tab, select Nonlinear ARX from the Model type list (if
it is not already selected).

3 (Optional) Edit the Model name by clicking the pencil icon. The name of
the model should be unique to all nonlinear ARX models in the System
Identification Tool GUI.

4-16
 Identifying Nonlinear ARX Models

4 (Optional) If you want to refine a previously estimated model, click

Initialize to select a previously estimated model from the Initial Model
list.

Note Refining a previously estimated model starts with the parameter

values of the initial model and uses the same model structure. The
algorithm uses default estimation settings unless you specify to use the
initial model settings, or change these settings.

The Initial Model list includes models that:

• Exist in the System Identification Tool GUI.
• Have the same number of inputs and outputs as the dimensions of the
estimation data (selected as Working Data in the System Identification
Tool GUI).

5 Keep the default settings in the Nonlinear Models dialog box that specify
the model structure and the algorithm, or modify these settings:

Note For more information about available options, click Help in the
Nonlinear Models dialog box to open the GUI help.

What to Configure Options in Comment

Nonlinear Models
GUI
Model order In the Regressors Model order na is the
tab, edit the No. of output No. of Terms,
Terms corresponding and nb is the input No.
to each input and of Terms.
output channel.
Input delay In the Regressors If you do not know the
tab, edit the Delay input delay value, click
corresponding to an Infer Input Delay.
input channel. This action opens the
Infer Input Delay

4-17
 4 Nonlinear Black-Box Model Identification

What to Configure Options in Comment

Nonlinear Models
GUI

dialog box to suggest

possible delay values.
Regressors In the Regressors This action opens
tab, click Edit the Model Regressors
Regressors. dialog box. Use this
dialog box to create
custom regressors
or to include specific
regressors in the
nonlinear block.
Nonlinearity estimator In the Model To use all standard
Properties tab. and custom regressors
in the linear block
only, you can exclude
the nonlinear block by
setting Nonlinearity
to None.
Estimation algorithm In the Estimate
tab, click Algorithm
Options.

6 Click Estimate to add this model to the System Identification Tool GUI.

The Estimate tab displays the estimation progress and results.

7 Validate the model response by selecting the desired plot in the Model
Views area of the System Identification Tool GUI. For more information
about validating models, see “How to Plot Nonlinear ARX Plots Using the
GUI” on page 4-35.

If you get a poor fit, try changing the model structure or algorithm
configuration in step 5.

You can export the model to the MATLAB workspace by dragging it to To

Workspace in the System Identification Tool GUI.

4-18
 Identifying Nonlinear ARX Models

How to Estimate Nonlinear ARX Models at the

Command Line
Prerequisites

• Learn about the nonlinear ARX model structure in “Structure of Nonlinear

ARX Models” on page 4-9.
• Prepare your data, as described in “Preparing Data for Nonlinear
Identification” on page 4-7.
• (Optional) Estimate model orders and delays the same way you would for
linear ARX models. See “Preliminary Step – Estimating Model Orders and
Input Delays” on page 3-48.
• (Optional) Choose a nonlinearity estimator in “Nonlinearity Estimators for
Nonlinear ARX Models” on page 4-10.
• (Optional) Estimate or construct an linear ARX model for initialization
of nonlinear ARX model. See “Using Linear Model for Nonlinear ARX
Estimation” on page 4-28.

Estimate model using nlarx.

Use nlarx to both construct and estimate a nonlinear ARX model. After
each estimation, validate the model by comparing it to other models and
simulating or predicting the model response.

Basic Estimation

Start with the simplest estimation using m = nlarx(data,[na nb nk]). For

example:

m = nlarx(data,[2 2 1]) % na=nb=2 and nk=1

By default, the nonlinearity estimator is the wavelet network (see the wavenet
reference page), which takes all standard regressors as inputs to its linear
and nonlinear functions. m is an idnlarx object.

For MIMO systems, nb, nf, and nk are ny-by-nu matrices. See the nlarx
reference page for more information about MIMO estimation.

4-19
 4 Nonlinear Black-Box Model Identification

Specify a different nonlinearity estimator (for example, sigmoid network):

M = nlarx(data,[2 2 1],'sigmoid')

Set the Focus property of the idnlarx object estimation to simulation error
minimization:

M = nlarx(data,[2 2 1],'sigmoid','Focus','simulation')

Configure model regressors.

Standard Regressors

Change the model order to create a model structure with different model
regressors, which are delayed input and output variables that are inputs to
the nonlinearity estimator.

Custom Regressors

Explore including custom regressors in the nonlinear ARX model structure.

Custom regressors are in addition to the standard model regressors (see
“Nonlinear ARX Order and Delay” on page 4-14).

Use polyreg or customreg to construct custom regressors in terms of

model input-output variables. You can specify custom regressors using the
CustomRegressors property of the idnlarx class or addreg to append custom
regressors to an existing model.

For example, generate regressors as polynomial functions of inputs and

outputs:

load iddata1
m = nlarx(z1,[2 2 1],'sigmoidnet');
getreg(m) % displays all regressors
% Generate polynomial regressors up to order 2:
reg = polyreg(m)
% Append polynomial regressors to CustomRegressors:
m = addreg(m,reg);
getreg(m) % now includes polynomial regressors

4-20
 Identifying Nonlinear ARX Models

You can also specify arbitrary functions of input and output variables. For
example:

load iddata1
m = nlarx(z1,[2 2 1],'sigmoidnet',...
'CustomReg',{'y1(t-1)^2','y1(t-2)*u1(t-3)'});
getreg(m) % displays all regressors
% Append polynomial regressors to CustomRegressors:
m = addreg(m,reg);
getreg(m) % polynomial regressors

Manipulate custom regressors using the CustomRegressors property of the

idnlarx class. For example, to get the function handle of the first custom
regressor in the array:

CReg1 = m.CustomReg(1).Function;

To view the regressor expression as a string, use:

m.CustomReg(1).Display

You can exclude all standard regressors and use only custom regressors in the
model structure by setting na=nb=nk=0:

m = nlarx(data,[0 0 0],'CustomReg',{'y1(t-1)^2','y1(t-2)*u1(t-3)'})

In advanced applications, you can specify advanced estimation options for

nonlinearity estimators. For example, wavenet and treepartition classes
provide the Options property for setting such estimation options.

Linear and nonlinear regressors.

By default, all model regressors enter as inputs to both linear and nonlinear
function blocks of the nonlinearity estimator. To reduce model complexity and
keep the estimation well-conditioned, use a subset of regressors as inputs to
the nonlinear function of the nonlinear estimator block.

For example, specify a nonlinear ARX model to be linear in past outputs

and nonlinear in past inputs:

m = nlarx(data,[2 2 1]) % all standard regressors are

% inputs to the nonlinear function

4-21
 4 Nonlinear Black-Box Model Identification

getreg(m) % lists all standard regressors

m = nlarx(data,[4 4 1],sigmoidnet,'nlreg',[5 6 7 8])

This example uses getreg to determine the index of each regressor from
the complete list of all model regressors. Only regressor numbers 5 through
8 are inputs to the nonlinear function—getreg shows that these regressors
are functions of the input variable u1. nlreg is an abbreviation for the
NonlinearRegressors property of the idnlarx class.

Alternatively, include only input regressors in the nonlinear function block

using:

m = nlarx(data,[4 4 1],sigmoidnet,'nlreg','input')

When you are not sure which regressors to include as inputs to the nonlinear
function block, specify to search during estimation for the optimum regressor
combination:

m = nlarx(data,[4 4 1],sigmoidnet,'nlreg','search')

After estimation, use m.NonlinearRegressors to view which regressors were

selected by the automatic regressor search. This search typically takes a long
time, and you can display the search progress using:

m = nlarx(data,[4 4 1],sigmoidnet,'nlreg','search',...
'Display', 'on')

Configure the nonlinearity estimator.

Specify the nonlinearity estimator directly in the estimation command as:

• A string of the nonlinearity name, which uses the default nonlinearity

configuration.

m = nlarx(data, [2 2 1],'sigmoidnet')

or

m = nlarx(data,[2 2 1],'sig') % abbreviated string

• Nonlinearity object.

4-22
 Identifying Nonlinear ARX Models

m = nlarx(data,[2 2 1],wavenet('num',5))

This estimation uses a nonlinear ARX model with a wavelet nonlinearity

that has 5 units.

To construct the nonlinearity object before providing it as an input to the

nonlinearity estimator:

w = wavenet('num', 5);
m = nlarx(data,[2 2 1],w)

or

w = wavenet;
w.NumberOfUnits = 5;
m = nlarx(data,[2 2 1],w)

For MIMO systems, you can specify a different nonlinearity for each output.
For example, to specify sigmoidnet for the first output and wavenet for the
second output:

M = nlarx(data,[na nb nk],[sigmoidnet; wavenet])

If you want the same nonlinearity for all output channels, specify one
nonlinearity.

This table summarizes values that specify nonlinearity estimators.

Nonlinearity Value (Default Nonlinearity Class

Configuration)
Wavelet network 'wavenet' or 'wave' wavenet
(default)
One layer sigmoid 'sigmoidnet' or 'sigm' sigmoidnet
network
Tree partition 'treepartition' or 'tree' treepartition
F is linear in x 'linear' or [ ] linear

Additional available nonlinearities include multilayered neural networks

and custom networks that you create.

4-23
 4 Nonlinear Black-Box Model Identification

Specify a multilayered neural network using:

m = nlarx(data,[na nb nk],NNet)

where NNet is the neural network object you create using the Neural Network
Toolbox software. See the neuralnet reference page.

Specify a custom network by defining a function called gaussunit.m, as

described in the customnet reference page. Define the custom network object
CNetw and estimate the model:

CNetw = cutomnet(@gaussunit);
m = nlarx(data,[na nb nk],CNetw)

Include only nonlinear function in nonlinearity estimator.

If your model includes wavenet, sigmoidnet, and customnet nonlinearity

estimators, you can exclude the linear function using the LinearTerm
property of the nonlinearity estimator. The nonlinearity estimator becomes
F(x)= g ( Q( x − r) ) .

Nonlinearity Estimator

u
Regressors Nonlinear y
u(t),u(t-1),y(t-1), ... Function

For example:

SNL = sigmoidnet('LinearTerm','off'')
m = nlarx(data,[2 2 1],SNL);

Note You cannot exclude the linear function from tree partition and neural
network nonlinearity estimators.

4-24
 Identifying Nonlinear ARX Models

Include only linear function in nonlinearity estimator.

Configure the nonlinear ARX structure to include only the linear function in
the nonlinearity estimator by setting the nonlinearity to linear. In this case,
F(x)= LT ( x) + d is a weighted sum of model regressors plus an offset. Such
models provide a bridge between purely linear ARX models and fully flexible
nonlinear models.

Nonlinearity Estimator

u
Regressors Linear y
u(t),u(t-1),y(t-1), ... Function

In the simplest case, a model with only standard regressors is linear (affine).
For example, this structure:

m = nlarx(data,[na nb nk],'linear');

is similar to the linear ARX model:

lin_m = arx(data,[na nb nk]);

However, the nonlinear ARX model m is more flexible than the linear ARX
model lin_m because it contains the offset term, d. This offset term provides
the additional flexibility of capturing signal offsets, which is not available
in linear models.

A popular nonlinear ARX configuration in many applications uses polynomial

regressors to model system nonlinearities. In such cases, the system is
considered to be a linear combination of products of (delayed) input and
output variables. Use the polyreg command to easily generate combinations
of regressor products and powers.

For example, suppose that you know the output y(t) of a system to be a linear
combination of (y(t − 1))2 and y(t − 2)*u(t − 3). To model such as system, use:

4-25
 4 Nonlinear Black-Box Model Identification

M = nlarx(data,[0 0 0],'linear',...
'CustomReg',{'y1(t-1)^2','y1(t-2)*u1(t-3)'})

M has no standard regressors and the nonlinearity in the model is described

only by the custom regressors.

Iteratively refine the model.

If your model structure includes nonlinearities that support iterative search

(see “Estimation Algorithm for Nonlinear ARX Models” on page 4-15), you can
use pem to refine model parameters:

m = nlarx(data,[2 2 1],'sigmoidnet')
m2 = pem(data,m)

You can also use nlarx to refine the original model:

m1 = nlarx(data, [2 2 1],'sigmoidnet','wavenet');
m2 = nlarx(data,m1) % can repeatedly run this command

Check the search termination criterion m.EstimationInfo.WhyStop. If

WhyStop indicates that the estimation reached the maximum number of
iterations, try repeating the estimation and possibly specifying a larger value
for the MaxIter idnlarx property:

m2 = pem(data,m1,'MaxIter',30) % runs 30 more iterations

% starting from m1

When the m.EstimationInfo.WhyStop value is Near (local) minimum,

(norm( g) < tol or No improvement along the search direction
with line search, validate your model to see if this model adequately fits
the data. If not, the solution might be stuck in a local minimum of the
cost-function surface. Try adjusting the Algorithm.Tolerance property value
of the idnlarx class or the Algorithm.SearchMethod property, and repeat
the estimation. You can also try perturbing the parameters of the last model
using init (called randomization) and refining the model using pem:

M1 = nlarx(data, [2 2 1], `sigm'); % original model

M1p = init(M1); % randomly perturbs parameters about nominal values
M2 = pem(data, M1p); % estimates parameters of perturbed model

4-26
 Identifying Nonlinear ARX Models

You can display the progress of the iterative search in the MATLAB Command
Window using the Display property of the idnlarx class:

M2= pem(data,M1p,'Display','On')

What if you cannot get a satisfactory model?

If you do not get a satisfactory model after many trials with various model
structures and algorithm settings, it is possible that the data is poor. For
example, your data might be missing important input or output variables and
does not sufficiently cover all the operating points of the system.

Nonlinear black-box system identification usually requires more data than

linear model identification to gain enough information about the system.

Example – Using nlarx to Estimate Nonlinear ARX Models

Use nlarx to estimate a nonlinear ARX model for the data in the “Tutorial –
Identifying Nonlinear Black-Box Models Using the GUI”.

1 Prepare the data for estimation:

load twotankdata
z = iddata(y, u, 0.2);
ze = z(1:1000); zv = z(1001:3000);

2 Estimate several models using different model orders, delays, and

nonlinearity settings:

m1 = nlarx(ze,[2 2 1]);
m2 = nlarx(ze,[2 2 3]);
m3 = nlarx(ze,[2 2 3],wavenet('num',8));
m4 = nlarx(ze,[2 2 3],wavenet('num',8),...
'nlr', [1 2]);

An alternative way to perform the estimation is to configure the model

structure first, and then to estimate this model:

m5 = idnlarx([2 2 3],sigmoidnet('num',14),'nlr',[1 2])

m5 = pem(ze,m5);

4-27
 4 Nonlinear Black-Box Model Identification

3 Compare the resulting models by plotting the model outputs with the
measured output:

compare(zv, m1,m2,m3,m4,m5)

The following examples

Using Linear Model for Nonlinear ARX Estimation

• “About Using Linear Models” on page 4-29
• “How to Initialize Nonlinear ARX Estimation Using Linear ARX Models”
on page 4-30
• “Example – Using Linear ARX Models to Estimate Nonlinear ARX Models”
on page 4-30

4-28
 Identifying Nonlinear ARX Models

About Using Linear Models

You can use the following discrete-time linear models for nonlinear ARX
estimation. The linear model must sufficiently represent the linear dynamics
of your system.

Discrete-Time Linear Model Use for Initializing...

Single-output polynomial model of Single-output nonlinear ARX model
ARX structure (idpoly) estimation
Multi-output polynomial model of Multi-output nonlinear ARX model
ARX structure (idarx) estimation

Tip To learn more about when to use linear models, see “When to Fit
Nonlinear Models” on page 4-2.

Typically, you create a linear ARX model using the arx command. You can
provide the linear model only at the command line when constructing (see
idnlarx) or estimating (see nlarx) a nonlinear ARX model.

The software uses the linear model for initializing the nonlinear ARX
estimation:

• Assigns the linear model orders as initial values of nonlinear model orders
(na and nb properties of the idnlarx object) and delays (nk property) to
compute standard regressors in the nonlinear ARX model structure.
• Uses the A and B polynomials of the linear model to compute the linear
function of the nonlinearity estimators (LinearCoef parameter of the
nonlinearity estimator object), except for neural network nonlinearity
estimator.

During estimation, the estimation algorithm uses these values to further

adjust the nonlinear model to the data. The initialization always provides a
better fit to the estimation data than the linear ARX model.

4-29
 4 Nonlinear Black-Box Model Identification

How to Initialize Nonlinear ARX Estimation Using Linear ARX

Models
Estimate a nonlinear ARX model initialized using a linear model by typing

m = nlarx(data,LinARXModel)

LinARXModel is an idpoly object of ARX structure for single-output models,

and idarx object for multi-output models. m is an idnlarx object. data is a
time-domain iddata object.

By default, the nonlinearity estimator is the wavelet network (wavenet

object). This network takes all standard regressors computed using orders
and delay of LinARXModel as inputs to its linear and nonlinear functions. The
software computes the LinearCoef parameter of the wavenet object using the
A and B polynomials of the linear ARX model.

Tip When you use the same data set, a nonlinear ARX model initialized using
a linear ARX model produces a better fit than the linear ARX model.

Specify a different nonlinearity estimator, for example a sigmoid network:

m = nlarx(data,LinARXModel,'sigmoid')

Set the Focus property of the idnlarx object estimation to simulation error
minimization:

m = nlarx(data,LinARXModel,'sigmoid','Focus','simulation')

After each estimation, validate the model by comparing the simulated

response to the data. To improve the fit of the nonlinear ARX model, adjust
various elements of the nonlinear ARX structure. For more information, see
“Ways to Configure Nonlinear ARX Estimation” on page 4-12.

Example – Using Linear ARX Models to Estimate Nonlinear

ARX Models

1 Load the estimation data.

4-30
 Identifying Nonlinear ARX Models

load throttledata.mat

This command loads the data object ThrottleData into the MATLAB
workspace. The object contains input and output samples collected from an
engine throttle system, sampled at a rate of 100 Hz.

A DC motor controls the opening angle of the butterfly valve in the throttle
system. A step signal (in volts) drives the DC motor. The output is the
angular position (in degrees) of the valve.

2 Plot the data to view and analyze the data characteristics.

plot(ThrottleData)

In the normal operating range of 15 90 degrees, the input and output

variables have a linear relationship, as shown in the following figure. You
use a linear model of low order to model this relationship.

4-31
 4 Nonlinear Black-Box Model Identification

In the throttle system, a hard stop limits the valve position to 90 degrees,
and a spring brings the valve to 15 degrees when the DC motor is turned
off. These physical components introduce nonlinearities that a linear model
cannot capture.

3 Estimate an ARX model to model the linear behavior of this single-input

single-output system in the normal operating range.

% Detrend the data because linear models cannot capture offsets.

Tr = getTrend(ThrottleData);
Tr.OutputOffset = 15;
DetrendedData = detrend(ThrottleData, Tr);

4-32
 Identifying Nonlinear ARX Models

% Estimate a linear ARX model with na=2, nb=1, nk=1.

LinearModel = arx(DetrendedData, [2 1 1], 'Focus', 'Simulation');

4 Compare the simulated model response with estimation data.

compare(DetrendedData, LinearModel)

The linear model captures the rising and settling behavior in the linear
operating range but does not account for output saturation at 90 degrees,
as shown in the next figure.

5 Estimate a nonlinear ARX model to model the output saturation.

4-33
 4 Nonlinear Black-Box Model Identification

NonlinearModel = nlarx(ThrottleData, LinearModel, 'sigmoidnet',...

'Focus', 'Simulation');

The software uses the orders and delay of the linear model for the orders of
the nonlinear model. In addition, the software computes the linear function
of sigmoidnet nonlinearity estimator.

6 Compare the nonlinear model with data.

compare(ThrottleData, NonlinearModel)

The model captures the nonlinear effects (output saturation) and improves
the overall fit to data, as shown in the next figure.

4-34
 Identifying Nonlinear ARX Models

Validating Nonlinear ARX Models

• “About Nonlinear ARX Plots” on page 4-35
• “How to Plot Nonlinear ARX Plots Using the GUI” on page 4-35
• “How to Validate Nonlinear ARX Models at the Command Line” on page
4-36
• “Configuring the Nonlinear ARX Plot” on page 4-38
• “Axis Limits, Legend, and 3-D Rotation” on page 4-39

About Nonlinear ARX Plots

The Nonlinear ARX plot displays the characteristics of model nonlinearities
as a function of one or two regressors. For more information about estimating
nonlinear ARX models, see “Identifying Nonlinear ARX Models” on page 4-8.

Examining a nonlinear ARX plot can help you gain insight into which
regressors have the strongest effect on the model output. Understanding the
relative importance of the regressors on the output can help you decide which
regressors should be included in the nonlinear function.

Furthermore, you can create several nonlinear models for the same data
set using different nonlinearity estimators, such a wavelet network and
tree partition, and then compare the nonlinear surfaces of these models.
Agreement between nonlinear surfaces increases the confidence that these
nonlinear models capture the true dynamics of the system.

How to Plot Nonlinear ARX Plots Using the GUI

You can plot linear and nonlinear blocks of nonlinear ARX models.

To create a nonlinear ARX plot in the System Identification Tool GUI, select
the Nonlinear ARX check box in the Model Views area. For general
information about creating and working with plots, see “Working with Plots
in the System Identification Tool GUI” on page 11-15.

4-35
 4 Nonlinear Black-Box Model Identification

Note The Nonlinear ARX check box is unavailable if you do not have a
nonlinear ARX model in the Model Board.

The following figure shows a sample nonlinear ARX plot.

How to Validate Nonlinear ARX Models at the Command Line

You can use the following approaches to validate nonlinear ARX models at
the command line:

Compare Model Output to Measured Output

Compare estimated models using compare. Use an independent validation

data set whenever possible. For more information about validating models,
see “Overview of Model Validation and Plots” on page 8-2.

4-36
 Identifying Nonlinear ARX Models

For example, compare linear and nonlinear ARX models of same order:

load iddata1
LM = arx(z1,[2 2 1]) % estimates linear ARX model
M = nlarx(z1,[2 2 1],'sigmoidnet') % estimates nonlinear ARX model
compare(z1,LM,M) % compares responses of LM and M
% against measured data

Compare the performance of several models using the

properties M.EstimationInfo.FPE (final prediction error) and
M.EstimationInfo.LossFcn (value of loss function at estimation
termination). Smaller values typically indicate better performance. However,
m.EstimationInfo.FPE values might be unreliable when the model contains
a large number of parameters relative to the estimation data size.

Simulate and Predict Model Response

Use sim(idnlarx) and predict(idnlarx) to simulate and predict model

response, respectively. To compute the step response of the model, use step.
See the corresponding reference page for more information.

Analyze Residuals

Residuals are differences between the one-step-ahead predicted output from

the model and the measured output from the validation data set. Thus,
residuals represent the portion of the validation data output not explained by
the model. Use resid to compute and plot the residuals.

Plot Nonlinearity

Use plot to view the shape of the nonlinearity. For example:

plot(M)

where M is the nonlinear ARX (idnlarx) model. The plot command opens the
Nonlinear ARX Model Plot window. For more information about working with
this plot window, see “Configuring the Nonlinear ARX Plot” on page 4-38 and
“Axis Limits, Legend, and 3-D Rotation” on page 4-39.

4-37
 4 Nonlinear Black-Box Model Identification

If the shape of the plot looks like a plane for all the chosen regressor values,
then the model is probably linear in those regressors. In this case, you can
remove the corresponding regressors from nonlinear block by specifying the
M.NonlinearRegressors property and repeat the estimation.

You can use additional plot arguments to specify the following information:

• Include multiple nonlinear ARX models on the plot.

• Configure the regressor values for computing the nonlinearity values.

For detailed information about plot, type the following command at the
prompt:

help idnlarx/plot

Check Iterative Search Termination Conditions

If your idnlarx model structure uses iterative search to minimize prediction

or simulation errors, use M.EstimationInfo to display the estimation
termination conditions. M is the estimated idnlarx model. For example,
check the WhyStop field of the EstimationInfo property, which describes why
the estimation stopped—the algorithm might have reached the maximum
number of iterations or the required tolerance value. For more information
about iterative search, see “Estimation Algorithm for Nonlinear ARX Models”
on page 4-15.

Configuring the Nonlinear ARX Plot

To include or exclude a model on the plot, click the corresponding model icon
in the System Identification Tool GUI. Active models display a thick line
inside the Model Board icon.

To configure the plot:

1 If your model contains multiple output, select the output channel in the
Select nonlinearity at output list. Selecting the output channel displays
the nonlinearity values that correspond to this output channel.

2 If the regressor selection options are not visible, click to expand the
Nonlinear ARX Model Plot window.

4-38
 Identifying Nonlinear ARX Models

3 Select Regressor 1 from the list of available regressors. In the Range

field, enter the range of values to include on the plot for this regressor. The
regressor values are plotted on the Reg1 axis.

4 Specify a second regressor for a 3-D plot by selecting one of the following
types of options:
• Select Regressor 2 to display three axes. In the Range field, enter the
range of values to include on the plot for this regressor. The regressor
values are plotted on the Reg2 axis.
• Select <none> in the Regressor 2 list to display only two axes.

5 To fix the values of the regressor that are not displayed, click Fix Values.
In the Fix Regressor Values dialog box, double-click the Value cell to edit
the constant value of the corresponding regressor. The default values are
determined during model estimation. Click OK.

6 In the Nonlinear ARX Model Plot window, click Apply to update the plot.

7 To change the grid of the regressor space along each axis, Options > Set
number of samples, and enter the number of samples to use for each
regressor. Click Apply and then Close.

For example, if the number of samples is 20, each regressor variable

contains 20 points in its specified range. For a 3-D plots, this results in
evaluating the nonlinearity at 20 x 20 = 400 points.

Axis Limits, Legend, and 3-D Rotation

The following table summarizes the commands to modify the appearance
of the Nonlinear ARX plot.

4-39
 4 Nonlinear Black-Box Model Identification

Changing Appearance of the Nonlinear ARX Plot

Action Command
Change axis limits. Select Options > Set axis limits to
open the Axis Limits dialog box, and
edit the limits. Click Apply.
Hide or show the legend. Select Style > Legend. Select this
option again to show the legend.
(Three axes only) Select Style > 3D Rotate and
Rotate in three dimensions. drag the axes on the plot to
a new orientation. To disable
three-dimensional rotation, select
Note Available only when you Style > 3D Rotate again.
have selected two regressors as
independent variables.

Using Nonlinear ARX Models

Simulation and Prediction
Use sim(idnlarx) to simulate the model output, and predict(idnlarx)
to predict the model output. To compare models to measured output and
to each other, use compare.

Simulation and prediction commands provide default handling of the model’s

initial conditions, or initial state values. See the idnlarx reference page for a
definition of the nonlinear ARX model states.

This toolbox provides several options to facilitate how you specify initial states.
For example, you can use findstates(idnlarx) and data2state(idnlarx)
to compute state values based on operating conditions or the requirement
to maximize fit to measured output.

To learn more about how sim and predict compute the model output, see
“Low-Level Manipulation of Nonlinear ARX Components” on page 4-41.

4-40
 Identifying Nonlinear ARX Models

Linearization

Compute linear approximation of nonlinear ARX models using

linearize(idnlarx) or linapp.

linearize provides a first-order Taylor series approximation of the system

about an operation point (also called tangent linearization). linapp computes
a linear approximation of a nonlinear model for a given input data. For more
information, see the “Linear Approximation of Nonlinear Black-Box Models”
on page 4-81.

You can compute the operating point for linearization using findop(idnlarx).

After computing a linear approximation of a nonlinear model, you can perform

linear analysis and control design on your model using Control System
Toolbox commands. For more information, see Chapter 9, “Control Design
Applications”.

Simulation and Code Generation Using Simulink

You can import estimated Nonlinear ARX models into Simulink software
using the Nonlinear ARX block from the System Identification Toolbox block
library. Import the idnlarx object from the workspace into Simulink using
this block to simulate the model output.

The Nonlinear ARX block supports code generation with Real-Time

Workshop® software, using both generic and embedded targets. Code
generation does not work when the model contains customnet or neuralnet
nonlinearity estimator, or custom regressors.

For more information about working in the Simulink environment, see

Chapter 10, “System Identification Toolbox Blocks”.

Low-Level Manipulation of Nonlinear ARX

Components
In most applications, sim(idnlarx) and predict(idnlarx) are sufficient
for computing the simulated and predicted model response, respectively.
This advanced topic describes how the software evaluates the output of
nonlinearity estimators and uses this output to compute the model response.

4-41
 4 Nonlinear Black-Box Model Identification

Evaluating Nonlinearities
Evaluating the predicted output of a nonlinearity for a specific regressor
value x requires that you first extract the nonlinearity F and regressors
from the model:

F = get(m,'Nonlinearity') % equivalent to F = m.nl

x = getreg(m,'all',data) % computes regressors

Evaluate F(x):

y = evaluate(F,x)

where x is a row vector of regressor values.

You can also evaluate predicted output values at multiple time instants by
evaluating F for several regressor vectors simultaneously:

y = evaluate(F,[x1;x2;x3])

Example – Low-level Simulation and Prediction of Sigmoid

Network
This example shows how the software computes the simulated and predicted
output of the model as a result of evaluating the output of its nonlinearity
estimator for given regressor values.

Estimating and Exploring a Nonlinear ARX Model

1 Estimate nonlinear ARX model with sigmoid network nonlinearity:

load twotankdata
estData = iddata(y,u,0.2,'Tstart',0);
M = nlarx(estData,[1 1 0],'sig');

2 Inspect the model properties and estimation result:

present(M)

which provides information about input and output variables, regressors,

and nonlinearity estimator:

Input name: u1

4-42
 Identifying Nonlinear ARX Models

Output name: y1
Standard regressors corresponding to the orders
na = 1, nb = 1, nk = 0
No custom regressor
Nonlinear regressors:
y1(t-1)
u1(t)
Nonlinearity estimator: sigmoidnet with 10 units

3 Inspect the nonlinearity estimator:

NL = M.Nonlinearity % equivalent to M.nl

class(NL) % nonlinearity class
display(NL) % equivalent to NL

Inspect the sigmoid network parameter values:

NL.Parameters

Prediction of Output

The model output is:

y1(t)=f(y1(t-1),u1(t))

where f is the sigmoid network function. The model regressors y1(t-1) and
u1(t) are inputs to the nonlinearity estimator. Time t is a discrete variable
representing kT, where k = 0, 1,…, and T is the sampling interval. In this
example, T=0.2 second.

The output prediction equation is:

yp(t)=f(y1_meas(t-1),u1_meas(t))

where yp(t) is the predicted value of the response at time t. y1_meas(t-1)

and u1_meas(t) are the measured output and input values at times t-1 and t,
respectively.

Computing the predicted response includes:

• Computing regressor values from input-output data.

4-43
 4 Nonlinear Black-Box Model Identification

• Evaluating the nonlinearity for given regressor values.

To compute the predicted value of the response using initial conditions and
current input:

1 Estimate model from data and get nonlinearity parameters:

load twotankdata
estData = iddata(y,u,0.2,'Tstart',0);
M = nlarx(estData,[1 1 0],'sig');
NL = M.Nonlinearity;

2 Specify zero initial states:

x0 = 0;

The model has one state because there is only one delayed term y1(t-1).
The number of states is equal to sum(getDelayInfo(M)).

3 Compute the predicted output at time t=0.

RegValue = [0,estData.u(1)] % input to nonlinear function f

yp_0 = evaluate(NL,RegValue)

RegValue is the vector of regressors at t=0. The predicted output is

yp(t=0)=f(y1_meas(t=-1),u1_meas(t=0)). In terms of MATLAB variables,
this output is f(0,estData.u(1)), where
• y1_meas(t=-1) is the initial state x0 (=0).
• u1_meas(t=0) is the value of the input at t=0, which is the first input
data sample estData.u(1).

4 Compute the predicted output at time t=1.

RegValue = [estData.y(1),estData.u(2)];
yp_1 = evaluate(NL,RegValue)

The predicted output is yp(t=1)=f(y1_meas(t=0),u1_meas(t=1)). In terms of

MATLAB variables, this output is f(estData.y(1),estData.u(2)), where
• y1_meas(t=0) is the measured output value at t=0, which is to
estData.y(1).

4-44
 Identifying Nonlinear ARX Models

• u1_meas(t=1) is the second input data sample estData.u(2).

5 Perform one-step-ahead prediction at all time values for which data is

available.

RegMat = getreg(M,[],estData,x0);
yp = evaluate(NL,RegMat)

This code obtains a matrix of regressors RegMat for all the time samples
using getreg. RegMat has as many rows as there are time samples, and as
many columns as there are regressors in the model—two, in this example.

These steps are equivalent to the predicted response computed in a single

step using predict(idnlarx):

yp = predict(M,estData,1,'InitialState',x0)

Simulation of Output

The model output is:

y1(t)=f(y1(t-1),u1(t))

where f is the sigmoid network function. The model regressors y1(t-1) and
u1(t) are inputs to the nonlinearity estimator. Time t is a discrete variable
representing kT, where k = 0, 1,…, and T is the sampling interval. In this
example, T=0.2 second.

The simulated output is:

ys(t)=f(ys(t-1),u1_meas(t))

where ys(t) is the simulated value of the response at time t. The simulation
equation is the same as the prediction equation, except that the past output
value ys(t-1) results from the simulation at the previous time step, rather
than the measured output value.

Computing the simulated response includes:

• Computing regressor values from input-output data using simulated

output values.

4-45
 4 Nonlinear Black-Box Model Identification

• Evaluating the nonlinearity for given regressor values.

To compute the simulated value of the response using initial conditions and
current input:

1 Estimate model from data and get nonlinearity parameters:

load twotankdata
estData = iddata(y,u,0.2,'Tstart',0);
M = nlarx(estData,[1 1 0],'sig');
NL = M.Nonlinearity

2 Specify zero initial states:

x0 = 0;

The model has one state because there is only one delayed term y1(t-1).
The number of states is equal to sum(getDelayInfo(M)).

3 Compute the simulated output at time t=0, ys(t=0).

RegValue = [0,estData.u(1)]
ys_0 = evaluate(NL,RegValue)

RegValue is the vector of regressors at t=0. ys(t=0)=f(y1(t=-1),u1_meas(t=0)).

In terms of MATLAB variables, this output is f(0,estData.u(1)), where
• y1(t=-1) is the initial state x0 (=0).
• u1_meas(t=0) is the value of the input at t=0, which is the first input
data sample estData.u(1).

4 Compute the simulated output at time t=1, ys(t=1).

RegValue = [ys_0,estData.u(2)];
ys_1 = evaluate(NL,RegValue)

The simulated output ys(t=1)=f(ys(t=0),u1_meas(t=1)). In terms of

MATLAB variables, this output is f(ys_0,estData.u(2)), where
• ys(t=0) is the simulated value of the output at t=0.
• u1_meas(t=1) is the second input data sample estData.u(2).

4-46
 Identifying Nonlinear ARX Models

5 Compute the simulated output at time t=2:

RegValue = [ys_1,estData.u(3)];
ys_2 = evaluate(NL,RegValue)

Unlike for output prediction, you cannot use getreg to compute regressor
values for all time values. You must compute regressors values at each time
sample separately because the output samples required for forming the
regressor vector are available iteratively, one sample at a time.

These steps are equivalent to the simulated response computed in a single

step using sim(idnlarx):

ys = sim(M,estData,x0)

Low-Level Nonlinearity Evaluation

This examples performs a low-level computation of the nonlinearity response

for the sigmoidnet network function:

F ( x) = ( x − r) PL + a1 f ( ( x − r ) Qb1 + c1 ) + …
+ an f ( ( x − r ) Qbn + cn ) + d

where f is the sigmoid function, given by the following equation:

1
f ( z) = .
−z
e +1
In F(x), the input to the sigmoid function is x-r. x is the regressor value and
r is regressor mean, computed from the estimation data. an, bn, and cn are
the network parameters stored in the model property M.nl.par, where M
is an idnlarx object.

Compute the output value at time t=1, when the regressor values are
x=[estData.y(1),estData.u(2)]:

% Estimate model from sample data:

load twotankdata
estData = iddata(y,u,0.2,'Tstart',0);
M = nlarx(estData,[1 1 0],'sig');

4-47
 4 Nonlinear Black-Box Model Identification

NL = M.Nonlinearity
% Assign values to the parameters in the expression for F(x):
x = [estData.y(1),estData.u(2)]; % regressor values at t=1
r = NL.Parameters.RegressorMean;
P = NL.Parameters.LinearSubspace;
L = NL.Parameters.LinearCoef;
d = NL.Parameters.OutputOffset;
Q = NL.Parameters.NonLinearSubspace;
aVec = NL.Parameters.OutputCoef; %[a_1; a_2; ...]
cVec = NL.Parameters.Translation; %[c_1; c_2; ...]
bMat = NL.Parameters.Dilation; %[b_1; b_2; ...]
% Compute the linear portion of the response (plus offset):
yLinear = (x-r)*P*L+d
% Compute the nonlinear portion of the response:
f = @(z)1/(exp(-z)+1); % anonymous function for sigmoid unit
yNonlinear = 0;
for k = 1:length(aVec)
fInput = (x-r)*Q* bMat(:,k)+cVec(k);
yNonlinear = yNonlinear+aVec(k)*f(fInput);
end
% Total response y = F(x) = yLinear + yNonlinear
y = yLinear + yNonlinear; % y is equal to evaluate(NL,x)

4-48
 Identifying Hammerstein-Wiener Models

Identifying Hammerstein-Wiener Models

In this section...
“Applications of Hammerstein-Wiener Models” on page 4-49
“Structure of Hammerstein-Wiener Models” on page 4-50
“Nonlinearity Estimators for Hammerstein-Wiener Models” on page 4-52
“Ways to Configure Hammerstein-Wiener Estimation” on page 4-53
“Estimation Algorithm for Hammerstein-Wiener Models” on page 4-55
“How to Estimate Hammerstein-Wiener Models in the GUI” on page 4-55
“How to Estimate Hammerstein-Wiener Models at the Command Line”
on page 4-58
“Using Linear Model for Hammerstein-Wiener Estimation” on page 4-64
“Validating Hammerstein-Wiener Models” on page 4-70
“Using Hammerstein-Wiener Models” on page 4-76
“Low-Level Manipulation of Hammerstein-Wiener Components” on page
4-78

Applications of Hammerstein-Wiener Models

If the output of a system depends nonlinearly on its inputs, it might be possible
to decompose the input-output relationship into two or more interconnected
elements. In this case, you can represent the dynamics by a linear transfer
function and capture the nonlinearities using nonlinear functions of inputs
and outputs of the linear system. The Hammerstein-Wiener model achieves
this configuration as a series connection of static nonlinear blocks with a
dynamic linear block.

Hammerstein-Wiener model applications span several areas, such as

modeling electro-mechanical system and radio frequency components, audio
and speech processing and predictive control of chemical processes. These
models are popular because they have a convenient block representation,
transparent relationship to linear systems, and are easier to implement than
heavy-duty nonlinear models (such as neural networks and Volterra models).

4-49
 4 Nonlinear Black-Box Model Identification

You can use the Hammerstein-Wiener model as a black-box model structure

because it provides a flexible parameterization for nonlinear models. For
example, you might estimate a linear model and try to improve its quality by
adding an input or output nonlinearity to this model.

You can also use a Hammerstein-Wiener model as a grey-box structure to

capture physical knowledge about process characteristics. For example,
the input nonlinearity might represent typical physical transformations
in actuators and the output nonlinearity might describe common sensor
characteristics.

Structure of Hammerstein-Wiener Models

This block diagram represents the structure of a Hammerstein-Wiener model:

u(t) Input w(t) Linear x(t) Output y(t)

Nonlinearity Block Nonlinearity
f B/F h

where:

• w(t) = f(u(t)) is a nonlinear function transforming input data u(t). w(t) has
the same dimension as u(t).
• x(t) = (B/F)w(t) is a linear transfer function. x(t) has the same dimension
as y(t).
where B and F are similar to polynomials in the linear Output-Error model,
as described in “What Are Black-Box Polynomial Models?” on page 3-39.
For ny outputs and nu inputs, the linear block is a transfer function matrix
containing entries:

B j,i (q)
F j,i (q)

where j = 1,2,...,ny and i = 1,2,...,nu.

• y(t) = h(x(t)) is a nonlinear function that maps the output of the linear
block to the system output.

4-50
 Identifying Hammerstein-Wiener Models

w(t) and x(t) are internal variables that define the input and output of the
linear block, respectively.

Because f acts on the input port of the linear block, this function is called
the input nonlinearity. Similarly, because h acts on the output port of the
linear block, this function is called the output nonlinearity. If system contains
several inputs and outputs, you must define the functions f and h for each
input and output signal.

You do not have to include both the input and the output nonlinearity in the
model structure. When a model contains only the input nonlinearity f, it is
called a Hammerstein model. Similarly, when the model contains only the
output nonlinearity h), it is called a Wiener model.

The nonlinearities f and h are scalar functions, one nonlinear function for
each input and output channel.

The Hammerstein-Wiener model computes the output y in three stages:

1 Computes w(t) = f(u(t)) from the input data.

w(t) is an input to the linear transfer function B/F.

The input nonlinearity is a static (memoryless) function, where the value of

the output a given time t depends only on the input value at time t.

You can configure the input nonlinearity as a sigmoid network, wavelet

network, saturation, dead zone, piecewise linear function, one-dimensional
polynomial, or a custom network. You can also remove the input
nonlinearity.

2 Computes the output of the linear block using w(t) and initial conditions:
x(t) = (B/F)w(t).

You can configure the linear block by specifying the numerator B and
denominator F orders.

3 Compute the model output by transforming the output of the linear block
x(t) using the nonlinear function h: y(t) = h(x(t)).

4-51
 4 Nonlinear Black-Box Model Identification

Similar to the input nonlinearity, the output nonlinearity is a static

function. Configure the output nonlinearity in the same way as the input
nonlinearity. You can also remove the output nonlinearity, such that y(t) =
x(t).

Resulting models are idnlhw objects that store all model data, including
model parameters and nonlinearity estimator. See the idnlhw reference page
for more information.

Nonlinearity Estimators for Hammerstein-Wiener

Models
System Identification Toolbox software provides several scalar nonlinearity
estimators F(x) for Hammerstein-Wiener models. The nonlinearity estimators
are available for both the input and output nonlinearities f and h, respectively.
For more information about F(x), see “Structure of Hammerstein-Wiener
Models” on page 4-50.

Each nonlinearity estimator corresponds to an object class in this toolbox.

When you estimate Hammerstein-Wiener models in the GUI, System
Identification Toolbox creates and configures objects based on these classes.
You can also create and configure nonlinearity estimators at the command
line. For a detailed description of each estimator, see the references page of
the corresponding nonlinearity class.

Nonlinearity Class Structure Comments

Piecewise pwlinear A piecewise linear function By default,
linear parameterized by breakpoint locations. the number of
(default) breakpoints is 10.
One layer sigmoidnet Default number of
sigmoid units n is 10.
n
network
g( x) = ∑ α k κ ( βk ( x − γ k ) )
k=1

κ(s) is the sigmoid function

( )
−1
 (s) = es + 1 . βk is a row vector such
that βk ( x − γ k ) is a scalar.

4-52
 Identifying Hammerstein-Wiener Models

Nonlinearity Class Structure Comments

Wavelet wavenet By default,
network the estimation
n
algorithm
g( x) = ∑ α k κ ( βk ( x − γ k ) ) determines the
k=1
number of units
where κ(s) is the wavelet function. nautomatically.
Saturation saturation Parameterize hard limits on the signal Use to model
value as upper and lower saturation known saturation
limits. effects on signal
amplitudes.
Dead zone deadzone Parameterize dead zones in signals as Use to model
the duration of zero response. known dead
zones in signal
amplitudes.
One- poly1d Single-variable polynomial of a degree By default, the
dimensional that you specify. polynomial degree
polynomial is 1.
Unit gain unitgain Excludes the input or output Useful for
nonlinearity from the model structure configuring
to achieve a Wiener or Hammerstein multi-input,
configuration, respectively. multi-output
(MIMO) models
to exclude
Note Excluding both the input and nonlinearities from
output nonlinearities reduces the specific input and
Hammerstein-Wiener structure to a output channels.
linear transfer function.

Custom customnet Similar to sigmoid network but you (For advanced use)
network specify κ(s) . Uses the unit
(user-defined) function that you
specify.

Ways to Configure Hammerstein-Wiener Estimation

Estimate a Hammerstein-Wiener model with default configuration by:

4-53
 4 Nonlinear Black-Box Model Identification

• Specifying model order and input delay:

- nb—The number of zeros plus one.
- nf—The number of poles.
- nk—The delay from input to the output in terms of the number of
samples.
nb is the order of the transfer function numerator (B polynomial), and
nf is the order of the transfer function denominator (F polynomial). As
you fit different Hammerstein-Wiener models to your data, you can
configure the linear block structure by specifying a different order and
delay. For MIMO systems with ny outputs and nu inputs, nb, nf, and
nk are ny-by-nu matrices.
• Initializing using one of the following discrete-time linear models:
- An input-output polynomial model of Output-Error (OE) structure
(idpoly)
- A linear state-space model with no disturbance component (idss object
with K=0)
You can perform this operation only at the command line. The initialization
configures the Hammerstein-Wiener model to use orders and delay
of the linear model, and the B and F polynomials as the transfer
function numerator and denominator. See “Using Linear Model for
Hammerstein-Wiener Estimation” on page 4-64.

By default, the input and output nonlinearity estimators are both piecewise
linear functions, parameterized by breakpoint locations (see the pwlinear
reference page). You can configure the input and output nonlinearity
estimators by:

• Configuring the input and output nonlinearity properties.

• Excluding the input or output nonlinear block.

See these topics for detailed steps to change the model structure:

• “Tutorial – Identifying Nonlinear Black-Box Models Using the GUI”

• “How to Estimate Hammerstein-Wiener Models in the GUI” on page 4-55

4-54
 Identifying Hammerstein-Wiener Models

• “How to Estimate Hammerstein-Wiener Models at the Command Line”

on page 4-58

Estimation Algorithm for Hammerstein-Wiener

Models
Estimation of Hammerstein-Wiener models uses iterative search to minimize
the simulation error between the model output and the measured output.

You can configure the estimation method using the Algorithm properties of
the idnlhw class. The most common of these properties are:

• MaxIter — Maximum number of iterations.

• SearchMethod — Search method for minimization of prediction or
simulation errors, such as Gauss-Newton and Levenburg-Marquardt line
search, and Trust-region reflective Newton approach.
• Tolerance — Condition for terminating iterative search when the expected
improvement of the parameter values is less than a specified value.
• Display — Shows progress of iterative minimization in the MATLAB
Command Window.

By default, the initial states of the model are zero and not estimated.
However, you can choose to estimate initial states during model estimation,
which sometimes helps to achieve better results.

How to Estimate Hammerstein-Wiener Models in the

GUI
Prerequisites

• Learn about the Hammerstein-Wiener model structure (see “Structure of

Hammerstein-Wiener Models” on page 4-50).
• Import data into the System Identification Tool GUI (see “Preparing Data
for Nonlinear Identification” on page 4-7).
• (Optional) Choose a nonlinearity estimator in “Nonlinearity Estimators for
Hammerstein-Wiener Models” on page 4-52.

4-55
 4 Nonlinear Black-Box Model Identification

• (Optional) Estimate or construct an OE or state-space model to use

for initialization. See “Using Linear Model for Hammerstein-Wiener
Estimation” on page 4-64.

1 In the System Identification Tool GUI, select Estimate > Nonlinear

models to open the Nonlinear Models dialog box.

2 In the Configure tab, select Hammerstein-Wiener from the Model type

list.

3 (Optional) Edit the Model name by clicking the pencil icon. The name
of the model should be unique to all Hammerstein-Wiener models in the
System Identification Tool GUI.

4 (Optional) If you want to refine a previously estimated model, click

Initialize to select a previously estimated model from the Initial Model
list.

Note Refining a previously estimated model starts with the parameter

values of the initial model and uses the same model structure. You can
change these settings.

The Initial Model list includes models that:

• Exist in the System Identification Tool GUI.
• Have the same number of inputs and outputs as the dimensions of the
estimation data (selected as Working Data in the System Identification
Tool GUI).

5 Keep the default settings in the Nonlinear Models dialog box that specify
the model structure and the algorithm, or modify these settings:

Note For more information about available options, click Help in the
Nonlinear Models dialog box to open the GUI help.

4-56
 Identifying Hammerstein-Wiener Models

What to Configure Options in Comment

Nonlinear Models
GUI
Input or output In the I/O If you do not know
nonlinearity Nonlinearity which nonlinearity to
tab, select the try, use the (default)
Nonlinearity and piecewise linear
specify the No. of nonlinearity.
Units.
When you estimate
from binary input
data, you cannot
reliably estimate the
input nonlinearity.
In this case, set
Nonlinearity for
the input channel to
None.
For multiple-input
and multiple-output
systems, you can
assign nonlinearities
to specific input and
output channels.
Model order and delay In the Linear Block If you do not know the
tab, specify B Order, input delay values,
F Order, and Input click Infer Input
Delay. For MIMO Delay. This action
systems, select the opens the Infer Input
output channel and Delay dialog box which
specify the orders and suggests possible delay
delays from each input values.
channel.
Estimation algorithm In the Estimate You can specify to
tab, click Algorithm estimate initial states.
Options.

6 Click Estimate to add this model to the System Identification Tool GUI.

4-57
 4 Nonlinear Black-Box Model Identification

The Estimate tab displays the estimation progress and results.

7 Validate the model response by selecting the desired plot in the Model
Views area of the System Identification Tool GUI. For more information
about validating models, see Chapter 8, “Model Analysis”.

If you get a poor fit, try changing the model structure or algorithm
configuration in step 5.

You can export the model to the MATLAB workspace by dragging it to To

Workspace in the System Identification Tool GUI.

How to Estimate Hammerstein-Wiener Models at the

Command Line
Prerequisites

• Learn about the Hammerstein-Wiener model structure described in

“Structure of Hammerstein-Wiener Models” on page 4-50.
• Prepare your data, as described in “Preparing Data for Nonlinear
Identification” on page 4-7.
• (Optional) Choose a nonlinearity estimator in “Nonlinearity Estimators for
Hammerstein-Wiener Models” on page 4-52.
• (Optional) Estimate or construct an input-output polynomial model
of Output-Error (OE) structure (idpoly) or a state-space model
with no disturbance component (idss with K=0) for initialization
of Hammerstein-Wiener model. See “Using Linear Model for
Hammerstein-Wiener Estimation” on page 4-64.

Estimate model using nlhw.

Use nlhw to both construct and estimate a Hammerstein-Wiener model. After

each estimation, validate the model by comparing it to other models and
simulating or predicting the model response.

Basic Estimation

4-58
 Identifying Hammerstein-Wiener Models

Start with the simplest estimation using m = nlhw(data,[nb nf nk]). For

example:

m = nlhw(data,[2 2 1]) % nb=nf=2 and nk=1

The second input argument [nb nf nk] sets the order of the linear transfer
function, where nb is the number of zeros plus 1, nf is the number of poles,
and nk is the input delay. By default, both the input and output nonlinearity
estimators are piecewise linear functions (see the pwlinear reference page).
m is an idnlhw object.

For MIMO systems, nb, nf, and nk are ny-by-nu matrices. See the nlhw
reference page for more information about MIMO estimation.

Configure the nonlinearity estimator.

Specify a different nonlinearity estimator using m = nlhw(data,[nb nf

nk],InputNL,OutputNL). InputNL and OutputNL are nonlinearity estimator
objects.

Note If your input signal is binary, set InputNL to unitgain.

To use nonlinearity estimators with default settings, specify InputNL and

OutputNL using strings (such as 'wave' for wavelet network or 'sig' for
sigmoid network).

If you need to configure the properties of a nonlinearity estimator, use its

object representation. For example, to estimate a Hammerstein-Wiener model
that uses saturation as its input nonlinearity and one-dimensional polynomial
of degree 3 as its output nonlinearity:

m = nlhw(data,[2 2 1],'saturation',poly1d('Degree',3))

The third input 'saturation' is a string representation of the saturation

nonlinearity with default property values. poly1d('Degree',3) creates a
one-dimensional polynomial object of degree 3.

For MIMO models, specify the nonlinearities using objects unless you want to
use the same nonlinearity with default configuration for all channels.

4-59
 4 Nonlinear Black-Box Model Identification

This table summarizes values that specify the nonlinearity estimators.

Nonlinearity Value (Default Nonlinearity Class

Configuration)
Piecewise linear 'pwlinear' or 'pwlin' pwlinear
(default)
One layer sigmoid 'sigmoidnet' or 'sigm' sigmoidnet
network
Wavelet network 'wavenet' or 'wave' wavenet
Saturation 'saturation' or 'sat' saturation
Dead zone 'deadzone' or 'dead' deadzone
One- 'poly1d' or 'poly' poly1d
dimensional polynomial
Unit gain 'unitgain' or [ ] unitgain

Additional available nonlinearities include custom networks that you create.

Specify a custom network by defining a function called gaussunit.m, as
described in the customnet reference page. Define the custom network object
CNetw as:

CNetw = cutomnet(@gaussunit);
m = nlhw(data,[na nb nk],CNetw)

Exclude the input or output nonlinearity.

Exclude a nonlinearity for a specific channel by specifying the unitgain value

for the InputNonlinearity or OutputNonlinearity properties.

If the input signal is binary, set InputNL to unitgain.

For more information about model estimation and properties, see the nlhw
and idnlhw reference pages.

For a description of each nonlinearity estimator, see “Nonlinearity Estimators

for Hammerstein-Wiener Models” on page 4-52.

4-60
 Identifying Hammerstein-Wiener Models

Iteratively refine the model.

Use pem to refine the original model. For example:

m1 = nlhw(data, [2 2 1],'sigmoidnet','wavenet');
m2 = pem(data,m1) % can repeatedly run this command

You can also use nlhw to refine the original model:

m1 = nlhw(data, [2 2 1],'sigmoidnet','wavenet');
m2 = nlhw(data,m1) % can repeatedly run this command

Check the search termination criterion in m.EstimationInfo.WhyStop. If

WhyStop indicates that the estimation reached the maximum number of
iterations, try repeating the estimation and possibly specifying a larger value
for the MaxIter idnlhw property:

m2 = pem(data,m1,'MaxIter',30) % runs 30 more iterations starting at m1

When the m.EstimationInfo.WhyStop value is Near (local) minimum,

(norm( g) < tol or No improvement along the search direction
with line search, validate your model to see if this model adequately fits
the data. If not, the solution might be stuck in a local minimum of the
cost-function surface. Try adjusting the Algorithm.Tolerance property value
of the idnlhw class or the Algorithm.SearchMethod property, and repeat the
estimation. You can also try perturbing the parameters of the last model
using init (called randomization) and refining the model using pem:

M1 = nlhw(data, [2 2 1], `sigm','wave'); % original model

M1p = init(M1); % randomly perturbs parameters about nominal values
M2 = pem(data, M1p); % estimates parameters of perturbed model

You can display the progress of the iterative search in the MATLAB Command
Window using the Display property of the idnlhw class:

M2= pem(data,M1p,'Display','On')

Improve estimation results using initial states.

If your estimated Hammerstein-Wiener model provides a poor fit to measured

data, you can repeat the estimation using the initial state values estimated

4-61
 4 Nonlinear Black-Box Model Identification

from the data. By default, the initial states corresponding to the linear block
of the Hammerstein-Wiener model are zero.

To specify estimating initial states during model estimation, use:

m = nlhw(data,[nb nf nk],[sigmoidnet;pwlinear],[],...
'InitialState','e');

What if you cannot get a satisfactory model?

If you do not get a satisfactory model after many trials with various model
structures and algorithm settings, it is possible that the data is poor. For
example, your data might be missing important input or output variables and
does not sufficiently cover all the operating points of the system.

Nonlinear black-box system identification usually requires more data than

linear model identification to gain enough information about the system.

Example – Using nlhw to Estimate Hammerstein-Wiener

Models
Use nlhw to estimate a Hammerstein-Wiener model for the data in “Tutorial –
Identifying Nonlinear Black-Box Models Using the GUI”.

1 Prepare the data for estimation:

load twotankdata
z = iddata(y, u, 0.2);
ze = z(1:1000); zv = z(1001:3000);

2 Estimate several models using different model orders, delays, and

nonlinearity settings:

m1 = nlhw(ze,[2 3 1]);
m2 = nlhw(ze,[2 2 3]);
m3 = nlhw(ze,[2 2 3], pwlinear('num',13),...
pwlinear('num',10));
m4 = nlhw(ze,[2 2 3], sigmoidnet('num',2),...
pwlinear('num',10));

4-62
 Identifying Hammerstein-Wiener Models

An alternative way to perform the estimation is to configure the model

structure first, and then to estimate this model:

m5 = idnlhw([2 2 3], 'dead','sat')

m5 = pem(ze,m5);

3 Compare the resulting models by plotting the model outputs on top of the
measured output:

compare(zv,m1,m2,m3,m4,m5)

Example – Improving a Linear Model Using

Hammerstein-Wiener Structure
Use the Hammerstein-Wiener model structure to improve a previously
estimated linear model. After estimating the linear model, insert it into the
Hammerstein-Wiener structure that includes input or output nonlinearities.

4-63
 4 Nonlinear Black-Box Model Identification

1 Estimate a linear model:

load iddata1
LM = arx(z1,[2 2 1]);

2 Extract the transfer function coefficients from the linear model:

[Num, Den] = tfdata(LM);

3 Create a Hammerstein-Wiener model, where you initialize the linear block

properties B and F using Num and Den, respectively:

nb = 1; % In general, nb = ones(ny,nu)
% ny is number of outputs
% nu is number of inputs
nf = nb;
nk = 0; % In general, nk = zeros(ny,nu)
% ny is number of outputs
% nu is number of inputs
M = idnlhw([nb nf nk],'poly1d','pwlinear');
M.b = Num;
M.f = Den;

4 Estimate the model coefficients, which refines the linear model coefficients
in Num and Den:

M = pem(z1, M);

5 Compare responses of linear and nonlinear model against measured data:

compare(z1,LM,M)

Using Linear Model for Hammerstein-Wiener

Estimation
• “About Using Linear Models” on page 4-65
• “How to Initialize Hammerstein-Wiener Estimation Using Linear
Polynomial Output-Error or State-Space Models” on page 4-66
• “Example – Using Linear OE Models to Estimate Hammerstein-Wiener
Models” on page 4-66

4-64
 Identifying Hammerstein-Wiener Models

About Using Linear Models

You can use the following discrete-time linear models for Hammerstein-Wiener
estimation. The linear model must sufficiently represent the linear dynamics
of your system.

Discrete-Time Linear Model Use for Initializing...

Single-output polynomial model Single-output Hammerstein-Wiener
of Output-Error (OE) structure model estimation
(idpoly) or state-space model with
no disturbance component (idss
model with K = 0)
State-space with no disturbance Multi-output Hammerstein-Wiener
component (idss model with K = 0) model estimation

Tip To learn more about when to use linear models, see “When to Fit
Nonlinear Models” on page 4-2.

Typically, you use the oe or n4sid command to obtain the linear model. You
can provide the linear model only at the command line when constructing (see
idnlhw) or estimating (see nlhw) a Hammerstein-Wiener model.

The software uses the linear model for initializing the Hammerstein-Wiener
estimation:

• Assigns the linear model orders as initial values of nonlinear model orders
(nb and nf properties of the Hammerstein-Wiener (idnlhw) and delays
(nk property).
• Sets the B and F polynomials of the linear transfer function in the
Hammerstein-Wiener model structure.

During estimation, the estimation algorithm uses these values to further

adjust the nonlinear model to the data.

4-65
 4 Nonlinear Black-Box Model Identification

How to Initialize Hammerstein-Wiener Estimation Using Linear

Polynomial Output-Error or State-Space Models
Estimate a Hammerstein-Wiener model using either a linear input-output
polynomial model of OE structure or state-space model by typing

m = nlhw(data,LinModel)

LinModel must be an idpoly object of OE structure or state-space (idss)

model for single-output models, and state-space model for multi-output
models. m is an idnlhw object. data is a time-domain iddata object.

By default, both the input and output nonlinearity estimators are piecewise
linear functions (see pwlinear).

Specify different input and output nonlinearity, for example sigmoid and
deadzone:

m = nlarx(data,LinModel, 'sigmoid','deadzone')

After each estimation, validate the model by comparing the simulated

response to the data. To improve the fit of the Hammerstein-Wiener model,
adjust various elements of the Hammerstein-Wiener structure. For more
information, see “Ways to Configure Hammerstein-Wiener Estimation” on
page 4-53.

Example – Using Linear OE Models to Estimate

Hammerstein-Wiener Models

1 Load the estimation data.

load throttledata.mat

This command loads the data object ThrottleData into the MATLAB
workspace. The object contains input and output samples collected from an
engine throttle system, sampled at a rate of 100 Hz.

A DC motor controls the opening angle of the butterfly valve in the throttle
system. A step signal (in volts) drives the DC motor. The output is the
angular position (in degrees) of the valve.

4-66
 Identifying Hammerstein-Wiener Models

2 Plot the data to view and analyze the data characteristics.

plot(ThrottleData)

In the normal operating range of 15 90 degrees, the input and output

variables have a linear relationship, as shown in the following figure. You
use a linear model of low order to model this relationship.

In the throttle system, a hard stop limits the valve position to 90 degrees,
and a spring brings the valve to 15 degrees when the DC motor is turned
off. These physical components introduce nonlinearities that a linear model
cannot capture.

4-67
 4 Nonlinear Black-Box Model Identification

3 Estimate a Hammerstein-Wiener model to model the linear behavior of this

single-input single-output system in the normal operating range.

% Detrend the data because linear models cannot capture offsets.

Tr = getTrend(ThrottleData);
Tr.OutputOffset = 15;
DetrendedData = detrend(ThrottleData, Tr);

% Estimate a linear OE model with na=2, nb=1, nk=1.

LinearModel = oe(DetrendedData, [2 1 1], 'Focus', 'Simulation');

4 Compare the simulated model response with estimation data.

compare(DetrendedData, LinearModel)

The linear model captures the rising and settling behavior in the linear
operating range but does not account for output saturation at 90 degrees,
as shown in the next figure.

4-68
 Identifying Hammerstein-Wiener Models

5 Estimate a Hammerstein-Wiener model to model the output saturation.

NonlinearModel = nlhw(ThrottleData, LinearModel, [], 'saturation')

The software uses the orders and delay of the linear model for the orders
of the nonlinear model. In addition, the software uses the B and F
polynomials of the linear transfer function.

6 Compare the nonlinear model with data.

compare(ThrottleData, NonlinearModel)

4-69
 4 Nonlinear Black-Box Model Identification

Validating Hammerstein-Wiener Models

• “About Hammerstein-Wiener Plots” on page 4-71
• “How to Create Hammerstein-Wiener Plots in the GUI” on page 4-71
• “How to Validate Hammerstein-Wiener Models at the Command Line”
on page 4-72
• “Plotting Nonlinear Block Characteristics” on page 4-74
• “Plotting Linear Block Characteristics” on page 4-75

4-70
 Identifying Hammerstein-Wiener Models

About Hammerstein-Wiener Plots

Hammerstein-Wiener model plot lets you explore the characteristics of the
linear block and the static nonlinearities of the Hammerstein-Wiener model.
For more information about estimating nonlinear Hammerstein-Wiener
models, see “Identifying Hammerstein-Wiener Models” on page 4-49.

Examining a Hammerstein-Wiener plot can help you determine whether you

chose an unnecessarily complicated nonlinearity for modeling your system.
For example, if you chose a piece-wise-linear nonlinearity (which is very
general), but the plot indicates saturation behavior, then you can estimate a
new model using the simpler saturation nonlinearity instead.

For multivariable systems, you can use the Hammerstein-Wiener plot

to determine whether to exclude nonlinearities for specific channels. If
the nonlinearity for a specific input or output channel does not exhibit
strong nonlinear behavior, you can estimate a new model after setting the
nonlinearity at that channel to unit gain.

How to Create Hammerstein-Wiener Plots in the GUI

To create a Hammerstein-Wiener plot in the System Identification Tool GUI,
select the Hamm-Wiener check box in the Model Views area. For general
information about creating and working with plots, see “Working with Plots
in the System Identification Tool GUI” on page 11-15.

Note The Hamm-Wiener check box is unavailable if you do not have a

Hammerstein-Wiener model in the Model Board.

To include or exclude a model on the plot, click the corresponding model icon
in the System Identification Tool GUI. By default, the input nonlinearity
block UNL is selected. You can select the output nonlinearity block YNL or
Linear Block, as shown in the next figure.

4-71
 4 Nonlinear Black-Box Model Identification

Selected block Supported plots

is highlighted. for linear block

Hide or show
the top pane.

After you generate a plot, you can learn more about your model by:

• “Plotting Nonlinear Block Characteristics” on page 4-74

• “Plotting Linear Block Characteristics” on page 4-75

How to Validate Hammerstein-Wiener Models at the Command

Line
You can use the following approaches to validate Hammerstein-Wiener
models at the command line:

4-72
 Identifying Hammerstein-Wiener Models

Compare Model Output to Measured Output

Compare estimated models using compare. Use an independent validation

data set whenever possible. For more information about validating models,
see “Overview of Model Validation and Plots” on page 8-2.

For example, compare linear and nonlinear ARX models of same order:

load iddata1
LM = arx(z1,[2 2 1]) % estimates linear ARX model
M = nlhw(z1,[2 2 1]) % estimates Hammerstein-Wiener model
compare(z1,LM,M) % compares responses of LM and M
% against measured data

Compare the performance of several models using the

properties M.EstimationInfo.FPE (final prediction error) and
M.EstimationInfo.LossFcn (value of loss function at estimation
termination). Smaller values typically indicate better performance. However,
m.EstimationInfo.FPE values might be unreliable when the model contains
a large number of parameters relative to the estimation data size. Use these
indicators in combination with other validation techniques to draw reliable
conclusions.

Simulate and Predict Model Response

Use sim(idnlhw) and predict(idnlhw) to simulate and predict model

response, respectively. To compute the step response of the model, use step.
See the corresponding reference page for more information.

Analyze Residuals

Residuals are differences between the model output and the measured output.
Thus, residuals represent the portion of the output not explained by the
model. Use resid to compute and plot the residuals.

Plot Nonlinearity

Access the object representing the nonlinearity estimator and its parameters
using M.InputNonlinearity (or M.unl) and M.OutputNonlinearity (or
M.ynl), where M is the estimated model.

4-73
 4 Nonlinear Black-Box Model Identification

Use plot to view the shape of the nonlinearity and properties of the linear
block. For example:

plot(M)

You can use additional plot arguments to specify the following information:

• Include several Hammerstein-Wiener models on the plot.

• Configure how to evaluate the nonlinearity at each input and output
channel.
• Specify the time or frequency values for computing transient and frequency
response plots of the linear block.

The plot command opens the Hammerstein-Wiener Model Plot window.

For more information about working with this plot window, see “Plotting
Nonlinear Block Characteristics” on page 4-74 and “Plotting Linear Block
Characteristics” on page 4-75.

For detailed information about plot, type the following command at the
prompt:

help idnlhw/plot

Check Iterative Search Termination Conditions

Use M.EstimationInfo to display the estimation termination conditions,

where M is the estimated idnlhw model. For example, check the WhyStop
field of the EstimationInfo property, which describes why the estimation
was stopped. For example, the algorithm might have reached the maximum
number of iterations or the required tolerance value.

Plotting Nonlinear Block Characteristics

The Hammerstein-Wiener model can contain up to two nonlinear blocks. The
nonlinearity at the input to the Linear Block is labeled uNL and is called the
input nonlinearity. The nonlinearity at the output of the Linear Block is
labeled yNL and is called the output nonlinearity.

To configure the plot, perform the following steps:

4-74
 Identifying Hammerstein-Wiener Models

1 If the top pane is not visible, click to expand the Hammerstein-Wiener

Model Plot window.

2 Select the nonlinear block you want to plot:

• To plot uNL as a command of the input data, click the uNL block.
• To plot yNL as a command of its inputs, click the yNL block.

The selected block is highlighted in green.

Note The input to the output nonlinearity block yNL is the output from the
Linear Block and not the measured input data.

3 If your model contains multiple inputs or outputs, select the channel in the
Select nonlinearity at channel list. Selecting the channel updates the
plot and displays the nonlinearity values versus the corresponding input to
this nonlinear block.

4 To change the range of the horizontal axis, select Options > Set input
range to open the Range for Input to Nonlinearity dialog box. Enter the
range using the format [MinValue MaxValue]. Click Apply and then
Close to update the plot.

Plotting Linear Block Characteristics

The Hammerstein-Wiener model contains one Linear Block that represents
the embedded linear model.

To configure the plot:

1 If the top pane is not visible, click to expand the Hammerstein-Wiener

Model Plot window.

2 Click the Linear Block to select it. The Linear Block is highlighted in green.

3 In the Select I/O pair list, select the input and output data pair for which
to view the response.

4-75
 4 Nonlinear Black-Box Model Identification

4 In the Choose plot type list, select the linear plot from the following
options:
• Step
• Impulse
• Bode
• Pole-Zero Map

5 If you selected to plot step or impulse response, you can set the time span.
Select Options > Time span and enter a new time span in units of time
you specified for the model.

For a time span T, the resulting response is plotted from -T/4 to T. The
default time span is 10.

Click Apply and then Close.

6 If you selected to plot a Bode plot, you can set the frequency range.

The default frequency vector is 128 linearly distributed values, greater

than zero and less than or equal to the Nyquist frequency. To change the
range, select Options > Frequency range, and specify a new frequency
vector in units of rad per model time units.

Enter the frequency vector using any one of following methods:

• MATLAB expression, such as [1:100]*pi/100 or logspace(-3,-1,200).
Cannot contain variables in the MATLAB workspace.
• Row vector of values, such as [1:.1:100].

Click Apply and then Close.

Using Hammerstein-Wiener Models

Simulation and Prediction
Use sim(idnlhw) to simulate the model output, and predict(idnlhw) to
predict the model output. To compare models to measured output and to
each other, use compare.

4-76
 Identifying Hammerstein-Wiener Models

This toolbox provides a number of options to facilitate how you specify initial
states. For example, you can use findstates(idnlhw) to automatically
search for state values in simulation and prediction applications. You can also
specify the states manually.

If you need to specify the states manually, see the idnlhw reference page for a
definition of the Hammerstein-Wiener model states.

To learn more about how sim and predict compute the model output, see
“Low-Level Manipulation of Hammerstein-Wiener Components” on page 4-78.

Linearization

Compute linear approximation of Hammerstein-Wiener models using

linearize(idnlhw) or linapp.

linearize provides a first-order Taylor series approximation of the system

about an operation point (also called tangent linearization). linapp computes
a linear approximation of a nonlinear model for a given input data. For more
information, see the “Linear Approximation of Nonlinear Black-Box Models”
on page 4-81.

You can compute the operating point for linearization using findop(idnlhw).

After computing a linear approximation of a nonlinear model, you can perform

linear analysis and control design on your model using Control System
Toolbox commands. For more information, see Chapter 9, “Control Design
Applications”.

Simulation and Code Generation Using Simulink

You can import estimated Hammerstein-Wiener Model block into Simulink

software using the Hammerstein-Wiener block from the System Identification
Toolbox block library. After you bring the idnlhw object from the workspace
into Simulink, you can simulate the model output.

The Hammerstein-Wiener block supports code generation with Real-Time

Workshop software, using both generic and embedded targets. Code
generation does not work when the model contains customnet as the input or
output nonlinearity.

4-77
 4 Nonlinear Black-Box Model Identification

For more information about working in the Simulink environment, see

Chapter 10, “System Identification Toolbox Blocks”.

Low-Level Manipulation of Hammerstein-Wiener

Components
In most applications, sim(idnlhw) and predict(idnlhw) are sufficient
for computing the simulated and predicted model response, respectively.
This advanced topic describes how the software evaluates the output of
nonlinearity estimators and uses this output to compute the model response.

Evaluating Nonlinearities (SISO)

Evaluating the predicted output of a nonlinearity for a input u requires that
you first extract the input or output nonlinearity F from the model:

F = M.InputNonlinearity % equivalent to F = M.unl

H = M.OutputNonlinearity % equivalent to F = M.ynl

Evaluate F(u):

w = evaluate(F,u)

where u is a scalar representing the value of the input signal at a given time.

You can evaluate predicted output values at multiple time instants by

evaluating F for several time values simultaneously using a column vector
of input values:

w = evaluate(F,[u1;u2;u3])

Similarly, you can evaluate the value of the nonlinearity H using the output
of the linear block x(t) as its input:

y = evaluate(H,x)

Evaluating Nonlinearities (MIMO)

For MIMO models, F and H are vectors of length nu and ny, respectively. nu
is the number of inputs and ny is the number of outputs. In this case, you
must evaluate the predicted output of each nonlinearity separately.

4-78
 Identifying Hammerstein-Wiener Models

For example, suppose that you estimate a two-input model:

M = nlhw(data,[nb nf nk],[wavenet;poly1d],'saturation')

In the input nonlinearity:

F = M.InputNonlinearity
F1 = F(1);
F2 = F(2);

F is a vector function containing two elements: F=[F1(u1_value);

F2(u2_value)], where F1 is a wavenet object and F2 is a poly1d object.
u1_value is the first input signal and u2_value is the second input signal.

Evaluate F by evaluating F1 and F2 separately:

w1 = evaluate(F(1), u1_value);
w2 = evaluate(F(2), u2_value);

The total input to the linear block, w, is a vector of w1 and w2 (w = [w1 w2]).

Similarly, you can evaluate the value of the nonlinearity H:

H = M.OutputNonlinearity %equivalent to H = M.ynl

Example – Low-level Simulation of Hammerstein-Wiener Model

This example shows how the software evaluates the simulated output by first
computing the output of the input and output nonlinearity estimators. For
same initial conditions, the prediction results match the simulation results.

1 Estimate Hammerstein-Wiener model:

load twotankdata
estData = iddata(y,u,0.2)
M = nlhw(estData,[1 5 3],'pwlinear','poly1d');

2 Extract the input nonlinearity, linear model, and output nonlinearity as

separate variables:

uNL = M.InputNonlinearity;
linModel = M.LinearModel;

4-79
 4 Nonlinear Black-Box Model Identification

yNL = M.OutputNonlinearity;

3 Simulate the output of the input nonlinearity estimator:

u = estData.u; %input data for simulation

% Compute output of input nonlinearity:
w = evaluate(uNL, u);
% Response of linear model to input w and zero
% initial conditions:
x = sim(linModel, w);
% Compute the output of the Hammerstein-Wiener model M
% as the output of the output nonlinearity estimator to input x:
y = evaluate(yNL, x);
% Previous commands are equivalent to:
ysim = sim(M, u);
% Compare low-level and direct simulation results:
time = estData.SamplingInstants;
plot(time, y, time, ysim, '.')

4-80
 Linear Approximation of Nonlinear Black-Box Models

Linear Approximation of Nonlinear Black-Box Models

In this section...
“Why Compute a Linear Approximation of a Nonlinear Model?” on page 4-81
“Choosing Your Linear Approximation Approach” on page 4-81
“Linear Approximation of Nonlinear Black-Box Models for a Given Input”
on page 4-82
“Tangent Linearization of Nonlinear Black-Box Models” on page 4-82
“Computing Operating Points for Nonlinear Black-Box Models” on page 4-83

Why Compute a Linear Approximation of a

Nonlinear Model?
Control design and linear analysis techniques using Control System Toolbox
software require linear models. You can use your estimated nonlinear model
in these applications after you linear the model. After you linearize your
model, you can use the model for control design and linear analysis. For more
information, see Chapter 9, “Control Design Applications”.

Choosing Your Linear Approximation Approach

System Identification Toolbox software provides two approaches for computing
a linear approximation of nonlinear ARX and Hammerstein-Wiener models.

To compute a linear approximation of a nonlinear model for a given input

signal, use the linapp command. The resulting model is only valid for the
same input that you use to compute the linear approximation. For more
information, see “Linear Approximation of Nonlinear Black-Box Models for a
Given Input” on page 4-82.

If you want a tangent approximation of the nonlinear dynamics that is

accurate near the system operating point, use the linearize command. The
resulting model is a first-order Taylor series approximation for the system
about the operating point, which is defined by a constant input and model
state values. For more information, see “Tangent Linearization of Nonlinear
Black-Box Models” on page 4-82.

4-81
 4 Nonlinear Black-Box Model Identification

Linear Approximation of Nonlinear Black-Box

Models for a Given Input
linapp computes the best linear approximation, in a mean-square-error
sense, of a nonlinear ARX or Hammerstein-Wiener model for a given input
or a randomly generated input. The resulting linear model might only be
valid for the same input signal as you the one you used to generate the linear
approximation.

linapp estimates the best linear model that is structurally similar to the
original nonlinear model and provides the best fit between a given input and
the corresponding simulated response of the nonlinear model.

To compute a linear approximation of a nonlinear black-box model for a given

input, you must have these variables:

• Nonlinear ARX model (idnlarx object) or Hammerstein-Wiener model

(idnlhw object)
• Input signal for which you want to obtain a linear approximation, specified
as a real matrix or an iddata object

linapp uses the specified input signal to compute a linear approximation:

• For nonlinear ARX models, linapp estimates a linear ARX model using the
same model orders na, nb, and nk as the original model.
• For Hammerstein-Wiener models, linapp estimates a linear Output-Error
(OE) model using the same model orders nb, nf, and nk.

To compute a linear approximation of a nonlinear black-box model for a

randomly generated input, you must specify the minimum and maximum
input values for generating white-noise input with a magnitude in this
rectangular range, umin and umax.

For more information, see the linapp reference page.

Tangent Linearization of Nonlinear Black-Box Models

linearize computes a first-order Taylor series approximation for nonlinear
system dynamics about an operating point, which is defined by a constant

4-82
 Linear Approximation of Nonlinear Black-Box Models

input and model state values. The resulting linear model is accurate in the
local neighborhood of this operating point.

To compute a tangent linear approximation of a nonlinear black-box model,

you must have these variables:

• Nonlinear ARX model (idnlarx object) or Hammerstein-Wiener model

(idnlhw object)
• Operating point

To specify the operating point of your system, you must specify the constant
input and the states. For more information about state definitions for each
type of parametric model, see these reference pages:

• idnlarx — Nonlinear ARX model

• idnlhw — Nonlinear Hammerstein-Wiener model

If you do not know the operating point values for your system, see “Computing
Operating Points for Nonlinear Black-Box Models” on page 4-83.

For more information, see the linearize(idnlarx) or linearize(idnlhw)

reference page.

Computing Operating Points for Nonlinear Black-Box

Models
An operating point is defined by a constant input and model state values.

If you do not know the operating conditions of your system for linearization,
you can use findop to compute the operating point from specifications:

• “Computing Operating Point from Steady-State Specifications” on page 4-83

• “Computing Operating Points at a Simulation Snapshot” on page 4-84

Computing Operating Point from Steady-State Specifications

Use findop to compute an operating point from steady-state specifications:

4-83
 4 Nonlinear Black-Box Model Identification

• Values of input and output signals.

If either the steady-state input or output value is unknown, you can specify
it as NaN to estimate this value. This is especially useful when modeling
MIMO systems, where only a subset of the input and output steady-state
values are known.
• More complex steady-state specifications.
Construct an object that stores specifications for computing the
operating point, including input and output bounds, known values,
and initial guesses. For more information, see operspec(idnlarx) or
operspec(idnlhw).

For more information, see the findop(idnlarx) or findop(idnlhw) reference

page.

Computing Operating Points at a Simulation Snapshot

Compute an operating point at a specific time during model simulation
(snapshot) by specifying the snapshot time and the input value. To use this
method for computing the equilibrium operating point, choose an input that
leads to a steady-state output value. Use that input and the time value
at which the output reaches steady state (snapshot time) to compute the
operating point.

It is optional to specify the initial conditions for simulation when using this
method because initial conditions often do not affect the steady-state values.
By default, the initial conditions are zero.

However, for nonlinear ARX models, the steady-state output value might
depend on initial conditions. For these models, you should investigate
the effect of initial conditions on model response and use the values that
produce the desired output. You can use data2state(idnlarx) to map the
input-output signal values from before the simulation starts to the model’s
initial states. Because the initial states are a function of the past history of
the model’s input and output values, data2state generates the initial states
by transforming the data.

4-84
 5

ODE Parameter Estimation

(Grey-Box Modeling)

• “Supported Grey-Box Models” on page 5-2

• “Data Supported by Grey-Box Models” on page 5-3
• “Choosing idgrey or idnlgrey Model Object” on page 5-4
• “Estimating Linear Grey-Box Models” on page 5-6
• “Estimating Nonlinear Grey-Box Models” on page 5-16
• “After Estimating Grey-Box Models” on page 5-23
 5 ODE Parameter Estimation (Grey-Box Modeling)

Supported Grey-Box Models

If you understand the physics of your system and can represent the system
using ordinary differential or difference equations (ODEs) with unknown
parameters, then you can use System Identification Toolbox commands to
perform linear or nonlinear grey-box modeling. Grey-box model ODEs specify
the mathematical structure of the model explicitly, including couplings
between parameters and known parameter values. Grey-box modeling is
useful when you know the relationships between variables, constraints on
model behavior, or explicit equations representing system dynamics.

The toolbox supports both continuous-time and discrete-time models.

However, because most laws of physics are expressed in continuous time, it is
easier to construct models with physical insight in continuous time, rather
than in discrete time.

In addition to dynamic input-output models, you can also create time-series

models that have no inputs and static models that have no states.

If it is too difficult to describe your system using known physical laws, you
can perform black-box modeling.

You can also use the idss model object to perform structured model
estimation by using structure matrices As, Bs, Cs, Ds, X0s, Ks to fix or free
specific parameters. However, you cannot use this approach to estimate
arbitrary structures (arbitrary parameterization). For more information
about structure matrices, see “How to Estimate State-Space Models with
Structured Parameterization” on page 3-94.

5-2
 Data Supported by Grey-Box Models

Data Supported by Grey-Box Models

You can estimate both continuous-time or discrete-time grey-box models for
data with the following characteristics:

• Time-domain or frequency-domain data, including time-series data with no

inputs.

Note Nonlinear grey-box models support only time-domain data.

• Single-output or multiple-output data

You must first import your data into the MATLAB workspace. If you are
using the System Identification Tool GUI, then import the data into the GUI
to make the data available to the toolbox. However, if you prefer to work at
the command line, then represent your data as an iddata or idfrd object.
For more information about preparing data for identification, see Chapter 2,
“Data Import and Processing”.

5-3
 5 ODE Parameter Estimation (Grey-Box Modeling)

Choosing idgrey or idnlgrey Model Object

Grey-box models require that you specify the structure of the ODE model in
a file. You use this file to create the idgrey or idnlgrey model object. You
can use both the idgrey and the idnlgrey objects to model linear systems.
However, you can only represent nonlinear dynamics using the idnlgrey
model object.

The idgrey object requires that you write a function to describe the linear
dynamics in the state-space form, such that this file returns the state-space
matrices as a function of your parameters. For more information, see
“Specifying the Linear Grey-Box Model Structure” on page 5-6.

The idnlgrey object requires that you write a function or MEX-file to describe
the dynamics as a set of first-order differential equations, such that this file
returns the output and state derivatives as a function of time, input, state,
and parameter values. For more information, see “Specifying the Nonlinear
Grey-Box Model Structure” on page 5-17.

The following table compares idgrey and idnlgrey model objects.

Comparison of idgrey and idnlgrey Objects

Settings and Supported by Supported by

Operations idgrey? idnlgrey?
Set bounds on No Yes
parameter values.
Handle initial states No Yes
individually.
Perform linear analysis. Yes No
For example, use the
bode command.

5-4
 Choosing idgrey or idnlgrey Model Object

Comparison of idgrey and idnlgrey Objects (Continued)

Settings and Supported by Supported by

Operations idgrey? idnlgrey?
Honor stability Yes No
constraints. Specify
constraints using
Algorithm.Advanced. Note You can use
Threshold.Zstability parameter bounds to
and ensure stability of
Algorithm.Advanced. an idnlgrey model,
Threshold.Sstability if these bounds are
model properties. known.

Estimate a disturbance Yes No

model. The disturbance model
is represented by K in
state-space equations.
Optimize estimation Yes No
results for simulation Set the Because idnlgrey
or prediction. Algorithm.Focus models are
property to Output-Error models,
'Simulation' or there is no difference
'Prediction'. between simulation and
prediction results.

5-5
 5 ODE Parameter Estimation (Grey-Box Modeling)

Estimating Linear Grey-Box Models

In this section...
“Specifying the Linear Grey-Box Model Structure” on page 5-6
“Example – Creating a Function for Representing a Grey-Box Model” on
page 5-7
“Example – Estimating a Continuous-Time Grey-Box Model for Heat
Diffusion” on page 5-9
“Example – Estimating a Discrete-Time Grey-Box Model with
Parameterized Disturbance” on page 5-12

Specifying the Linear Grey-Box Model Structure

You can estimate linear discrete-time and continuous-time grey-box models
for arbitrary ordinary differential or difference equations using single-output
and multiple-output time-domain data, or output-only time-series data.

You must represent your system equations in state-space form. State-space

models use state variables x(t) to describe a system as a set of first-order
differential equations, rather than by one or more nth-order differential
equations.

In continuous-time, the state-space description has the following form:

 (t)
x (t) = Fx(t) + Gu(t) + Kw
y(t) = Hx(t) + Du(t) + w(t)
x(0) = x0

The discrete-time state-space model structure is often written in the

innovations form:

x(kT + T ) = Ax(kT ) + Bu(kT ) + Ke(kT )

y(kT ) = Cx(kT ) + Du(kT ) + e(kT )
x(0) = x0

5-6
 Estimating Linear Grey-Box Models

The first step in grey-box modeling is to write a function that returns

state-space matrices as a function of user-defined parameters and information
about the model.

Use the following format to implement the linear grey-box model in the file:

[A,B,C,D,K,x0] = myfunc(par,T,aux)

where the matrices A, B, C, D, K, and x0 represent both the continuous-time

and discrete-time state-space description of the system, myfunc is the name
of the file, par contains the parameters as a column vector, and T is the
sampling interval. aux contains auxiliary variables in your system. You use
auxiliary variables to vary system parameters at the input to the function,
and avoid editing the file.

CDmfile is an optional argument that describes whether the resulting

state-space matrices are in discrete time or continuous time. By default,
CDmfile='cd', which means that the sampling interval property of the model
Ts determines whether the model is continuous or discrete in time. For more
information about these arguments, see the idgrey reference page.

Use pem to estimate your grey-box model.

Example – Creating a Function for Representing a

Grey-Box Model
In this example, you represent the structure of the following continuous-time
model:

⎡0 1 ⎤ ⎡0⎤
x (t) = ⎢ ⎥ x(t) + ⎢ ⎥ u(t)
⎣0 θ1 ⎦ ⎣ θ2 ⎦
⎡1 0 ⎤
y(t) = ⎢ ⎥ x(t) + e(t)
⎣0 1 ⎦
⎡ θ3 ⎤
x(0) = ⎢ ⎥
⎣0⎦

5-7
 5 ODE Parameter Estimation (Grey-Box Modeling)

This equation represents an electrical motor, where y1 (t) = x1 (t) is the

angular position of the motor shaft, and y2 (t) = x2 (t) is the angular velocity.

The parameter −θ1 is the inverse time constant of the motor, and − θ2 θ is the
1
static gain from the input to the angular velocity.

The motor is at rest at t=0, but its angular position θ3 is unknown. Suppose
that the approximate nominal values of the unknown parameters are θ1 = −1
and θ2 = 0.25 . For more information about this example, see the section
on state-space models in System Identification: Theory for the User, Second
Edition, by Lennart Ljung, Prentice Hall PTR, 1999.

The continuous-time state-space model structure is defined by the following

equation:

 (t)
x (t) = Fx(t) + Gu(t) + Kw
y(t) = Hx(t) + Du(t) + w(t)
x(0) = x0

To prepare this model for identification:

1 Create the following file to represent the model structure in this example:

function [A,B,C,D,K,x0] = myfunc(par,T,aux)

A = [0 1; 0 par(1)];
B = [0;par(2)];
C = eye(2);
D = zeros(2,1);
K = zeros(2,2);
x0 =[par(3);0];

5-8
 Estimating Linear Grey-Box Models

2 Use the following syntax to define an idgrey model object based on the
myfunc file:

m = idgrey('myfunc',par,'c',T,aux)

where par represents user-defined parameters and contains their nominal

(initial) values. 'c' specifies that the underlying parameterization is in
continuous time. aux contains the values of the auxiliary parameters.

Note You must specify T and aux even if they are not used by the myfunc
code.

Use pem to estimate the grey-box parameter values:

m = pem(data,m)

where data is the estimation data and m is the idgrey object with unknown
parameters.

Note Compare this example to “Example – Estimating Structured

Continuous-Time State-Space Models” on page 3-98, where the same problem
is solved using a structured state-space representation.

Example – Estimating a Continuous-Time Grey-Box

Model for Heat Diffusion
In this example, you estimate the heat conductivity and the heat-transfer
coefficient of a continuous-time grey-box model for a heated-rod system.

This system consists of a well-insulated metal rod of length L and a

heat-diffusion coefficient κ . The input to the system is the heating power u(t)
and the measured output y(t) is the temperature at the other end.

5-9
 5 ODE Parameter Estimation (Grey-Box Modeling)

Under ideal conditions, this system is described by the heat-diffusion

equation—which is a partial differential equation in space and time.

∂x(t, ξ) ∂ 2 x(t, ξ)
=κ
∂t ∂ξ2

To get a continuous-time state-space model, you can represent the

second-derivative using the following difference approximation:

∂ 2 x(t, ξ) x ( t, ξ + ΔL ) − 2 x(t, ξ) + x ( t, ξ − ΔL )
=
∂ξ 2
( ΔL )2
where ξ = k ⋅ ΔL

This transformation produces a state-space model of order n = ΔLL , where

the state variables x(t, k ⋅ ΔL) are lumped representations for x(t, ξ) for the
following range of values:

k ⋅ ΔL ≤ ξ < ( k + 1) ΔL

The dimension of x depends on the spatial grid size ΔL in the approximation.

The heat-diffusion equation is mapped to the following continuous-time

state-space model structure to identify the state-space matrices:

 (t)
x (t) = Fx(t) + Gu(t) + Kw
y(t) = Hx(t) + Du(t) + w(t)
x(0) = x0

5-10
 Estimating Linear Grey-Box Models

The following code describes the state-space equation for this model. In this
case, the auxiliary variables specify grid-size variables, so that you can modify
the grid size without the code file.

function [A,B,C,D,K,x0] = heatd(pars,T,aux)

% Number of points in the space-discretization
Ngrid = aux(1);
% Length of the rod
L = aux(2);
% Initial rod temperature (uniform)
temp = aux(3);
% Space interval
deltaL = L/Ngrid;
% Heat-diffusion coefficient
kappa = pars(1);
% Heat transfer coefficient at far end of rod
htf = pars(2);
A = zeros(Ngrid,Ngrid);
for kk = 2:Ngrid-1
A(kk,kk-1) = 1;
A(kk,kk) = -2;
A(kk,kk+1) = 1;
end
% Boundary condition on insulated end
A(1,1) = -1; A(1,2) = 1;
A(Ngrid,Ngrid-1) = 1;
A(Ngrid,Ngrid) = -1;
A = A*kappa/deltaL/deltaL;
B = zeros(Ngrid,1);
B(Ngrid,1) = htf/deltaL;
C = zeros(1,Ngrid);
C(1,1) = 1;
D = 0;
K = zeros(Ngrid,1);
x0 = temp*ones(Ngrid,1);

5-11
 5 ODE Parameter Estimation (Grey-Box Modeling)

Use the following syntax to define an idgrey model object based on the heatd
code file:

m = idgrey('heatd',[0.27 1],'c',[10,1,22])

This command specifies the auxiliary parameters as inputs to the function,

include the model order 10, the rod length of 1 meter, and an initial
temperature of 22 degrees Celsius. The command also specifies the initial
values for heat conductivity as 0.27, and for the heat transfer coefficient as 1.

For given data, you can use pem to estimate the grey-box parameter values:

me = pem(data,m)

The following command shows how you can specify to estimate a new model
with different auxiliary variables directly in the estimator command:

me = pem(data,m,'FileArgument',[20,1,22])

This syntax uses the FileArgument model property to specify a finer grid
using a larger value for Ngrid. For more information about linear grey-box
model properties, see the idgrey reference page.

Example – Estimating a Discrete-Time Grey-Box

Model with Parameterized Disturbance
This example shows how to create a single-input and single-output grey-box
model structure when you know the variance of the measurement noise. The
code in this example uses the Control System Toolbox command kalman for
computing the Kalman gain from the known and estimated noise variance.

5-12
 Estimating Linear Grey-Box Models

Description of the SISO System

This example is based on a discrete, single-input and single-output (SISO)
system represented by the following state-space equations:

⎡ par1 par 2⎤ ⎡1 ⎤
x(kT + T ) = ⎢ ⎥ x(kT ) + ⎢ ⎥ u(kT ) + w(kT )
⎣ 1 0 ⎦ ⎣0 ⎦
y(kT ) = [ par 3 par 4 ] x(kT ) + e(kT )
x(0) = x0

where w and e are independent white-noise terms with covariance matrices

R1 and R2, respectively. R1=E{ww’} is a 2–by-2 matrix and R2=E{ee’} is a
scalar. par1, par2, par3, and par4 represent the unknown parameter values
to be estimated.

Assume that you know the variance of the measurement noise R2 to be 1.

R1(1,1) is unknown and is treated as an additional parameter par5. The
remaining elements of R1 are known to be zero.

Estimating the Parameters of an idgrey Model

You can represent the system described in “Description of the SISO System”
on page 5-13 as an idgrey (grey-box) model using a function. Then, you can
use this file and the pem command to estimate the model parameters based
on initial parameter guesses.

To run this example, you must load an input-output data set and represent
it as an iddata or idfrd object called data. For more information about
this operation, see “Representing Time- and Frequency-Domain Data Using
iddata Objects” on page 2-49 or “Representing Frequency-Response Data
Using idfrd Objects” on page 2-69.

To estimate the parameters of a grey-box model:

5-13
 5 ODE Parameter Estimation (Grey-Box Modeling)

1 Create the file mynoise that computes the state-space matrices as a

function of the five unknown parameters and the auxiliary variable that
represents the known variance R2.

Note R2 is treated as an auxiliary variable rather than assigned a value

in the file to let you change this value directly at the command line and
avoid editing the file.

function [A,B,C,D,K,x0] = mynoise(par,T,aux)

R2 = aux(1); % Known measurement noise variance
A = [par(1) par(2);1 0];
B = [1;0];
C = [par(3) par(4)];
D = 0;
R1 = [par(5) 0;0 0];
[est,K] = kalman(ss(A,eye(2),C,0,T),R1,R2);
% Uses Control System Toolbox product
% u=[]
x0 = [0;0];

2 Specify initial guesses for the unknown parameter values and the auxiliary
parameter value R2:

par1 = 0.1; % Initial guess for A(1,1)

par2 = -2; % Initial guess for A(1,2)
par3 = 1; % Initial guess for C(1,1)
par4 = 3; % Initial guess for C(1,2)
par5 = 0.2; % Initial guess for R1(1,1)
Pvec = [par1; par2; par3; par4; par5]
auxVal = 1; % R2=1

3 Construct an idgrey model using the mynoise file:

Minit = idgrey('mynoise',Pvec,'d',auxVal);

The third input argument 'd' specifies a discrete-time system.

4 Estimate the model parameter values from data:

5-14
 Estimating Linear Grey-Box Models

Model = pem(data,Minit)

5-15
 5 ODE Parameter Estimation (Grey-Box Modeling)

Estimating Nonlinear Grey-Box Models

In this section...
“Supported Nonlinear Grey-Box Models” on page 5-16
“Nonlinear Grey-Box Demos and Examples” on page 5-16
“Specifying the Nonlinear Grey-Box Model Structure” on page 5-17
“Constructing the idnlgrey Object” on page 5-18
“Using pem to Estimate Nonlinear Grey-Box Models” on page 5-19
“Options for the Estimation Algorithm” on page 5-20

Supported Nonlinear Grey-Box Models

You can estimate nonlinear discrete-time and continuous-time grey-box
models for arbitrary nonlinear ordinary differential equations using
single-output and multiple-output time-domain data, or output-only
time-series data. Your grey-box models can be static or dynamic.

Grey-box models describe the system behavior as a set of nonlinear ordinary

differential or difference equations (ODEs) with unknown parameters.

Nonlinear Grey-Box Demos and Examples

The System Identification Toolbox product provides several demos and case
studies on creating, manipulating, and estimating nonlinear grey-box models.
You can access these demos by typing the following command at the prompt:

iddemo

For examples of code files and MEX-files that specify model structure, see the
toolbox/ident/iddemos/examples folder. For example, the model of a DC
motor—used in the demo idnlgreydemo1—is described in files dcmotor_m
and dcmotor_c.

5-16
 Estimating Nonlinear Grey-Box Models

Specifying the Nonlinear Grey-Box Model Structure

You must represent your system as a set of first-order nonlinear difference or
differential equations:

x† (t) = F (t, x(t), u(t), par1, par 2,..., parN )

y(t) = H (t, x(t), u(t), par1, par 2,..., parN ) + e(t)
x(0) = x0

where x† (t) = dx(t) dt for continuous-time representation and x† (t) = x(t + Ts )

for discrete-time representation with Ts as the sampling interval. F and
H are arbitrary linear or nonlinear functions with Nx and Ny components,
respectively. Nx is the number of states and Ny is the number of outputs.

After you establish the equations for your system, create a function or
MEX-file. MEX-files, which can be created in C or Fortran, are dynamically
linked subroutines that can be loaded and executed by the MATLAB
interpreter. For more information about MEX-files, see the MATLAB
documentation.

The purpose of the model file is to return the state derivatives and model
outputs as a function of time, states, inputs, and model parameters, as follows:

[dx,y] = MODFILENAME(t,x,u,p1,p2, ...,pN,FileArgument)

Tip The template file for writing the C MEX-file,

IDNLGREY_MODEL_TEMPLATE.c, is located in matlab/toolbox/ident/nlident.

The output variables are:

• dx — Represents the right side(s) of the state-space equation(s). A column

vector with Nx entries. For static models, dx=[].
For discrete-time models. dx is the value of the states at the next time
step x(t+Ts).

For continuous-time models. dx is the state derivatives at time t, or dx

dt
.

5-17
 5 ODE Parameter Estimation (Grey-Box Modeling)

• y — Represents the right side(s) of the output equation(s). A column vector

with Ny entries.

The file inputs are:

• t — Current time.
• x — State vector at time t. For static models, equals [].
• u — Input vector at time t. For time-series models, equals [].
• p1,p2, ...,pN — Parameters, which can be real scalars, column vectors
or two-dimensional matrices. N is the number of parameter objects. For
scalar parameters, N is the total number of parameter elements.
• FileArgument — Contains auxiliary variables that might be required for
updating the constants in the state equations.

Tip After creating a model file, call it directly from the MATLAB software
with reasonable inputs and verify the output values.

For an example of creating grey-box model files and idnlgrey model object,
see the demo Creating idnlgrey Model Files.

Constructing the idnlgrey Object

After you create the function or MEX-file with your model structure, you must
define an idnlgrey object. This object shares many of the properties of the
linear idgrey model object.

Use the following syntax to define the idnlgrey model object:

m = idnlgrey('filename',Order,Parameters,InitialStates)

The idnlgrey arguments are defined as follows:

• 'filename' — Name of the function or MEX-file storing the model

structure. This file must be on the MATLAB path when you use this model
object for model estimation, prediction, or simulation.

5-18
 Estimating Nonlinear Grey-Box Models

• Order — Vector with three entries [Ny Nu Nx], specifying the number of
model outputs Ny, the number of inputs Nu, and the number of states Nx.
• Parameters — Parameters, specified as struct arrays, cell arrays, or
double arrays.
• InitialStates — Specified in the same way as parameters. Must be the
fourth input to the idnlgrey constructor.

For detailed information about this object and its properties, see the idnlgrey
reference page.

Use pem to estimate your grey-box model.

Using pem to Estimate Nonlinear Grey-Box Models

You can use the pem command to estimate the unknown idnlgrey model
parameters and initial states using measured data.

The input-output dimensions of the data must be compatible with the input
and output orders you specified for the idnlgrey model.

Use the following general estimation syntax:

m = pem(data,m)

where data is the estimation data and m is the idnlgrey model object you
constructed.

You can pass additional property-value pairs to pem to specify the properties
of the model or the estimation algorithm. Assignable properties include
the ones returned by the get(idnlgrey) command and the algorithm
properties returned by the get(idnlgrey, 'Algorithm'), such as MaxIter
and Tolerance. For detailed information about these model properties, see
the idnlgrey reference page.

For more information about validating your models, see Chapter 8, “Model
Analysis”.

5-19
 5 ODE Parameter Estimation (Grey-Box Modeling)

Options for the Estimation Algorithm

The Algorithm property of the model specifies the estimation algorithm,
which simulates the model several times by trying various parameter values
to reduce the prediction error.

The following algorithm properties can affect the quality of the results:

• “Simulation Method” on page 5-20

• “Search Method” on page 5-20
• “Gradient Options” on page 5-21
• “Example – Specifying Algorithm Properties” on page 5-21

For detailed information about these and other model properties, see the
idnlgrey reference page.

Simulation Method
You can specify the simulation method using the SimulationOptions
(struct) fields of the model Algorithm property.

System Identification Toolbox software provides several variable-step and

fixed-step solvers for simulating idnlgrey models. To view a list of available
solvers and their properties, type the following command at the prompt:

idprops idnlgrey algorithm.simulationoptions

For discrete-time systems, the default solver is 'FixedStepDiscrete'. For

continuous-time systems, the default solver is 'ode45'.

By default, SimulationOptions.Solver is set to 'Auto', which automatically

selects either 'ode45' or 'FixedStepDiscrete' during estimation and
simulation—depending on whether the system is continuous or discrete in
time.

Search Method
You can specify the search method for estimating model parameters using the
SearchMethod field of the Algorithm property. Two categories of methods are
available for nonlinear grey-box modeling.

5-20
 Estimating Nonlinear Grey-Box Models

One category of methods consists of the minimization schemes that are

based on line-search methods, including Gauss-Newton type methods,
steepest-descent methods, and Levenberg-Marquardt methods.

The Trust-Region Reflective Newton method of nonlinear least-squares

(lsqnonlin), where the cost is the sum of squares of errors between the
measured and simulated outputs, requires Optimization Toolbox™ software.
When the parameter bounds differ from the default +/- Inf, this search
method handles the bounds better than the schemes based on a line search.
However, unlike the line-search-based methods, lsqnonlin only works with
Criterion='Trace'.

By default, SearchMethod is set to Auto, which automatically selects a

method from the available minimizers. If the Optimization Toolbox product is
installed, SearchMethod is set to 'lsqnonlin'. Otherwise, SearchMethod is a
combination of line-search based schemes.

Gradient Options
You can specify the method for calculating gradients using the
GradientOptions field of the Algorithm property. Gradients are the
derivatives of errors with respect to unknown parameters and initial states.

Gradients are calculated by numerically perturbing unknown quantities and

measuring their effects on the simulation error.

Option for gradient computation include the choice of the differencing scheme
(forward, backward or central), the size of minimum perturbation of the
unknown quantities, and whether the gradients are calculated simultaneously
or individually.

Example – Specifying Algorithm Properties

You can specify the Algorithm fields directly in the estimation syntax, as
property-value pairs.

For example, you can specify the following properties as part of the pem syntax:

m = pem(data,init_model,'Search','gn',...
'MaxIter',5,...
'Display','On')

5-21
 5 ODE Parameter Estimation (Grey-Box Modeling)

5-22
 After Estimating Grey-Box Models

After Estimating Grey-Box Models

After estimating linear and nonlinear grey-box models, you can simulate the
model output using the sim command. For more information, see “Overview
of Model Validation and Plots” on page 8-2.

The toolbox represents linear grey-box models using the idgrey model object.
To convert grey-box models to state-space form, use the idss command,
as described in “Transforming Between Linear Model Representations” on
page 3-118. You can then analyze the model behavior using transient- and
frequency-response plots and other linear analysis plots.

The toolbox represents nonlinear grey-box models as idnlgrey model objects.

These model objects store the parameter values resulting from the estimation.
You can access these parameters from the model objects to use these variables
in computation in the MATLAB workspace.

Note Linearization of nonlinear grey-box models is not supported.

You can import grey box models into a Simulink model using the System
Identification Toolbox Block Library. For more information, see “Simulating
Model Output” on page 10-6.

5-23
 5 ODE Parameter Estimation (Grey-Box Modeling)

5-24
 6

Time Series Identification

• “What Are Time-Series Models?” on page 6-2

• “Preparing Time-Series Data” on page 6-3
• “Estimating Time-Series Power Spectra” on page 6-4
• “Estimating AR and ARMA Models” on page 6-7
• “Estimating State-Space Time-Series Models” on page 6-12
• “Example – Identifying Time-Series Models at the Command Line” on page
6-14
• “Estimating Nonlinear Models for Time-Series Data” on page 6-15
 6 Time Series Identification

What Are Time-Series Models?

A time series is one or more measured output channels with no measured
input.

You can estimate time-series spectra using both time- and frequency-domain
data. Time-series spectra describe time-series variations using cyclic
components at different frequencies.

You can also estimate parametric autoregressive (AR), autoregressive and

moving average (ARMA), and state-space time-series models. For a definition
of these models, see “Definition of AR and ARMA Models” on page 6-7.

Note ARMA and state-space models are supported for time-domain data
only. Only single-output ARMA models are supported.

6-2
 Preparing Time-Series Data

Preparing Time-Series Data

Before you can estimate models for time-series data, you must import your
data into the MATLAB software. You can estimate models from either
time-domain and frequency-domain data. For information about which
variables you need to represent time-series data, see “Importing Time-Series
Data into MATLAB” on page 2-8.

For more information about preparing data for modeling, see “Ways to Process
Data for System Identification” on page 2-4.

If your data is already in the MATLAB workspace, you can import it directly
into the System Identification Tool GUI. If you prefer to work at the command
line, you must represent the data as a System Identification Toolbox data
object instead.

In the System Identification Tool GUI. When you import scalar or

multiple-output time series data into the GUI, leave the Input field empty.
For more information about importing data, see “Importing Data into the
GUI” on page 2-14.

At the command line. To represent a time series vector or a matrix s as an

iddata object, use the following syntax:

y = iddata(s,[],Ts);

s contains as many columns as there are measured outputs. For time-domain

data, set Ts to the sampling interval. For continuous-time frequency domain
data, set Ts to 0.

6-3
 6 Time Series Identification

Estimating Time-Series Power Spectra

In this section...
“How to Estimate Time-Series Power Spectra Using the GUI” on page 6-4
“How to Estimate Time-Series Power Spectra at the Command Line” on
page 6-5

How to Estimate Time-Series Power Spectra Using

the GUI
You must have already imported your data into the GUI, as described in
“Preparing Time-Series Data” on page 6-3.

To estimate time-series spectral models in the System Identification Tool GUI:

1 In the System Identification Tool GUI, select Estimate > Spectral

models to open the Spectral Model dialog box.

2 In the Method list, select the spectral analysis method you want to use.
For information about each method, see “Options for Computing Spectral
Models” on page 3-5.

3 Specify the frequencies at which to compute the spectral model in either of

the following ways:
• In the Frequencies field, enter either a vector of values, a MATLAB
expression that evaluates to a vector, or a variable name of a vector in
the MATLAB workspace. For example, logspace(-1,2,500).
• Use the combination of Frequency Spacing and Frequencies to
construct the frequency vector of values:
– In the Frequency Spacing list, select Linear or Logarithmic
frequency spacing.

Note For etfe, only the Linear option is available.

– In the Frequencies field, enter the number of frequency points.

6-4
 Estimating Time-Series Power Spectra

For time-domain data, the frequency ranges from 0 to the Nyquist

frequency. For frequency-domain data, the frequency ranges from the
smallest to the largest frequency in the data set.

4 In the Frequency Resolution field, enter the frequency resolution, as

described in “Options for Frequency Resolution” on page 3-6. To use the
default value, enter default or leave the field empty.

5 In the Model Name field, enter the name of the correlation analysis model.
The model name should be unique in the Model Board.

6 Click Estimate to add this model to the Model Board in the System
Identification Tool GUI.

7 In the Spectral Model dialog box, click Close.

8 To view the estimated disturbance spectrum, select the Noise spectrum

check box in the System Identification Tool GUI. For more information
about working with this plot, see “Noise Spectrum Plots” on page 8-46.

To export the model to the MATLAB workspace, drag it to the To Workspace

rectangle in the System Identification Tool GUI. You can view the power
spectrum and the confidence intervals of the resulting idfrd model object
using the bode command.

How to Estimate Time-Series Power Spectra at the

Command Line
You can use the etfe, spa, and spafdr commands to estimate power spectra
of time series for both time-domain and frequency-domain data. The following
table provides a brief description of each command.

You must have already prepared your data, as described in “Preparing

Time-Series Data” on page 6-3.

The resulting models are stored as an idfrd model object, which contains
SpectrumData and its variance. For multiple-output data, SpectrumData
contains power spectra of each output and the cross-spectra between each
output pair.

6-5
 6 Time Series Identification

Estimating Frequency Response of Time Series

Command Description
etfe Estimates a periodogram using Fourier
analysis.
spa Estimates the power spectrum with its
standard deviation using spectral analysis.
spafdr Estimates the power spectrum with
its standard deviation using a variable
frequency resolution.

For example, suppose y is time-series data. The following commands estimate

the power spectrum g and the periodogram p, and plot both models with three
standard deviation confidence intervals:

g = spa(y)
p = etfe(y)
bode(g,p,'sd',3)

For detailed information about these commands, see the corresponding

reference pages.

6-6
 Estimating AR and ARMA Models

Estimating AR and ARMA Models

In this section...
“Definition of AR and ARMA Models” on page 6-7
“Estimating Polynomial Time-Series Models in the GUI” on page 6-7
“Estimating AR and ARMA Models at the Command Line” on page 6-10

Definition of AR and ARMA Models

For a single-output signal y(t), the AR model is given by the following equation:

A(q) y(t) = e(t)

The AR model is a special case of the ARX model with no input.

The ARMA model for a single-output time-series is given by the following

equation:

A(q) y(t) = C(q) e(t)

The ARMA structure reduces to the AR structure for C(q)=1. The ARMA
model is a special case of the ARMAX model with no input.

For more information about polynomial models, see “What Are Black-Box
Polynomial Models?” on page 3-39.

Estimating Polynomial Time-Series Models in the GUI

Before you begin, you must have accomplished the following:

• Prepared the data, as described in “Preparing Time-Series Data” on page

6-3
• Estimated model order, as described in “Preliminary Step – Estimating
Model Orders and Input Delays” on page 3-48

6-7
 6 Time Series Identification

• (Multiple-output AR models only) Specified the model-order matrix in

the MATLAB workspace before estimation, as described in “Options for
Multiple-Input and Multiple-Output ARX Orders” on page 3-64

To estimate AR and ARMA models using the System Identification Tool GUI:

1 In the System Identification Tool GUI, select Estimate > Linear

parametric models to open the Linear Parametric Models dialog box.

2 In the Structure list, select the polynomial model structure you want to
estimate from the following options:
• AR:[na]
• ARMA:[na nc]

This action updates the options in the Linear Parametric Models dialog box
to correspond with this model structure. For information about each model
structure, see “Definition of AR and ARMA Models” on page 6-7.

Note OE and BJ structures are not available for time-series models.

3 In the Orders field, specify the model orders, as follows:

• For single-output models. Enter the model orders according to the

sequence displayed in the Structure field.
• For multiple-output ARX models. (AR models only) Enter the
model orders directly, as described in “Options for Multiple-Input and
Multiple-Output ARX Orders” on page 3-64. Alternatively, enter the
name of the matrix NA in the MATLAB Workspace browser that stores
model orders, which is Ny-by-Ny.

Tip To enter model orders and delays using the Order Editor dialog box,
click Order Editor.

6-8
 Estimating AR and ARMA Models

4 (AR models only) Select the estimation Method as ARX or IV

(instrumental variable method). For more information about these
methods, see “Algorithms for Estimating Polynomial Models” on page 3-66.

Note IV is not available for multiple-output data.

5 In the Name field, edit the name of the model or keep the default. The
name of the model should be unique in the Model Board.

6 In the Initial state list, specify how you want the algorithm to treat initial
states. For more information about the available options, see “Options for
Initial States” on page 3-66.

Tip If you get an inaccurate fit, try setting a specific method for handling
initial states rather than choosing it automatically.

7 In the Covariance list, select Estimate if you want the algorithm to

compute parameter uncertainties. Effects of such uncertainties are
displayed on plots as model confidence regions.

To omit estimating uncertainty, select None. Skipping uncertainty

computation might reduce computation time for complex models and large
data sets.

8 (ARMA only) To view the estimation progress at the command line, select
the Trace check box. During estimation, the following information is
displayed for each iteration:
• Loss function — Equals the determinant of the estimated covariance
matrix of the input noise.
• Parameter values — Values of the model structure coefficients you
specified.
• Search direction — Changes in parameter values from the previous
iteration.

6-9
 6 Time Series Identification

• Fit improvements — Shows the actual versus expected improvements in

the fit.

9 Click Estimate to add this model to the Model Board in the System
Identification Tool GUI.

10 (Prediction-error method only) To stop the search and save the results
after the current iteration has been completed, click Stop Iterations. To
continue iterations from the current model, click the Continue iter button
to assign current parameter values as initial guesses for the next search.

11 To plot the model, select the appropriate check box in the Model Views
area of the System Identification Tool GUI. For more information about
validating models, see Chapter 8, “Model Analysis”.

You can export the model to the MATLAB workspace for further analysis
by dragging it to the To Workspace rectangle in the System Identification
Tool GUI.

Estimating AR and ARMA Models at the Command

Line
You can estimate AR and ARMA models at the command line. For
single-output time-series, the resulting models are idpoly model objects. For
multiple-output time-series, the resulting models are idarx model objects.
For more information about models objects, see “Creating Model Structures at
the Command Line” on page 1-14.

The following table summarizes the commands and specifies whether

single-output or multiple-output models are supported.

6-10
 Estimating AR and ARMA Models

Commands for Estimating Polynomial Time-Series Models

Method Name Description Supported Data

ar Noniterative, least-squares method Time-domain, time-series iddata
to estimate linear, discrete-time data object.
single-output AR models.
armax Iterative prediction-error method Time-domain, time-series iddata
to estimate linear, single-output data object.
ARMAX models.
arx Noniterative, least-squares method Supports time- and frequency-domain
for estimating single-output and time-series iddata data.
multiple-output linear AR models.
ivar Noniterative, instrumental variable Supports time-domain, time-series
method for estimating single-output iddata data.
AR models.

The following code shows usage examples for estimating AR models:

% For scalar signals

m = ar(y,na)
% For multiple-output vector signals
m = arx(y,na)
% Instrumental variable method
m = ivar(y,na)
% For ARMA, do not need to specify nb and nk
th = armax(y,[na nc])

The ar command provides additional options to let you choose the algorithm
for computing the least-squares from a group of several popular techniques
from the following methods:

• Burg (geometric lattice)

• Yule-Walker
• Covariance

For more information about validating models, see “Overview of Model

Validation and Plots” on page 8-2.

6-11
 6 Time Series Identification

Estimating State-Space Time-Series Models

In this section...
“Definition of State-Space Time-Series Model” on page 6-12
“Estimating State-Space Models at the Command Line” on page 6-12

Definition of State-Space Time-Series Model

The discrete-time state-space model for a time series is given by the following
equations:

x(kT + T ) = Ax(kT ) + Ke(kT )

y(kT ) = Cx(kT ) + e(kT )

where T is the sampling interval and y(kT) is the output at time instant kT.

The time-series structure corresponds to the general structure with empty B

and D matrices.

For information about general discrete-time and continuous-time structures

for state-space models, see “What Are State-Space Models?” on page 3-73.

Estimating State-Space Models at the Command Line

You can estimate single-output and multiple-output state-space models at the
command line for time-domain and frequency-domain data (iddata object).

The following table provides a brief description of each command. The

resulting models are idss model objects.

6-12
 Estimating State-Space Time-Series Models

Commands for Estimating State-Space Time-Series Models

Command Description
n4sid Noniterative subspace method for estimating
discrete-time linear state-space models.

Note When you use pem to estimate a state-space

model, n4sid creates the initial model.

pem Estimates linear, discrete-time time-series

models using an iterative estimation method that
minimizes the prediction error.

6-13
 6 Time Series Identification

Example – Identifying Time-Series Models at the Command

Line
The following example simulates a time-series model, compares spectral
estimates, covariance estimates, and predicts output of the model:

ts0 = idpoly([1 -1.5 0.7],[]);

ir = sim(ts0,[1;zeros(24,1)]);
% Define the true covariance function
Ry0 = conv(ir,ir(25:-1:1));
e = idinput(200,'rgs');
% Define y vector
y = sim(ts0,e);
% iddata object with sampling time 1
y = iddata(y)
plot(y)
per = etfe(y);
speh = spa(y);
ffplot(per,speh,ts0)
% Estimate a second-order AR model
ts2 = ar(y,2);
ffplot(speh,ts2,ts0,'sd',3)
% Get covariance function estimates
Ryh = covf(y,25);
Ryh = [Ryh(end:-1:2),Ryh]';
ir2 = sim(ts2,[1;zeros(24,1)]);
Ry2 = conv(ir2,ir2(25:-1:1));
plot([-24:24]'*ones(1,3),[Ryh,Ry2,Ry0])
% The prediction ability of the model
compare(y,ts2,5)

6-14
 Estimating Nonlinear Models for Time-Series Data

Estimating Nonlinear Models for Time-Series Data

When a linear model provides an insufficient description of the dynamics,
you can try estimating a nonlinear models. To learn more about when to
estimate nonlinear models, see “Building Models from Data” in the Getting
Started Guide.

Before you can estimate models for time-series data, you must have already
prepared the data as described in “Preparing Time-Series Data” on page 6-3.

For black-box modeling of time-series data, the toolbox supports nonlinear

ARX models. To learn how to estimate this type of model, see “Identifying
Nonlinear ARX Models” on page 4-8.

If you understand the underlying physics of the system, you can specify an
ordinary differential or difference equation and estimate the coefficients. To
learn how to estimate this type of model, see “Estimating Nonlinear Grey-Box
Models” on page 5-16.

For more information about validating models, see “Overview of Model

Validation and Plots” on page 8-2.

6-15
 6 Time Series Identification

6-16
 7

Recursive Model
Identification

• “What Is Recursive Estimation?” on page 7-2

• “Commands for Recursive Estimation” on page 7-3
• “Algorithms for Recursive Estimation” on page 7-6
• “Data Segmentation” on page 7-14
 7 Recursive Model Identification

What Is Recursive Estimation?

Many real-world applications, such as adaptive control, adaptive filtering, and
adaptive prediction, require a model of the system to be available online while
the system is in operation. Estimating models for batches of input-output
data is useful for addressing the following types of questions regarding
system operation:

• Which input should be applied at the next sampling instant?

• How should the parameters of a matched filter be tuned?
• What are the predictions of the next few outputs?
• Has a failure occurred? If so, what type of failure?

You might also use online models to investigate time variations in system
and signal properties.

The methods for computing online models are called recursive identification
methods. Recursive algorithms are also called recursive parameter estimation,
adaptive parameter estimation, sequential estimation, and online algorithms.

For examples of recursive estimation and data segmentation, run the

Recursive Estimation and Data Segmentation demo by typing the following
command at the prompt:

iddemo5

For detailed information about recursive parameter estimation algorithms,

see the corresponding chapter in System Identification: Theory for the User by
Lennart Ljung (Prentice Hall PTR, Upper Saddle River, NJ, 1999).

7-2
 Commands for Recursive Estimation

Commands for Recursive Estimation

You can recursively estimate linear polynomial models, such as ARX,
ARMAX, Box-Jenkins, and Output-Error models. If you are working with
time-series data that contains no inputs and a single output, you can
estimate AR (autoregressive) and ARMA (autoregressive and moving average)
single-output models.

Before estimating models using recursive algorithms, you must import your
data into the MATLAB workspace and represent your data in either of the
following formats:

• Matrix of the form [y u]. y represents the output data using one or more
column vectors. Similarly, u represents the input data using one or more
column vectors.
• iddata or idfrd object. For more information about creating these objects,
see Chapter 2, “Data Import and Processing”.

The general syntax for recursive estimation commands is as follows:

[params,y_hat]=command(data,nn,adm,adg)

params matrix contains the values of the estimated parameters, where the
kth row contains the parameters associated with time k, which are computed
using the data values in the rows up to and including the row k.

y_hat contains the predicted output values such that the kth row of y_hat is
computed based on the data values in the rows up to and including the row k.

Tip y_hat contains the adaptive predictions of the output and is useful for
adaptive filtering applications, such as noise cancelation.

nn specified the model orders and delay according to the specific polynomial
structure of the model. For example, nn=[na nb nk] for ARX models. For
more information about specifying polynomial model orders and delays, see
“Identifying Input-Output Polynomial Models” on page 3-39.

7-3
 7 Recursive Model Identification

adm and adg specify any of the four recursive algorithm, as described in
“Algorithms for Recursive Estimation” on page 7-6.

The following table summarizes the recursive estimation commands

supported by the System Identification Toolbox product. The command
description indicates whether you can estimate single-input, single-output,
multiple-input, and multiple-output, and time-series (no input) models. For
details about each command, see the corresponding reference page.

Tip For ARX and AR models, use rarx. For single-input/single-output

ARMAX or ARMA, Box-Jenkins, and Output-Error models, use rarmax, rbj,
and roe, respectively.

Commands for Linear Recursive Estimation

Command Description
rarmax Estimate parameters of
single-input/single-output ARMAX
and ARMA models.
rarx Estimate parameters of single- or
multiple-input and single-output ARX
and AR models. Does not support
multiple-output system.
rbj Estimate parameters of
single-input/single-output Box-Jenkins
models.
roe Estimate parameters of
single-input/single-output Output-Error
models.

7-4
 Commands for Recursive Estimation

Commands for Linear Recursive Estimation (Continued)

Command Description
rpem Estimate parameters of multiple-input
and single-output ARMAX/ARMA,
Box-Jenkins, or Output-Error models
using the general recursive prediction-error
algorithm for estimating the parameter
gradient.

Note Unlike pem, rpem does not support

state-space models.

rplr Use as an alternative to rpem to estimate

parameters of multiple-input and
single-output systems when you want
to use recursive pseudolinear regression
method.

7-5
 7 Recursive Model Identification

Algorithms for Recursive Estimation

In this section...
“Types of Recursive Estimation Algorithms” on page 7-6
“General Form of Recursive Estimation Algorithm” on page 7-6
“Kalman Filter Algorithm” on page 7-8
“Forgetting Factor Algorithm” on page 7-10
“Unnormalized and Normalized Gradient Algorithms” on page 7-11

Types of Recursive Estimation Algorithms

You can choose from the following four recursive estimation algorithms:

• “General Form of Recursive Estimation Algorithm” on page 7-6

• “Kalman Filter Algorithm” on page 7-8
• “Forgetting Factor Algorithm” on page 7-10
• “Unnormalized and Normalized Gradient Algorithms” on page 7-11

You specify the type of recursive estimation algorithms as arguments adm

and adg of the recursive estimation commands in “Commands for Recursive
Estimation” on page 7-3.

For detailed information about these algorithms, see the corresponding

chapter in System Identification: Theory for the User by Lennart Ljung
(Prentice Hall PTR, Upper Saddle River, NJ, 1999).

General Form of Recursive Estimation Algorithm

The general recursive identification algorithm is given by the following
equation:

θˆ ( t ) = θˆ ( t − 1) + K ( t ) ( y ( t ) − yˆ ( t ) )

7-6
 Algorithms for Recursive Estimation

θ̂ ( t ) is the parameter estimate at time t. y(t) is the observed output at time t

and ŷ ( t ) is the prediction of y(t) based on observations up to time t-1. The
gain, K(t), determines how much the current prediction error y ( t ) − yˆ ( t )
affects the update of the parameter estimate. The estimation algorithms
minimize the prediction-error term y ( t ) − yˆ ( t ) .

The gain has the following general form:

K (t) = Q (t) ψ (t)

The recursive algorithms supported by the System Identification Toolbox

product differ based on different approaches for choosing the form of Q(t) and
computing ψ ( t ) , where ψ ( t ) represents the gradient of the predicted model
output yˆ ( t | θ ) with respect to the parameters θ .

The simplest way to visualize the role of the gradient ψ ( t ) of the parameters,
is to consider models with a linear-regression form:

y ( t ) = ψ T ( t ) θ0 ( t ) + e ( t )

In this equation, ψ ( t ) is the regression vector that is computed based on

previous values of measured inputs and outputs. θ0 ( t ) represents the true
parameters. e(t) is the noise source (innovations), which is assumed to be
white noise. The specific form of ψ ( t ) depends on the structure of the
polynomial model.

For linear regression equations, the predicted output is given by the following
equation:

yˆ ( t ) = ψ T ( t ) θˆ ( t − 1)

7-7
 7 Recursive Model Identification

For models that do not have the linear regression form, it is not possible to
compute exactly the predicted output and the gradient ψ ( t ) for the current

parameter estimate θ̂ ( t − 1) . To learn how you can compute approximation for

ψ ( t ) and θ̂ ( t − 1) for general model structures, see the section on recursive

prediction-error methods in System Identification: Theory for the User by
Lennart Ljung (Prentice Hall PTR, Upper Saddle River, NJ, 1999).

Kalman Filter Algorithm

• “Mathematics of the Kalman Filter Algorithm” on page 7-8
• “Using the Kalman Filter Algorithm” on page 7-9

Mathematics of the Kalman Filter Algorithm

The following set of equations summarizes the Kalman filter adaptation
algorithm:

θˆ ( t ) = θˆ ( t − 1) + K ( t ) ( y ( t ) − yˆ ( t ) )

yˆ ( t ) = ψ T ( t ) θˆ ( t − 1)

K (t) = Q (t) ψ (t)

P ( t − 1)
Q (t) =
R2 + ψ ( t ) P ( t − 1) ψ ( t )
T

P ( t − 1) ψ ( t ) ψ ( t ) P ( t − 1)
T
P ( t ) = P ( t − 1) + R1 −
R2 + ψ ( t ) P ( t − 1) ψ ( t )
T

This formulation assumes the linear-regression form of the model:

y ( t ) = ψ T ( t ) θ0 ( t ) + e ( t )

7-8
 Algorithms for Recursive Estimation

The Kalman filter is used to obtain Q(t).

This formulation also assumes that the true parameters θ0 ( t ) are described
by a random walk:

θ0 ( t ) = θ0 ( t − 1) + w ( t )

w(t) is Gaussian white noise with the following covariance matrix, or drift
matrix R1:

Ew ( t ) w ( t )
T
= R1

R2 is the variance of the innovations e(t) in the following equation:

y ( t ) = ψ T ( t ) θ0 ( t ) + e ( t )

The Kalman filter algorithm is entirely specified by the sequence of data y(t),
the gradient ψ ( t ) , R1, R2, and the initial conditions θ ( t = 0 ) (initial guess of
the parameters) and P ( t = 0 ) (covariance matrix that indicates parameters
errors).

Note To simplify the inputs, you can scale R1, R2, and P ( t = 0 ) of the
original problem by the same value such that R2 is equal to 1. This scaling
does not affect the parameters estimates.

Using the Kalman Filter Algorithm

The general syntax for the command described in “Algorithms for Recursive
Estimation” on page 7-6 is the following:

[params,y_hat]=command(data,nn,adm,adg)

7-9
 7 Recursive Model Identification

To specify the Kalman filter algorithm, set adm to 'kf' and adg to the value
of the drift matrix R1 (described in “Mathematics of the Kalman Filter
Algorithm” on page 7-8).

Forgetting Factor Algorithm

• “Mathematics of the Forgetting Factor Algorithm” on page 7-10
• “Using the Forgetting Factor Algorithm” on page 7-11

Mathematics of the Forgetting Factor Algorithm

The following set of equations summarizes the forgetting factor adaptation
algorithm:

θˆ ( t ) = θˆ ( t − 1) + K ( t ) ( y ( t ) − yˆ ( t ) )

yˆ ( t ) = ψ T ( t ) θˆ ( t − 1)

K (t) = Q (t) ψ (t)

P ( t − 1)
Q (t) = P (t) =
λ + ψ ( t ) P ( t − 1) ψ ( t )
T

⎛ P ( t − 1) ψ ( t ) ψ ( t ) P ( t − 1) ⎞
T
P (t) = λ⎜ (
)
1 ⎜ P t −1 − ⎟
λ + ψ ( t ) P ( t − 1) ψ ( t ) ⎠⎟
T
⎝

To obtain Q(t), the following function is minimized at time t:

∑ k=1 λt−k e2 ( k)
t

This approach discounts old measurements exponentially such that an

observation that is τ samples old carries a weight that is equal to λ τ times
the weight of the most recent observation. τ = 1 1−λ represents the memory

7-10
 Algorithms for Recursive Estimation

horizon of this algorithm. Measurements older than τ = 1 1−λ typically carry a

weight that is less than about 0.3.

λ is called the forgetting factor and typically has a positive value between
0.97 and 0.995.

Note In the linear regression case, the forgetting factor algorithm is

known as the recursive least-squares (RLS) algorithm. The forgetting factor
algorithm for λ = 1 is equivalent to the Kalman filter algorithm with R1=0
and R2=1. For more information about the Kalman filter algorithm, see
“Kalman Filter Algorithm” on page 7-8.

Using the Forgetting Factor Algorithm

The general syntax for the command described in “Algorithms for Recursive
Estimation” on page 7-6 is the following:

[params,y_hat]=command(data,nn,adm,adg)

To specify the forgetting factor algorithm, set adm to 'ff' and adg to the
value of the forgetting factor λ (described in “Mathematics of the Forgetting
Factor Algorithm” on page 7-10).

Tip λ typically has a positive value from 0.97 to 0.995.

Unnormalized and Normalized Gradient Algorithms

• “Mathematics of the Unnormalized and Normalized Gradient Algorithm”
on page 7-12
• “Using the Unnormalized and Normalized Gradient Algorithms” on page
7-12

7-11
 7 Recursive Model Identification

Mathematics of the Unnormalized and Normalized Gradient

Algorithm
In the linear regression case, the gradient methods are also known as the
least mean squares (LMS) methods.

The following set of equations summarizes the unnormalized gradient and

normalized gradient adaptation algorithm:

θˆ ( t ) = θˆ ( t − 1) + K ( t ) ( y ( t ) − yˆ ( t ) )

yˆ ( t ) = ψ T ( t ) θˆ ( t − 1)

K (t) = Q (t) ψ (t)

In the unnormalized gradient approach, Q(t) is the product of the gain γ

and the identity matrix:

Q ( t ) = γI

In the normalized gradient approach, Q(t) is the product of the gain γ , and
the identity matrix is normalized by the magnitude of the gradient ψ ( t ) :

γ
Q (t) = 2
I
ψ (t)

These choices of Q(t) update the parameters in the negative gradient direction,
where the gradient is computed with respect to the parameters.

Using the Unnormalized and Normalized Gradient Algorithms

The general syntax for the command described in “Algorithms for Recursive
Estimation” on page 7-6 is the following:

[params,y_hat]=command(data,nn,adm,adg)

7-12
 Algorithms for Recursive Estimation

To specify the unnormalized gain algorithm, set adm to 'ug' and adg to the
value of the gain γ (described in “Mathematics of the Unnormalized and
Normalized Gradient Algorithm” on page 7-12).

To specify the normalized gain algorithm, set adm to 'ng' and adg to the
value of the gain γ .

7-13
 7 Recursive Model Identification

Data Segmentation
For systems that exhibit abrupt changes while the data is being collected,
you might want to develop models for separate data segments such that the
system does not change during a particular data segment. Such modeling
requires identification of the time instants when the changes occur in the
system, breaking up the data into segments according to these time instants,
and identification of models for the different data segments.

The following cases are typical applications for data segmentation:

• Segmentation of speech signals, where each data segment corresponds

to a phonem.
• Detection of trend breaks in time series.
• Failure detection, where the data segments correspond to operation with
and without failure.
• Estimating different working modes of a system.

Use segment to build polynomial models, such as ARX, ARMAX, AR, and
ARMA, so that the model parameters are piece-wise constant over time. For
detailed information about this command, see the corresponding reference
page.

To see an example of using data segmentation, run the Recursive Estimation

and Data Segmentation demonstration by typing to the following command at
the prompt:

iddemo5

7-14
 8

Model Analysis

• “Overview of Model Validation and Plots” on page 8-2

• “Simulating and Predicting Model Output” on page 8-8
• “Residual Analysis” on page 8-21
• “Impulse and Step Response Plots” on page 8-30
• “Frequency Response Plots” on page 8-38
• “Noise Spectrum Plots” on page 8-46
• “Pole and Zero Plots” on page 8-53
• “Akaike’s Criteria for Model Validation” on page 8-58
• “Computing Model Uncertainty” on page 8-61
• “Troubleshooting Models” on page 8-64
• “Next Steps After Getting an Accurate Model” on page 8-69
 8 Model Analysis

Overview of Model Validation and Plots

In this section...
“When to Validate Models” on page 8-2
“Ways to Validate Models” on page 8-2
“Data for Validating Models” on page 8-3
“Supported Model Plots” on page 8-4
“Plotting Models in the GUI” on page 8-5
“Getting Advice About Models” on page 8-7

When to Validate Models

After estimating each model, you can validate whether the model reproduces
system behavior within acceptable bounds. You iterate between estimation
and validation until you find the simplest model that best captures the
system dynamics.

For ideas on how to adjust your modeling strategy based on validation results,
see “Troubleshooting Models” on page 8-64.

Tip If you have installed the Control System Toolbox product, you can also
view models using the LTI Viewer. For more information, see “Viewing Model
Response Using the LTI Viewer” on page 9-5.

Ways to Validate Models

You can use the following approaches to validate models:

• Comparing simulated or predicted model output to measured output.

See “Simulating and Predicting Model Output” on page 8-8.
To simulate identified models in the Simulink environment, see
“Simulating Model Output” on page 10-6.
• Analyzing autocorrelation and cross-correlation of the residuals with input.

8-2
 Overview of Model Validation and Plots

See “Residual Analysis” on page 8-21.

• Analyzing model response. For more information, see the following:
- “Impulse and Step Response Plots” on page 8-30
- “Frequency Response Plots” on page 8-38
For information about the response of the noise model, see “Noise Spectrum
Plots” on page 8-46.
• Plotting the poles and zeros of the linear parametric model.
For more information, see “Pole and Zero Plots” on page 8-53.
• Comparing the response of nonparametric models, such as impulse-, step-,
and frequency-response models, to parametric models, such as linear
polynomial models, state-space model, and nonlinear parametric models.

Note Do not use this comparison when feedback is present in the system
because feedback makes nonparametric models unreliable. To test if
feedback is present in the system, use the advice command on the data.

• Compare models using Akaike Information Criterion or Akaike Final

Prediction Error.
For more information, see the aic and fpe reference page.
• Plotting linear and nonlinear blocks of Hammerstein-Wiener and nonlinear
ARX models.
For more information, see Chapter 4, “Nonlinear Black-Box Model
Identification”.

Displaying confidence intervals on supported plots helps you assess the

uncertainty of model parameters. For more information, see “Computing
Model Uncertainty” on page 8-61.

Data for Validating Models

For plots that compare model response to measured response, such as model
output and residual analysis plots, you designate two types of data sets: one
for estimating the models (estimation data), and the other for validating the

8-3
 8 Model Analysis

models (validation data). Although you can designate the same data set to be
used for estimating and validating the model, you risk overfitting your data.
When you validate a model using an independent data set, this process is
called cross-validation.

Note Validation data should be the same in frequency content as the

estimation data. If you detrended the estimation data, you must remove the
same trend from the validation data. For more information about detrending,
see “Handling Offsets and Trends in Data” on page 2-97.

Supported Model Plots

The following table summarizes the types of supported model plots.

Plot Type Supported Models Learn More

Model Output All linear and nonlinear “Simulating and
models Predicting Model
Output” on page 8-8
Residual Analysis All linear and nonlinear “Residual Analysis” on
models page 8-21
Transient Response • All linear parametric “Impulse and Step
models Response Plots” on
page 8-30
• Correlation analysis
(nonparametric)
models
• For nonlinear
models, only step
response.
Frequency Response • All linear parametric “Frequency Response
models Plots” on page 8-38
• Spectral analysis
(nonparametric)
models

8-4
 Overview of Model Validation and Plots

Plot Type Supported Models Learn More

Noise Spectrum • All linear parametric “Noise Spectrum Plots”
models on page 8-46
• Spectral analysis
(nonparametric)
models
Poles and Zeros All linear parametric “Pole and Zero Plots” on
models page 8-53
Nonlinear ARX Nonlinear ARX models Nonlinear ARX Plots
only
Hammerstein-Wiener Hammerstein-Wiener Hammerstein-Wiener
models only Plots

Plotting Models in the GUI

To create one or more plots of your models, select the corresponding check box
in the Model Views area of the System Identification Tool GUI. An active
model icon has a thick line in the icon, while an inactive model has a thin line.
Only active models appear on the selected plots.

To include or exclude a model on a plot, click the corresponding icon in the

System Identification Tool GUI. Clicking the model icon updates any plots
that are currently open.

8-5
 8 Model Analysis

For example, in the following figure, Model output is selected. In this case,
the models n4s4 is not included on the plot because only arx441 is active.

Active model

Inactive model

Plots the model

output of active
models.

Plots Include Only Active Models

To close a plot, clear the corresponding check box in the System Identification
Tool GUI.

Tip To get information about a specific plot, select a help topic from the Help
menu in the plot window.

For general information about working with plots in the System Identification
Toolbox product , see “Working with Plots in the System Identification Tool
GUI” on page 11-15.

8-6
 Overview of Model Validation and Plots

Getting Advice About Models

Use the advice command on an estimated model to answer the following
questions about the model:

• Should I increase or decrease the model order?

• Should I estimate a noise model?
• Is feedback present?

8-7
 8 Model Analysis

Simulating and Predicting Model Output

In this section...
“Using a Model for Simulation and Prediction” on page 8-8
“Simulation and Prediction in the GUI” on page 8-11
“Simulation and Prediction at the Command Line” on page 8-16

Using a Model for Simulation and Prediction

The primary use of a model is to calculate the output (y(t)) for given input
values. This calculation is called simulation.

Simulating a model means computing its response using input data and
initial conditions. The time samples of the model response match the time
samples of the input data used for simulation.

For a continuous-time system, simulation means solving a differential

equation. For a discrete-time system, simulation means directly applying
the model equations.

For example, consider a dynamic model described by a first-order difference

equation that uses a sampling interval of 1 second:

y(t) + ay(t–1) = bu(t–1),

where y is the output and u is the input. For parameter values a = –0.9 and b
= 1.5, the equation becomes:

y(t) – 0.9y(t–1) = 1.5u(t–1).

Suppose you want to compute the values y(1), y(2), y(3),... for given input
values u(0) = 2, u(1) = 1, u(2) = 4,...Here, y(1) is the value of output at the first
sampling instant. Using initial condition of y(0) = 0, the values of y(t) for
times t = 1, 2 and 3 can be computed as:

8-8
 Simulating and Predicting Model Output

y(1) = 0.9y(0) + 1.5u(0) = 0.9*0 + 1.5*2 = 3

y(2) = 0.9y(1) + 1.5u(1) = 0.9*3 + 1.5*1 = 4.2

y(3) = 0.9y(2) + 1.5u(2) = 0.9*4.2 + 1.5*4 = 9.78

...

A model also lets you compute a qualified guess of future output values based
on past observations of system’s inputs and outputs. This is called prediction.

Prediction forecasts the model response k steps ahead into the future using
the current and past values of measured input and output values. k is called
the prediction horizon, and corresponds to predicting output at time kTs,
where Ts is the sampling interval.

For example, suppose you use sensors to measure the input signal u(t) and
output signal y(t) of the physical system, described in the previous first-order
equation. At the tenth sampling instant (t = 10), the output y(10) is 16 mm
and the corresponding input u(10) is 12 N. Now, you want to predict the value
of the output at the future time t = 11. Using the previous equation:

y(11) = 0.9y(10) + 1.5u(10)

Hence, the predicted value of future output y(11) at time t = 10 is:

y(11) = 0.9*16 + 1.5*12 = 32.4

In general, to predict the model response k steps into the future (k≥1) from
the current time t, you should know the inputs up to time t+k and outputs
up to time t:

yp(t+k) = f(u(t+k),u(t+k–1),...,u(t),u(t–1),...,u(0)
y(t),y(t–1),y(t–2),...,y(0))

u(0) and y(0) are the initial states. f() represents the predictor, which is a
dynamic model whose form depends on the model structure. For example, the
one-step-ahead predictor yp of the model y(t) + ay(t–1) = bu(t) is:

yp(t+1) = –ay(t) + bu(t+1)

8-9
 8 Model Analysis

The difference between prediction and simulation is that in prediction, the

past values of outputs used for calculation are measured values while in
simulation the outputs are themselves a result of calculation using inputs
and initial conditions.

The way information in past outputs is used depends on the disturbance

1
model H of the model. For the previous dynamic model, H ( z) = . In
1 + az−1
models of Output-Error (OE) structure (H(z) = 1), there is no information in
past outputs that can be used for predicting future output values. In this
case, predictions and simulations coincide. For state-space models (idss),
output-error structure corresponds to models with K=0. For polynomial
models (idpoly), this corresponds to models with polynomials a=c=d=1.

Note Prediction with k=∞ means that no previous outputs are used in the
computation and prediction returns the same result as simulation.

Both simulation and prediction require initial conditions, which correspond to

the states of the model at the beginning of the simulation or prediction.

Tip If you do not know the initial conditions and have input and output
measurements available, you can estimate the initial condition using this
toolbox.

You can validate linear parametric models and nonlinear models by checking
how well the simulated or predicted output of the model matches the
measured output. You can use either time or frequency domain data for
simulation or prediction. For frequency domain data, the simulation and
prediction results are products of the Fourier transform of the input and
frequency function of the model.

Simulation provides a better validation test for the model than prediction.
However, how you validate the model output should match how you plan to
use the model. For example, if you plan to use your model for control design,
you can validate the model by predicting its response over a time horizon that
represents the dominating time constants of the model.

8-10
 Simulating and Predicting Model Output

See Also
“Simulation and Prediction in the GUI” on page 8-11

“Simulation and Prediction at the Command Line” on page 8-16

Chapter 10, “System Identification Toolbox Blocks”—Simulate models in

Simulink software.

Simulation and Prediction in the GUI

• “How to Plot Model Output” on page 8-11
• “Interpreting the Model Output Plot” on page 8-12
• “Changing Model Output Plot Settings” on page 8-13
• “Definition: Confidence Interval” on page 8-15

How to Plot Model Output

To create a model output plot for parametric linear and nonlinear models
in the System Identification Tool GUI, select the Model output check box
in the Model Views area. By default, this operation estimates the initial
states from the data.

To include or exclude a model on the plot, click the corresponding model icon
in the System Identification Tool GUI. Active models display a thick line
inside the Model Board icon.

To learn how to interpret the model output plot, see “Interpreting the Model
Output Plot” on page 8-12.

To change plot settings, see “Changing Model Output Plot Settings” on page
8-13.

For general information about creating and working with plots, see “Working
with Plots in the System Identification Tool GUI” on page 11-15.

8-11
 8 Model Analysis

Interpreting the Model Output Plot

The following figure shows a sample Model Output plot, created in the System
Identification Tool GUI.

The model output plot shows different information depending on the domain
of the input-output validation data, as follows:

• For time-domain validation data, the plot shows simulated or predicted

model output.
• For frequency-domain data, the plot shows the amplitude of the model
response to the frequency-domain input signal. The model response is
equal to the product of the Fourier transform of the input and the model’s
frequency function.
• For frequency-response data, the plot shows the amplitude of the model
frequency response.

8-12
 Simulating and Predicting Model Output

For linear models, you can estimate a model using time-domain data, and
then validate the model using frequency domain data. For nonlinear models,
you can only use time-domain data for both estimation and validation.

The right side of the plot displays the percentage of the output that the model
reproduces (Best Fit), computed using the following equation:

⎛ y − yˆ ⎞
Best Fit = ⎜⎜ 1 − ⎟⎟ × 100
⎝ y− y ⎠

In this equation, y is the measured output, ŷ is the simulated or predicted

model output, and y is the mean of y. 100% corresponds to a perfect fit,
and 0% indicates that the fit is no better than guessing the output to be a
constant ( ŷ = y ).

Because of the definition of Best Fit, it is possible for this value to be negative.
A negative best fit is worse than 0% and can occur for the following reasons:

• The estimation algorithm failed to converge.

• The model was not estimated by minimizing y − yˆ . Best Fit can

be negative when you minimized 1-step-ahead prediction during the
estimation, but validate using the simulated output ŷ .
• The validation data set was not preprocessed in the same way as the
estimation data set.

Changing Model Output Plot Settings

The following table summarizes the Model Output plot settings.

8-13
 8 Model Analysis

Model Output Plot Settings

Action Command
Display confidence intervals. • To display the dashed lines on
either side of the nominal model
curve, select Options > Show
Note Confidence intervals are only confidence intervals. Select
available for simulated model output this option again to hide the
of linear models. Confidence internal confidence intervals.
are not available for nonlinear ARX
• To change the confidence
and Hammerstein-Wiener models.
value, select Options > Set %
confidence level, and choose a
value from the list.
See “Definition: Confidence Interval”
on page 8-15. • To enter your own confidence
level, select Options > Set
confidence level > Other.
Enter the value as a probability
(between 0 and 1) or as the
number of standard deviations of
a Gaussian distribution.
Change between simulated output • Select Options > Simulated
or predicted output. output or Options > k step
ahead predicted output.
• To change the prediction horizon,
Note Prediction is only available
select Options > Set prediction
for time-domain validation data.
horizon, and select the number
of samples.
• To enter your own prediction
horizon, select Options > Set
prediction horizon > Other.
Enter the value in terms of the
number of samples.
Display the actual output values Select Options > Signal plot or
(Signal plot), or the difference Options > Error plot.
between model output and measured
output (Error plot).

8-14
 Simulating and Predicting Model Output

Model Output Plot Settings (Continued)

Action Command
(Time-domain validation data only) Select Options > Customized time
Set the time range for model output span for fit and enter the minimum
and the time interval for which the and maximum time values. For
Best Fit value is computed. example:

[1 20]
(Multiple-output system only) Select the output by name in the
Select a different output. Channel menu.

Definition: Confidence Interval

The confidence interval corresponds to the range of output values with a
specific probability of being the actual output of the system. The toolbox uses
the estimated uncertainty in the model parameters to calculate confidence
intervals and assumes the estimates have a Gaussian distribution.

For example, for a 95% confidence interval, the region around the nominal
curve represents the range of values that have a 95% probability of being the
true system response. You can specify the confidence interval as a probability
(between 0 and 1) or as the number of standard deviations of a Gaussian
distribution. For example, a probability of 0.99 (99%) corresponds to 2.58
standard deviations.

Note The calculation of the confidence interval assumes that the model
sufficiently describes the system dynamics and the model residuals pass
independence tests.

In the GUI, you can display a confidence interval on the plot to gain insight
into the quality of a linear model. To learn how to show or hide confidence
interval, see “Changing Model Output Plot Settings” on page 8-13.

8-15
 8 Model Analysis

Simulation and Prediction at the Command Line

• “Summary of Simulation and Prediction Commands” on page 8-16
• “Initial States in Simulation and Prediction” on page 8-17
• “Example – Simulating Model Output with Noise at the Command Line”
on page 8-19
• “Example – Simulating a Continuous-Time State-Space Model at the
Command Line” on page 8-19
• “Example – Predicting Time Series” on page 8-20

Summary of Simulation and Prediction Commands

Note If you estimated a linear model from detrended data and want to
simulate or predict the output at the original operation conditions, use the
retrend command to the simulated or predicted output.

Command Description Example

compare Use this command for To plot five-step-ahead
model validation to predicted output of the model
determine how closely mod against the validation
the simulated model data data, use the following
response matches command:
the measured output
signal . compare(data,mod,5)

Plots simulated or
predicted model
Note Omitting the third
output on top of the
argument assumes an
measured output.
infinite horizon and results in
You should use
simulation.
an independent
validation data set
as input to the model.

8-16
 Simulating and Predicting Model Output

Command Description Example

sim Simulate and plot the To simulate the response of
model output only. the model model using input
data data, use the following
command:

sim(model,data)
predict Predict and plot the To perform one-step-ahead
model output only. prediction of the response for
the model model and input
data data, use the following
command:

predict(model,data,1)

Use the following syntax

to compute k-step-ahead
prediction of the output signal
using model m:

yhat = predict(m,[y u],k)

Initial States in Simulation and Prediction

The process of computing simulated and predicted responses over a time
range starts by using the initial conditions to compute the first few output
values. Both sim and predict commands provide defaults for handling initial
conditions.

Simulation: Default initial conditions are zero for polynomial (idpoly and
idarx) models. For state-space (idss), low-order transfer functions (idproc),
and linear grey-box (idgrey) models, the default initial conditions are the
internal model initial states (model property x0). You can specify other initial
conditions using the InitialState argument in sim, as described in the next
example.

8-17
 8 Model Analysis

Use the compare command to validate models by simulation because its

algorithm estimates the initial states of a model to optimize the model fit to
a given data set.

If you use sim, the simulated and the measured responses might differ when
the initial conditions of the estimated model and the system that measured
the validation data set differ—especially at the beginning of the response. To
minimize this difference, estimate the initial state values from the data using
the findstates command and specify these initial states as input arguments
to the sim command. For example, to compute the initial states that optimize
the fit of the model m to the output data in z:

% Estimate the initial states

X0est = findstates(m,z);
% Simulate the response using estimated initial states
sim(m,z.InputData,'InitialState',X0est)

See Also: sim (for linear models), sim(idnlarx), sim(idnlgrey),

sim(idnlhw)

Prediction: Default initial conditions depend on the type of model. You can
specify other initial conditions using the InitialState argument in predict.
For example, to compute the initial states that optimize the 1-step-ahead
predicted response of the model m to the output data z:

[Yp,X0est] = predict(m,z,1,'InitialState','Estimate')

This command returns the estimated initial states as the output argument
X0est. For information about other ways to specify initials states, see the
predict reference page for the corresponding model type.

See Also: predict (for linear models), predict(idnlarx),

predict(idnlgrey), predict(idnlhw)

8-18
 Simulating and Predicting Model Output

Example – Simulating Model Output with Noise at the

Command Line
This example demonstrates how you can create input data and a model, and
then use the data and the model to simulate output data. In this case, you use
the following ARMAX model with Gaussian noise e:

y(t) − 1.5 y(t − 1) + 0.7 y(t − 2) =

u(t − 1) + 0.5u(t − 2) + e(t) − e(t − 1) + 0.2e(t − 1)

Create the ARMAX model and simulate output data with random binary
input u using the following commands:

% Create an ARMAX model

m_armax = idpoly([1 -1.5 0.7],[0 1 0.5],[1 -1 0.2]);
% Create a random binary input
u = idinput(400,'rbs',[0 0.3]);
% Simulate the output data
y = sim(m_armax,u,'noise');

Note The argument 'noise' specifies to include in the simulation the

Gaussian noise e present in the model. Omit this argument to simulate the
noise-free response to the input u, which is equivalent to setting e to zero.

Example – Simulating a Continuous-Time State-Space Model

at the Command Line
This example demonstrates how to simulate a continuous-time state-space
model using a random binary input u and a sampling interval of 0.1 s.

Consider the following state-space model:

⎡ −1 1 ⎤ ⎡ 1 ⎤ ⎡ 0 .5 ⎤
x = ⎢ ⎥ x + ⎢ ⎥u + ⎢ ⎥ e
⎣ −0.5 0 ⎦ ⎣ 0 .5 ⎦ ⎣ 0 .5 ⎦
y = [1 0] x + e

where e is Gaussian white noise with variance 7.

8-19
 8 Model Analysis

Use the following commands to simulate the model:

% Set up the model matrices

A = [-1 1;-0.5 0]; B = [1; 0.5];
C = [1 0]; D = 0; K = [0.5;0.5];
% Create a continuous-time state-space model
% Ts = 0 indicates continuous time
model_ss = idss(A,B,C,D,K,'Ts',0,'NoiseVariance',7)
% Create a random binary input
u = idinput(400,'rbs',[0 0.3]);
% Create an iddata object with empty output
data = iddata([],u);
data.ts = 0.1
% Simulate the output using the model
y=sim(model_ss,data,'noise');

Note The argument 'noise' specifies to simulate with the Gaussian noise e
present in the model. Omit this argument to simulate the noise-free response
to the input u, which is equivalent to setting e to zero.

Example – Predicting Time Series

You can evaluate how well a time-series model predicts future values. In this
example, y is the original series of monthly sales figures. You use the first
half of the measured data to estimate the time-series model and test the
model’s ability to forecast sales six months ahead using the entire data set.

% Select the first half of the data for estimation

% y1 = y(1:48)
% Estimate a fourth-order autoregressive model
% using the first half of the data.
m = ar(y1,4)
% Compute 6-step ahead prediction
yhat = predict(m,y,6)
% Plot the predicted and measured outputs
plot(y,yhat)

8-20
 Residual Analysis

Residual Analysis
In this section...
“What Is Residual Analysis?” on page 8-21
“Supported Model Types” on page 8-22
“What Residual Plots Show for Different Data Domains” on page 8-22
“Displaying the Confidence Interval” on page 8-23
“How to Plot Residuals Using the GUI” on page 8-24
“How to Plot Residuals at the Command Line” on page 8-26
“Example – Examining Model Residuals” on page 8-26

What Is Residual Analysis?

Residuals are differences between the one-step-predicted output from the
model and the measured output from the validation data set. Thus, residuals
represent the portion of the validation data not explained by the model.

Residual analysis consists of two tests: the whiteness test and the
independence test.

According to the whiteness test criteria, a good model has the residual
autocorrelation function inside the confidence interval of the corresponding
estimates, indicating that the residuals are uncorrelated.

According to the independence test criteria, a good model has residuals

uncorrelated with past inputs. Evidence of correlation indicates that the
model does not describe how part of the output relates to the corresponding
input. For example, a peak outside the confidence interval for lag k means
that the output y(t) that originates from the input u(t-k) is not properly
described by the model.

Your model should pass both the whiteness and the independence tests,
except in the following cases:

8-21
 8 Model Analysis

• For output-error (OE) models and when using instrumental-variable (IV)

methods, make sure that your model shows independence of e and u, and
pay less attention to the results of the whiteness of e.
In this case, the modeling focus is on the dynamics G and not the
disturbance properties H.
• Correlation between residuals and input for negative lags, is not necessarily
an indication of an inaccurate model.
When current residuals at time t affect future input values, there might
be feedback in your system. In the case of feedback, concentrate on the
positive lags in the cross-correlation plot during model validation.

Supported Model Types

You can validate parametric linear and nonlinear models by checking the
behavior of the model residuals. For a description of residual analysis, see
“What Residual Plots Show for Different Data Domains” on page 8-22.

Note For nonparametric models, including impulse-response, step-response,

and frequency-response models, residual analysis plots are not available. For
time-series models, you can only generate model-output plots for parametric
models using time-domain time-series (no input) measured data.

What Residual Plots Show for Different Data Domains

Residual analysis plots show different information depending on whether you
use time-domain or frequency-domain input-output validation data.

For time-domain validation data, the plot shows the following two axes:

• Autocorrelation function of the residuals for each output

• Cross-correlation between the input and the residuals for each input-output
pair

Note For time-series models, the residual analysis plot does not provide
any input-residual correlation plots.

8-22
 Residual Analysis

For frequency-domain validation data, the plot shows the following two axes:

• Estimated power spectrum of the residuals for each output

• Transfer-function amplitude from the input to the residuals for each
input-output pair

For linear models, you can estimate a model using time-domain data, and
then validate the model using frequency domain data. For nonlinear models,
the System Identification Toolbox product supports only time-domain data.

The following figure shows a sample Residual Analysis plot, created in the
System Identification Tool GUI.

Displaying the Confidence Interval

You can display a confidence interval on the plot in the GUI to gain insight
into the quality of the model. To learn how to show or hide confidence
interval, see the description of the plot settings in “How to Plot Residuals
Using the GUI” on page 8-24.

8-23
 8 Model Analysis

Note If you are working in the System Identification Tool GUI, you can
specify a custom confidence interval. If you are using the resid command, the
confidence interface is fixed at 99%.

The confidence interval corresponds to the range of residual values with a

specific probability of being statistically insignificant for the system. The
toolbox uses the estimated uncertainty in the model parameters to calculate
confidence intervals and assumes the estimates have a Gaussian distribution.

For example, for a 95% confidence interval, the region around zero represents
the range of residual values that have a 95% probability of being statistically
insignificant. You can specify the confidence interval as a probability
(between 0 and 1) or as the number of standard deviations of a Gaussian
distribution. For example, a probability of 0.99 (99%) corresponds to 2.58
standard deviations.

How to Plot Residuals Using the GUI

To create a residual analysis plot for parametric linear and nonlinear models
in the System Identification Tool GUI, select the Model resids check box in
the Model Views area. For general information about creating and working
with plots, see “Working with Plots in the System Identification Tool GUI” on
page 11-15.

To include or exclude a model on the plot, click the corresponding model icon
in the System Identification Tool GUI. Active models display a thick line
inside the Model Board icon.

The following table summarizes the Residual Analysis plot settings.

8-24
 Residual Analysis

Residual Analysis Plot Settings

Action Command
Display confidence • To display the dashed lines on either
intervals around zero. side of the nominal model curve, select
Options > Show confidence intervals.
Select this option again to hide the
Note Confidence internal confidence intervals.
are not available for
• To change the confidence value, select
nonlinear ARX and
Options > Set % confidence level and
Hammerstein-Wiener
choose a value from the list.
models.
• To enter your own confidence level, select
Options > Set confidence level > Other.
Enter the value as a probability (between
0 and 1) or as the number of standard
deviations of a Gaussian distribution.
Change the number of lags • Select Options > Number of lags and
(data samples) for which choose the value from the list.
to compute autocorellation
• To enter your own lag value, select
and cross-correlation
Options > Set confidence level > Other.
functions.
Enter the value as the number of data
samples.
Note For
frequency-domain
validation data, increasing
the number of lags
increases the frequency
resolution of the residual
spectrum and the transfer
function.

(Multiple-output system Select the input-output by name in the

only) Channel menu.
Select a different
input-output pair.

8-25
 8 Model Analysis

How to Plot Residuals at the Command Line

The following table summarizes commands that generate residual-analysis
plots for linear and nonlinear models. For detailed information about this
command, see the corresponding reference page.

Note Apply pe and resid to one model at a time.

Command Description Example

pe Computes and plots model To plot the prediction
prediction errors. errors for the model model
using data data, type the
following command:

pe(model,data)
resid Performs whiteness and To plot residual correlations
independence tests on model for the model model using
residuals, or prediction data data, type the following
errors. Uses validation data command:
input as model input.
resid(model,data)

Example – Examining Model Residuals

This example shows how you can use residual analysis to evaluate model
quality.

Creating Residual Plots

1 To load the sample System Identification Tool session that contains

estimated models, type the following command in the MATLAB Command
Window:

ident('dryer2_linear_models')

2 To generate a residual analysis plot, select the Model resids check box in
the System Identification Tool GUI.

8-26
 Residual Analysis

This opens an empty plot.

3 In the System Identification Tool window, click each model icon to display
it on the Residual Analysis plot.

Note For the nonparametric models, imp and spad, residual analysis plots
are not available.

Description of the Residual Plot Axes

The top axes show the autocorrelation of residuals for the output (whiteness
test). The horizontal scale is the number of lags, which is the time difference
(in samples) between the signals at which the correlation is estimated. The
horizontal dashed lines on the plot represent the confidence interval of the
corresponding estimates. Any fluctuations within the confidence interval are
considered to be insignificant. Four of the models, arxqs, n4s3, arx223 and
amx2222, produce residuals that enter outside the confidence interval. A good

8-27
 8 Model Analysis

model should have a residual autocorrelation function within the confidence

interval, indicating that the residuals are uncorrelated.

The bottom axes show the cross-correlation of the residuals with the
input. A good model should have residuals uncorrelated with past inputs
(independence test). Evidence of correlation indicates that the model does not
describe how the output is formed from the corresponding input. For example,
when there is a peak outside the confidence interval for lag k, this means that
the contribution to the output y(t) that originates from the input u(t-k) is
not properly described by the model. The models arxqs and amx2222 extend
beyond the confidence interval and do not perform as well as the other models.

8-28
 Residual Analysis

Validating Models Using Analyzing Residuals

To remove models with poor performance from the Residual Analysis plot,
click the model icons arxqs, n4s3, arx223, and amx2222 in the System
Identification Tool GUI.

The Residual Analysis plot now includes only the three models that pass the
residual tests: arx692, n4s6, and amx3322.

The plots for these models fall within the confidence intervals. Thus, when
choosing the best model among several estimated models, it is reasonable to
pick amx3322 because it is a simpler, low-order model.

8-29
 8 Model Analysis

Impulse and Step Response Plots

In this section...
“Supported Models” on page 8-30
“How Transient Response Helps to Validate Models” on page 8-30
“What Does a Transient Response Plot Show?” on page 8-31
“How to Plot Impulse and Step Response Using the GUI” on page 8-32
“Displaying the Confidence Interval” on page 8-35
“How to Plot Impulse and Step Response at the Command Line” on page
8-36

Supported Models
You can plot the simulated response of a model using impulse and step
signals as the input for all linear parametric models and correlation analysis
(nonparametric) models.

You can also create step-response plots for nonlinear models. These step
and impulse response plots, also called transient response plots, provide
insight into the characteristics of model dynamics, including peak response
and settling time.

Note For frequency-response models, impulse- and step-response plots are

not available. For nonlinear models, only step-response plots are available.

How Transient Response Helps to Validate Models

Transient response plots provide insight into the basic dynamic properties of
the model, such as response times, static gains, and delays.

Transient response plots also help you validate how well a linear parametric
model, such as a linear ARX model or a state-space model, captures the
dynamics. For example, you can estimate an impulse or step response from
the data using correlation analysis (nonparametric model), and then plot the

8-30
 Impulse and Step Response Plots

correlation analysis result on top of the transient responses of the parametric

models.

Because nonparametric and parametric models are derived using different

algorithms, agreement between these models increases confidence in the
parametric model results.

What Does a Transient Response Plot Show?

Transient response plots show the value of the impulse or step response on
the vertical axis. The horizontal axis is in units of time you specified for the
data used to estimate the model.

The impulse response of a dynamic model is the output signal that results
when the input is an impulse. That is, u(t) is zero for all values of t except at
t=0, where u(0)=1. In the following difference equation, you can compute the
impulse response by setting y(-T)=y(-2T)=0, u(0)=1, and u(t>0)=0.

y(t) − 1.5 y(t − T ) + 0.7 y(t − 2T ) =

0.9u(t) + 0.5u(t − T )

The step response is the output signal that results from a step input, where
u(t<0)=0 and u(t>0)=1.

If your model includes a noise model, you can display the transient response
of the noise model associated with each output channel. For more information
about how to display the transient response of the noise model, see “How to
Plot Impulse and Step Response Using the GUI” on page 8-32.

The following figure shows a sample Transient Response plot, created in

the System Identification Tool GUI.

8-31
 8 Model Analysis

How to Plot Impulse and Step Response Using the

GUI
To create a transient analysis plot in the System Identification Tool GUI,
select the Transient resp check box in the Model Views area. For general
information about creating and working with plots, see “Working with Plots
in the System Identification Tool GUI” on page 11-15.

To include or exclude a model on the plot, click the corresponding model icon
in the System Identification Tool GUI. Active models display a thick line
inside the Model Board icon.

The following table summarizes the Transient Response plot settings.

8-32
 Impulse and Step Response Plots

Transient Response Plot Settings

Action Command
Display step response for linear Select Options > Step response.
or nonlinear model.
Display impulse response for Select Options > Impulse response.
linear model.

Note Not available for nonlinear

models.

Display the confidence interval. • To display the dashed lines on either

side of the nominal model curve,
select Options > Show confidence
Note Only available for linear intervals. Select this option again to
models. hide the confidence intervals.
• To change the confidence value, select
Options > Set % confidence level,
and choose a value from the list.
• To enter your own confidence level,
select Options > Set confidence
level > Other. Enter the value as a
probability (between 0 and 1) or as
the number of standard deviations of
a Gaussian distribution.
Change time span over which • Select Options > Time span (time
the impulse or step response is units), and choose a new time span
calculated. For a scalar time in units of time you specified for the
span T, the resulting response is model.
plotted from -T/4 to T.

8-33
 8 Model Analysis

Transient Response Plot Settings (Continued)

Action Command

• To enter your own time span,

Note To change the time span select Options > Time span (time
of models you estimated using units) > Other, and enter the total
correlation analysis models, response duration.
select Estimate > Correlation • To use the time span based on model
models and reestimate the dynamics, type [] or default.
model using a new time span.
The default time span is computed
based on the model dynamics and
might be different for different
models. For nonlinear models, the
default time span is 10.
Toggle between line plot or stem Select Style > Line plot or
plot. Style > Stem plot.

Tip Use a stem plot for

displaying impulse response.

(Multiple-output system only) Select the output by name in the

Select an input-output pair Channel menu.
to view the noise spectrum
If the plotted models include a noise
corresponding to those channels.
model, you can display the transient
response properties associated with
each output channel. The name of the
channel has the format e@OutputName,
where OutputName is the name of the
output channel corresponding to the
noise model.
(Step response for nonlinear Select Options > Step Size, and then
models only) chose from two options:
Set level of the input step.

8-34
 Impulse and Step Response Plots

Transient Response Plot Settings (Continued)

Action Command

• 0–>1 sets the lower level to 0 and the

Note For multiple-input models, upper level to 1.
the input-step level applies only
• Other opens the Step Level dialog
to the input channel you selected
box, where you enter the values for
to display in the plot.
the lower and upper level values.

Displaying the Confidence Interval

In addition to the transient-response curve, you can display a confidence
interval on the plot. To learn how to show or hide confidence interval, see the
description of the plot settings in “How to Plot Impulse and Step Response
Using the GUI” on page 8-32.

The confidence interval corresponds to the range of response values with

a specific probability of being the actual response of the system. The
toolbox uses the estimated uncertainty in the model parameters to calculate
confidence intervals and assumes the estimates have a Gaussian distribution.

For example, for a 95% confidence interval, the region around the nominal
curve represents the range of values that have a 95% probability of being the
true system response. You can specify the confidence interval as a probability
(between 0 and 1) or as the number of standard deviations of a Gaussian
distribution. For example, a probability of 0.99 (99%) corresponds to 2.58
standard deviations.

Note The calculation of the confidence interval assumes that the model
sufficiently describes the system dynamics and the model residuals pass
independence tests.

8-35
 8 Model Analysis

How to Plot Impulse and Step Response at the

Command Line
You can plot impulse- and step-response plots using the impulse and step
commands, respectively.

All plot commands have the same basic syntax, as follows:

• To plot one model, use the syntax command(model).

• To plot several models, use the syntax
command(model1,model2,...,modelN).

In this case, command represents any of the plotting commands.

To display confidence intervals for a specified number of standard deviations,

use the following syntax:

command(model,'sd',sd)

where sd is the number of standard deviations of a Gaussian distribution. For

example, a confidence value of 99% for the nominal model curve corresponds
to 2.58 standard deviations.

To display a filled confidence region, use the following syntax:

command(model,'sd',sd,'fill')

The following table summarizes commands that generate impulse- and

step-response plots. For detailed information about each command, see the
corresponding reference page.

8-36
 Impulse and Step Response Plots

Command Description Example

impulse Plots impulse response To plot the impulse response
for idpoly, idproc, of the model mod, type the
idarx, idss, and following command:
idgrey model objects.
Estimates and plots impulse(mod)
impulse response
models for iddata
objects.

Note Does not

support nonlinear
models.

step Plots the step response To plot the step response

of all linear and of the model mod, type the
nonlinear models. following command:
Estimates and plots step(mod)
step response models
for iddata objects.
To specify step levels for a
nonlinear model, type the
following command:

step(mod,
'InputLevel',[u1;u2])

8-37
 8 Model Analysis

Frequency Response Plots

In this section...
“What Is Frequency Response?” on page 8-38
“How Frequency Response Helps to Validate Models” on page 8-39
“What Does a Frequency-Response Plot Show?” on page 8-40
“How to Plot Bode Plots Using the GUI” on page 8-41
“How to Plot Bode and Nyquist Plots at the Command Line” on page 8-43

What Is Frequency Response?

Frequency response plots show the complex values of a transfer function as a
function of frequency.

In the case of linear dynamic systems, the transfer function G is essentially

an operator that takes the input u of a linear system to the output y:

y = Gu

For a continuous-time system, the transfer function relates the Laplace

transforms of the input U(s) and output Y(s):

Y (s) = G(s)U (s)

In this case, the frequency function G(iw) is the transfer function evaluated
on the imaginary axis s=iw.

For a discrete-time system sampled with a time interval T, the transfer

function relates the Z-transforms of the input U(z) and output Y(z):

Y ( z) = G( z)U ( z)

In this case, the frequency function G(eiwT) is the transfer function G(z)
evaluated on the unit circle. The argument of the frequency function G(eiwT)

8-38
 Frequency Response Plots

is scaled by the sampling interval T to make the frequency function periodic

with the sampling frequency 2π T .

How Frequency Response Helps to Validate Models

You can plot the frequency response of a model to gain insight into the
characteristics of linear model dynamics, including the frequency of the peak
response and stability margins. Frequency-response plots are available for all
linear parametric models and spectral analysis (nonparametric) models.

Note Frequency-response plots are not available for nonlinear models. In

addition, Nyquist plots do not support time-series models that have no input.

The frequency response of a linear dynamic model describes how the model
reacts to sinusoidal inputs. If the input u(t) is a sinusoid of a certain frequency,
then the output y(t) is also a sinusoid of the same frequency. However, the
magnitude of the response is different from the magnitude of the input signal,
and the phase of the response is shifted relative to the input signal.

Frequency response plots provide insight into linear systems dynamics, such
as frequency-dependent gains, resonances, and phase shifts. Frequency
response plots also contain information about controller requirements and
achievable bandwidths. Finally, frequency response plots can also help you
validate how well a linear parametric model, such as a linear ARX model or a
state-space model, captures the dynamics.

One example of how frequency-response plots help validate other models

is that you can estimate a frequency response from the data using spectral
analysis (nonparametric model), and then plot the spectral analysis result
on top of the frequency response of the parametric models. Because
nonparametric and parametric models are derived using different algorithms,
agreement between these models increases confidence in the parametric
model results.

8-39
 8 Model Analysis

What Does a Frequency-Response Plot Show?

System Identification Tool GUI supports the following types of
frequency-response plots for linear parametric models, linear state-space
models, and nonparametric frequency-response models:

• Bode plot of the model response. A Bode plot consists of two plots. The top
plot shows the magnitude G by which the transfer function G magnifies
the amplitude of the sinusoidal input. The bottom plot shows the phase
ϕ = arg G by which the transfer function shifts the input. The input to
the system is a sinusoid, and the output is also a sinusoid with the same
frequency.
• Bode plot of the disturbance model, called noise spectrum. This plot is the
same as a Bode plot of the model response, but it shows the frequency
response of the noise model instead. For more information, see “Noise
Spectrum Plots” on page 8-46.
• (Only in the MATLAB Command Window)
Nyquist plot. Plots the imaginary versus the real part of the transfer
function.

The following figure shows a sample Bode plot of the model dynamics, created
in the System Identification Tool GUI.

8-40
 Frequency Response Plots

How to Plot Bode Plots Using the GUI

To create a frequency-response plot for parametric linear models in the System
Identification Tool GUI, select the Frequency resp check box in the Model
Views area. For general information about creating and working with plots,
see “Working with Plots in the System Identification Tool GUI” on page 11-15.

To include or exclude a model on the plot, click the corresponding model icon
in the System Identification Tool GUI. Active models display a thick line
inside the Model Board icon.

In addition to the frequency-response curve, you can display a confidence

interval on the plot. The confidence interval corresponds to the range of
response values with a specific probability of being the actual response of the
system. The toolbox uses the estimated uncertainty in the model parameters
to calculate confidence intervals and assumes the estimates have a Gaussian
distribution.

For example, for a 95% confidence interval, the region around the nominal
curve represents the range of values that have a 95% probability of being the

8-41
 8 Model Analysis

true system response. You can specify the confidence interval as a probability
(between 0 and 1) or as the number of standard deviations of a Gaussian
distribution. For example, a probability of 0.99 (99%) corresponds to 2.58
standard deviations.

The following table summarizes the Frequency Function plot settings.

Frequency Function Plot Settings

Action Command
Display the confidence • To display the dashed lines on either
interval. side of the nominal model curve, select
Options > Show confidence intervals.
Select this option again to hide the
confidence intervals.
• To change the confidence value, select
Options > Set % confidence level, and
choose a value from the list.
• To enter your own confidence level,
select Options > Set confidence
level > Other. Enter the value as a
probability (between 0 and 1) or as
the number of standard deviations of a
Gaussian distribution.
Change the frequency Select Options > Frequency range and
values for computing the specify a new frequency vector in units of
noise spectrum. rad/s.
The default frequency Enter the frequency vector using any one of
vector is 128 linearly following methods:
distributed values, greater
than zero and less than • MATLAB expression, such as
or equal to the Nyquist [1:100]*pi/100 or logspace(-3,-1,200).
frequency. Cannot contain variables in the MATLAB
workspace.
• Row vector of values, such as [1:.1:100]

8-42
 Frequency Response Plots

Frequency Function Plot Settings (Continued)

Action Command

Note To restore the default frequency vector,

enter [].

Change frequency units Select Style > Frequency (Hz) or

between hertz and radians Style > Frequency (rad/s).
per second.
Change frequency scale Select Style > Linear frequency scale or
between linear and Style > Log frequency scale.
logarithmic.
Change amplitude scale Select Style > Linear amplitude scale or
between linear and Style > Log amplitude scale.
logarithmic.
(Multiple-output system Select the output by name in the Channel
only) menu.
Select an input-output pair
to view the noise spectrum
corresponding to those
channels.

Note You cannot view

cross spectra between
different outputs.

How to Plot Bode and Nyquist Plots at the Command

Line
You can plot Bode and Nyquist plots for linear models using the bode, ffplot,
and nyquist commands.

All plot commands have the same basic syntax, as follows:

8-43
 8 Model Analysis

• To plot one model, use the syntax command(model).

• To plot several models, use the syntax
command(model1,model2,...,modelN).

In this case, command represents any of the plotting commands.

To display confidence intervals for a specified number of standard deviations,

use the following syntax:

command(model,'sd',sd)

where sd is the number of standard deviations of a Gaussian distribution. For

example, a confidence value of 99% for the nominal model curve corresponds
to 2.58 standard deviations.

To display a filled confidence region, use the following syntax:

command(model,'sd',sd,'fill')

The following table summarizes commands that generate Bode and Nyquist
plots for linear models. For detailed information about each command and
how to specify the frequency values for computing the response, see the
corresponding reference page.

Command Description Example

bode Plots the magnitude To create the bode plot
and phase of the of the model mod, use
frequency response on the following command:
a logarithmic frequency
scale. bode(mod)

ffplot Plots the magnitude To create the bode plot

and phase of the of the model mod, use
frequency response the following command:
on a linear frequency
scale (hertz). ffplot(mod)

nyquist Plots the imaginary To plot the frequency

versus real part of the response of the model
transfer function. mod, use the following
command:

8-44
 Frequency Response Plots

Command Description Example

nyquist(mod)
Note Does not support
time-series models.

8-45
 8 Model Analysis

Noise Spectrum Plots

In this section...
“Supported Models” on page 8-46
“What Does a Noise Spectrum Plot Show?” on page 8-46
“Displaying the Confidence Interval” on page 8-47
“How to Plot the Noise Spectrum Using the GUI” on page 8-48
“How to Plot the Noise Spectrum at the Command Line” on page 8-51

Supported Models
When you estimate the noise model of your linear system, you can plot the
spectrum of the estimated noise model. Noise-spectrum plots are available for
all linear parametric models and spectral analysis (nonparametric) models.

Note For nonlinear models and correlation analysis models, noise-spectrum

plots are not available. For time-series models, you can only generate
noise-spectrum plots for parametric and spectral-analysis models.

What Does a Noise Spectrum Plot Show?

The general equation of a linear dynamic system is given by:

y(t) = G( z)u(t) + v(t)

In this equation, G is an operator that takes the input to the output and
captures the system dynamics, and v is the additive noise term. The toolbox
treats the noise term as filtered white noise, as follows:

v(t) = H ( z) e(t)

The toolbox computes both H and λ during the estimation of the noise model
and stores these quantities as model properties. The H(z) operator represents
the noise model. e(t) is a white-noise source with variance λ .

8-46
 Noise Spectrum Plots

Whereas the frequency-response plot shows the response of G, the

noise-spectrum plot shows the frequency-response of the noise model H.

For input-output models, the noise spectrum is given by the following

equation:

( )
2
Φ v (ω) = λ H eiω

For time-series models (no input), the vertical axis of the noise-spectrum plot
is the same as the dynamic model spectrum. These axes are the same because
there is no input for time series and y = He .

Note You can avoid estimating the noise model by selecting the Output-Error
model structure or by setting the DisturbanceModel property value to 'None'
for a state space model. If you choose to not estimate a noise model for
your system, then H and the noise spectrum amplitude are equal to 1 at all
frequencies.

Displaying the Confidence Interval

In addition to the noise-spectrum curve, you can display a confidence interval
on the plot. To learn how to show or hide confidence interval, see the
description of the plot settings in “How to Plot the Noise Spectrum Using
the GUI” on page 8-48.

The confidence interval corresponds to the range of power-spectrum values

with a specific probability of being the actual noise spectrum of the system.
The toolbox uses the estimated uncertainty in the model parameters to
calculate confidence intervals and assumes the estimates have a Gaussian
distribution.

For example, for a 95% confidence interval, the region around the nominal
curve represents the range of values that have a 95% probability of being the
true system noise spectrum. You can specify the confidence interval as a
probability (between 0 and 1) or as the number of standard deviations of a

8-47
 8 Model Analysis

Gaussian distribution. For example, a probability of 0.99 (99%) corresponds

to 2.58 standard deviations.

Note The calculation of the confidence interval assumes that the model
sufficiently describes the system dynamics and the model residuals pass
independence tests.

How to Plot the Noise Spectrum Using the GUI

To create a noise spectrum plot for parametric linear models in the GUI,
select the Noise spectrum check box in the Model Views area. For general
information about creating and working with plots, see “Working with Plots
in the System Identification Tool GUI” on page 11-15.

To include or exclude a model on the plot, click the corresponding model icon
in the System Identification Tool GUI. Active models display a thick line
inside the Model Board icon.

The following figure shows a sample Noise Spectrum plot.

8-48
 Noise Spectrum Plots

The following table summarizes the Noise Spectrum plot settings.

Noise Spectrum Plot Settings

Action Command
Display the confidence • To display the dashed lines on either side of the
interval. nominal model curve, select Options > Show
confidence intervals. Select this option
again to hide the confidence intervals.
• To change the confidence value, select
Options > Set % confidence level, and
choose a value from the list.
• To enter your own confidence level, select
Options > Set confidence level > Other.
Enter the value as a probability (between 0
and 1) or as the number of standard deviations
of a Gaussian distribution.

8-49
 8 Model Analysis

Noise Spectrum Plot Settings (Continued)

Action Command
Change the frequency Select Options > Frequency range and specify
values for computing a new frequency vector in units of radians per
the noise spectrum. second.
The default frequency Enter the frequency vector using any one of
vector is 128 linearly following methods:
distributed values,
greater than zero and • MATLAB expression, such as [1:100]*pi/100
less than or equal to the or logspace(-3,-1,200). Cannot contain
Nyquist frequency. variables in the MATLAB workspace.
• Row vector of values, such as [1:.1:100]

Tip To restore the default frequency vector, enter

[].

Change frequency units Select Style > Frequency (Hz) or

between hertz and Style > Frequency (rad/s).
radians per second.
Change frequency scale Select Style > Linear frequency scale or
between linear and Style > Log frequency scale.
logarithmic.
Change amplitude scale Select Style > Linear amplitude scale or
between linear and Style > Log amplitude scale.
logarithmic.
(Multiple-output Select the output by name in the Channel menu.
system only)
Select an input-output
pair to view the noise
spectrum corresponding
to those channels.

8-50
 Noise Spectrum Plots

Noise Spectrum Plot Settings (Continued)

Action Command

Note You cannot view

cross spectra between
different outputs.

How to Plot the Noise Spectrum at the Command Line

You can plot the frequency-response of the noise model.

First, select the portion of the model object that corresponds to the noise
model H. For example, to select the noise model in the model object m, type
the following command:

m_noise=m('noise')

Tip You can abbreviate the command to m_noise=m('n').

To plot the frequency-response of the noise model, use the bode command:

bode(m_noise)

To determine if your estimated noise model is good enough, you can compare
the frequency-response of the estimated noise-model H to the estimated
frequency response of v(t). To compute v(t), which represents the actual noise
term in the system, use the following commands:

ysimulated = sim(m,data);
v = ymeasured-ysimulated;

ymeasured is data.y. v is the noise term v(t), as described in “What Does a

Noise Spectrum Plot Show?” on page 8-46 and corresponds to the difference
between the simulated response ysimulated and the actual response
ymeasured.

8-51
 8 Model Analysis

To compute the frequency-response model of the actual noise, use spa:

V = spa(v);

The toolbox uses the following equation to compute the noise spectrum of
the actual noise:

∞
Φ v (ω) = ∑ Rv ( τ ) e−iωτ
τ=−∞

The covariance function Rv is given in terms of E, which denotes the

mathematical expectation, as follows:

Rv ( τ ) = Ev ( t ) v ( t − τ )

To compare the parametric noise-model H to the (nonparametric)

frequency-response estimate of the actual noise v(t), use bode:

bode(V,m('noise'))

If the parametric and the nonparametric estimates of the noise spectra are
different, then you might need a higher-order noise model.

8-52
 Pole and Zero Plots

Pole and Zero Plots

In this section...
“Supported Models” on page 8-53
“What Does a Pole-Zero Plot Show?” on page 8-53
“How to Plot Model Poles and Zeros Using the GUI” on page 8-55
“How to Plot Poles and Zeros at the Command Line” on page 8-56
“Reducing Model Order Using Pole-Zero Plots” on page 8-57

Supported Models
You can create pole-zero plots of linear input-output polynomial, state-space,
and grey-box models.

What Does a Pole-Zero Plot Show?

The following figure shows a sample pole-zero plot of the model with
confidence intervals. x indicate poles and o indicate zeros.

8-53
 8 Model Analysis

The general equation of a linear dynamic system is given by:

y(t) = G( z)u(t) + v(t)

In this equation, G is an operator that takes the input to the output and
captures the system dynamics, and v is the additive noise term.

The poles of a linear system are the roots of the denominator of the transfer
function G. The poles have a direct influence on the dynamic properties of
the system. The zeros are the roots of the numerator of G. If you estimated
a noise model H in addition to the dynamic model G, you can also view the
poles and zeros of the noise model.

Zeros and the poles are equivalent ways of describing the coefficients of a
linear difference equation, such as the ARX model. Poles are associated with
the output side of the difference equation, and zeros are associated with the
input side of the equation. The number of poles is equal to the number of
sampling intervals between the most-delayed and least-delayed output. The
number of zeros) is equal to the number of sampling intervals between the

8-54
 Pole and Zero Plots

most-delayed and least-delayed input. For example, there two poles and one
zero in the following ARX model:

y(t) − 1.5 y(t − T ) + 0.7 y(t − 2T ) =

0.9u(t) + 0.5u(t − T )

How to Plot Model Poles and Zeros Using the GUI

To create a pole-zero plot for parametric linear models in the System
Identification Tool GUI, select the Zeros and poles check box in the Model
Views area. For general information about creating and working with plots,
see “Working with Plots in the System Identification Tool GUI” on page 11-15.

To include or exclude a model on the plot, click the corresponding model icon
in the System Identification Tool GUI. Active models display a thick line
inside the Model Board icon.

In addition, you can display a confidence interval for each pole and zero on the
plot. The confidence interval corresponds to the range of pole or zero values
with a specific probability of being the actual pole or zero of the system. The
toolbox uses the estimated uncertainty in the model parameters to calculate
confidence intervals and assumes the estimates have a Gaussian distribution.

For example, for a 95% confidence interval, the region around the nominal
pole or zero value represents the range of values that have a 95% probability
of being the true system pole or zero value. You can specify the confidence
interval as a probability (between 0 and 1) or as the number of standard
deviations of a Gaussian distribution. For example, a probability of 0.99 (99%)
corresponds to 2.58 standard deviations.

The following table summarizes the Zeros and Poles plot settings.

8-55
 8 Model Analysis

Zeros and Poles Plot Settings

Action Command
Display the confidence • To display the dashed lines on either side
interval. of the nominal pole and zero values, select
Options > Show confidence intervals.
Select this option again to hide the confidence
intervals.
• To change the confidence value, select
Options > Set % confidence level, and
choose a value from the list.
• To enter your own confidence level, select
Options > Set confidence level > Other.
Enter the value as a probability (between 0
and 1) or as the number of standard deviations
of a Gaussian distribution.
Show real and Select Style > Re/Im-axes. Select this option
imaginary axes. again to hide the axes.
Show the unit circle. Select Style > Unit circle. Select this option
again to hide the unit circle.
(Multiple-output system Select the output by name in the Channel menu.
only)
Select an input-output
pair to view the poles
and zeros corresponding
to those channels.

How to Plot Poles and Zeros at the Command Line

You can create a pole-zero plot for linear polynomial, linear state-space, and
linear grey-box models using the pzmap command. pzmap lets you include
several models on a plot.

To display confidence intervals for a specified number of standard deviations,

use the following syntax:

pzmap(model,'sd',sd)

8-56
 Pole and Zero Plots

where sd is the number of standard deviations of a Gaussian distribution. For

example, a confidence value of 99% for the nominal model curve corresponds
to 2.58 standard deviations.

Command Description Example

pzmap Plots zeros and poles To plot the poles and
of the model on the zeros of the model
S-plane or Z-plane for mod, use the following
continuous-time or command:
discrete-time model,
respectively. pzmap(mod)

For detailed information about pzmap, see the corresponding reference page.

Reducing Model Order Using Pole-Zero Plots

You can use pole-zero plots to evaluate whether it might be useful to reduce
model order. When confidence intervals for a pole-zero pair overlap, this
overlap indicates a possible pole-zero cancelation.

For example, you can use the following syntax to plot a 1-standard-deviation
confidence interval around model poles and zeros.

pzmap(model,'sd',1)

If poles and zeros overlap, try estimating a lower order model.

Always validate model output and residuals to see if the quality of the
fit changes after reducing model order. If the plot indicates pole-zero
cancellations, but reducing model order degrades the fit, then the extra
poles probably describe noise. In this case, you can choose a different model
structure that decouples system dynamics and noise. For example, try
ARMAX, Output-Error, or Box-Jenkins polynomial model structures with
an A or F polynomial of an order equal to that of the number of uncanceled
poles. For more information about estimating linear polynomial models, see
“Identifying Input-Output Polynomial Models” on page 3-39.

8-57
 8 Model Analysis

Akaike’s Criteria for Model Validation

In this section...
“Definition of FPE” on page 8-58
“Computing FPE” on page 8-59
“Definition of AIC” on page 8-59
“Computing AIC” on page 8-60

Definition of FPE
Akaike’s Final Prediction Error (FPE) criterion provides a measure of model
quality by simulating the situation where the model is tested on a different
data set. After computing several different models, you can compare them
using this criterion. According to Akaike’s theory, the most accurate model
has the smallest FPE.

Note If you use the same data set for both model estimation and validation,
the fit always improves as you increase the model order and, therefore, the
flexibility of the model structure.

Akaike’s Final Prediction Error (FPE) is defined by the following equation:

⎛1+ dN ⎞
FPE = V ⎜
⎜ 1 − d ⎟⎟
⎝ N⎠

where V is the loss function, d is the number of estimated parameters, and N

is the number of values in the estimation data set.

The toolbox assumes that the final prediction error is asymptotic for d<<N
and uses the following approximation to compute FPE:

FPE = V (1 + 2 d N )

The loss function V is defined by the following equation:

8-58
 Akaike’s Criteria for Model Validation

⎛ N ⎞
( ) ( ε ( t, θ N ))
T
⎜ ∑
1
V = det ⎜ N ε t, θ N ⎟
⎟
⎝ 1 ⎠

where θ N represents the estimated parameters.

Computing FPE
You can compute Akaike’s Final Prediction Error (FPE) criterion for linear
and nonlinear models.

Note FPE for nonlinear ARX models that include a tree partition
nonlinearity is not supported.

To compute FPE, use the fpe command, as follows:

FPE = fpe(m1,m2,m3,...,mN)

According to Akaike’s theory, the most accurate model has the smallest FPE.

You can also access the FPE value of an estimated model by accessing the
FPE field of the EstimationInfo property of this model. For example, if you
estimated the model m, you can access its FPE using the following command:

m.EstimationInfo.FPE

Definition of AIC
Akaike’s Information Criterion (AIC) provides a measure of model quality
by simulating the situation where the model is tested on a different data
set. After computing several different models, you can compare them using
this criterion. According to Akaike’s theory, the most accurate model has
the smallest AIC.

Note If you use the same data set for both model estimation and validation,
the fit always improves as you increase the model order and, therefore, the
flexibility of the model structure.

8-59
 8 Model Analysis

Akaike’s Information Criterion (AIC) is defined by the following equation:

2d
AIC = log V +
N

where V is the loss function, d is the number of estimated parameters, and N

is the number of values in the estimation data set.

The loss function V is defined by the following equation:

⎛ N ⎞
( ) ( ε ( t, θ N ))
T
⎜ ∑
1
V = det ⎜ N ε t, θ N ⎟
⎟
⎝ 1 ⎠

where θ N represents the estimated parameters.

For d<<N:

⎛ ⎛ 2d ⎞ ⎞
AIC = log ⎜ V ⎜ 1 + ⎟⎟
⎝ ⎝ N ⎠⎠

Note AIC is approximately equal to log(FPE).

Computing AIC
Use the aic command to compute Akaike’s Information Criterion (AIC) for
one or more linear or nonlinear models, as follows:

AIC = aic(m1,m2,m3,...,mN)

According to Akaike’s theory, the most accurate model has the smallest AIC.

8-60
 Computing Model Uncertainty

Computing Model Uncertainty

In this section...
“Why Analyze Model Uncertainty?” on page 8-61
“What Is Model Covariance?” on page 8-61
“Viewing Model Uncertainty Information” on page 8-62

Why Analyze Model Uncertainty?

In addition to estimating model parameters, the toolbox algorithms also
estimate variability of the model parameters that result from random
disturbances in the output.

Understanding model variability helps you to understand how different your

model parameters would be if you repeated the estimation using a different
data set (with the same input sequence as the original data set) and the same
model structure.

When validating your parametric models, check the uncertainty values.

Large uncertainties in the parameters might be caused by high model orders,
inadequate excitation, and poor signal-to-noise ratio in the data.

Note You can get model uncertainty data for linear parametric black-box
models, and both linear and nonlinear grey-box models. Supported model
objects include idproc, idpoly, idss, idarx, idgrey, idfrd, and idnlgrey.

What Is Model Covariance?

Uncertainty in the model is called model covariance.

If you estimate model uncertainty data, this information is stored in the

Model.CovarianceMatrix model property. The covariance matrix is used to
compute all uncertainties in model output, Bode plots, residual plots, and
pole-zero plots.

8-61
 8 Model Analysis

Computing the covariance matrix is based on the assumption that the model
structure gives the correct description of the system dynamics. For models
that include a disturbance model H, a correct uncertainty estimate assumes
that the model produces white residuals. To determine whether you can trust
the estimated model uncertainty values, perform residual analysis tests on
your model, as described in “Residual Analysis” on page 8-21. If your model
passes residual analysis tests, there is a good chance that the true system lies
within the confidence interval and any parameter uncertainties results from
random disturbances in the output.

In the case of output-error models, where the noise model H is fixed to 1,

computing the covariance matrix does not assume that the residuals are
white. Instead, the covariance is estimated based on the estimated color of the
residual correlations. This estimation of the noise color is also performed for
state-space models with K=0, which is equivalent to an output-error model.

Viewing Model Uncertainty Information

You can view the following uncertainty information from linear and nonlinear
grey-box models:

• Uncertainties of estimated parameters.

Type present(model) at the prompt, where model represents the name of
a linear or nonlinear model.
• Confidence intervals on the linear model plots, including step-response,
impulse-response, Bode, and pole-zero plots.
Confidence intervals are computed based on the variability in the model
parameters. For information about displaying confidence intervals, see the
corresponding plot section.
• Covariance matrix of the estimated parameters in linear and nonlinear
grey-box models.
Type model.CovarianceMatrix at the prompt, where model represents the
name of the model object.
• Estimated standard deviations of polynomial coefficients or state-space
matrices

8-62
 Computing Model Uncertainty

Type model.dA at the prompt to access the estimated standard deviations

of the model.A estimated property, where model represents the name
of the model object, and A represents any estimated model property. In
general, you prefix the name of the estimated property with a d to get the
standard deviation estimate for that property. For example, to get the
standard deviation value of the A polynomial in an estimated ARX model,
type model.da.

Note State-space models, estimated with free parameterization, do

not have well-defined standard deviations of the matrix elements. To
display matrix parameter uncertainties in this case, first transform the
model to a canonical parameterization by setting the ss model property
to model.ss = 'canon'. For more information about free and canonical
parameterizations, see “Identifying State-Space Models” on page 3-73.

• Simulated output values for linear models with standard deviations using
the sim command.
Call the sim command with output arguments, where the second output
argument is the estimated standard deviation of each output value.
For example, type [ysim,ysimsd]=sim(model,data), where ysim is
the simulated output, ysimsd contains the standard deviations on the
simulated output, and data is the simulation data.

8-63
 8 Model Analysis

Troubleshooting Models
In this section...
“About Troubleshooting Models” on page 8-64
“Model Order Is Too High or Too Low” on page 8-64
“Nonlinearity Estimator Produces a Poor Fit” on page 8-65
“Substantial Noise in the System” on page 8-66
“Unstable Models” on page 8-66
“Missing Input Variables” on page 8-68
“Complicated Nonlinearities” on page 8-68

About Troubleshooting Models

During validation, you might find that your model output fits the validation
data poorly. You might also find some unexpected or undesirable model
characteristics.

If the tips suggested in these sections do not help improve your models, then
a good model might not be possible for this data. For example, your data
might have poor signal-to-noise ratio, large and nonstationary disturbances,
or varying system properties.

Model Order Is Too High or Too Low

When the Model Output plot does not show a good fit, there is a good chance
that you need to try a different model order. System identification is largely
a trial-and-error process when selecting model structure and model order.
Ideally, you want the lowest-order model that adequately captures the system
dynamics.

You can estimate the model order as described in “Preliminary Step –

Estimating Model Orders and Input Delays” on page 3-48. Typically, you use
the suggested order as a starting point to estimate the lowest possible order
with different model structures. After each estimation, you monitor the Model
Output and the Residual Analysis plots, and then adjust your settings for
the next estimation.

8-64
 Troubleshooting Models

When a low-order model fits the validation data poorly, try estimating a
higher-order model to see if the fit improves. For example, if a Model Output
plot shows that a fourth-order model gives poor results, try estimating an
eighth-order model. When a higher-order model improves the fit, you can
conclude that higher-order models might be required and linear models might
be sufficient.

You should use an independent data set to validate your models. If you
use the same data set to both estimate and validate a model, the fit always
improves as you increase model order, and you risk overfitting. However, if
you use an independent data set to validate your models, the fit eventually
deteriorates if your model orders are too high.

High-order models are more expensive to compute and result in greater

parameter uncertainty.

Nonlinearity Estimator Produces a Poor Fit

In the case of nonlinear ARX and Hammerstein-Wiener models, the Model
Output plot does not show a good fit when the nonlinearity estimator has
incorrect complexity.

You specify the complexity of piece-wise-linear, wavelet, sigmoid, and

custom networks using the number of units (NumberOfUnits nonlinear
estimator property). A high number of units indicates a complex nonlinearity
estimator. In the case of neural networks, you specify the complexity using
the parameters of the network object. For more information, see the Neural
Network Toolbox documentation.

To select the appropriate complexity of the nonlinearity estimator, start

with a low complexity and validate the model output. Next, increate the
complexity and validate the model output again. The model fit degrades when
the nonlinearity estimator becomes too complex.

Note To see the model fit degrade when the nonlinearity estimator becomes
too complex, you must use an independent data set to validate the data that is
different from the estimation data set.

8-65
 8 Model Analysis

Substantial Noise in the System

There are a couple of indications that you might have substantial noise in
your system and might need to use linear model structures that are better
equipped to model noise.

One indication of noise is when a state-space model is better than an ARX

model at reproducing the measured output; whereas the state-space structure
has sufficient flexibility to model noise, the ARX model structure is less able
to model noise because the A polynomial must account for both the system
dynamics and the noise. The following equation represents the ARX model
and shows that A couples the dynamics and the noise by appearing in the
denominator of both the dynamics term and the noise terms:

B 1
y= u+ e
A A

Another indication that a noise model is needed appears in residual analysis

plots when you see significant autocorrelation of residuals at nonzero lags.
For more information about residual analysis, see “Residual Analysis” on
page 8-21.

To model noise more carefully, use the ARMAX or the Box-Jenkins model
structure, where the dynamics term and the noise term are modeled by
different polynomials.

Unstable Models

Unstable Linear Model

You can test whether a linear model is unstable is by examining the pole-zero
plot of the model, which is described in “Pole and Zero Plots” on page
8-53. The stability threshold for pole values differs for discrete-time and
continuous-time models, as follows:

• For stable continuous-time models, the real part of the pole is less than 0.
• For stable discrete-time models, the magnitude of the pole is less than 1.

8-66
 Troubleshooting Models

Note Linear trends might cause linear models to be unstable. However,

detrending the model does not guarantee stability.

When an unstable model is OK: In some cases, an unstable model is still

a useful model. For example, your system might be unstable without a
controller, and you plan to use your model for control design. In this case, you
can import your unstable model into Simulink or Control System Toolbox
products.

Forcing stability during estimation: If you believe that your system is

stable, but your model is unstable, then you can estimate the model with a
Focus set to Stability. This setting might result in a reduced model quality.
For more information about Focus, see the Algorithm Properties reference
page.

Allowing for some instability: A more advanced approach to achieving a

stable model is by setting the stability threshold property to allow a margin of
error. The threshold model property is accessed as a field in the algorithm
structure:

• For continuous-time models, set the value of

model.algorithm.advanced.sstability. The model is considered stable
if the pole on the far right is to the left of sstability threshold.
• For discrete-time models, set the value of
model.algorithm.advanced.zstability. The model is considered stable
if all poles inside the circle centered at the origin and with a radius
zstability.

For more information about Threshold fields for linear models, see the
Algorithm Properties reference page.

Unstable Nonlinear Models

To test if a nonlinear model is unstable is to plot the simulated model output
on top of the validation data. If the simulated output diverges from measured
output, the model is unstable. However, agreement between model output
and measured output does not guarantee stability.

8-67
 8 Model Analysis

Missing Input Variables

If the Model Output plot and Residual Analysis plot shows a poor fit and
you have already tried different structures and orders and modeled noise, it
might be that there are one or more missing inputs that have a significant
effect on the output.

Try including other measured signals in your input data, and then estimating
the models again.

Inputs need not be control signals. Any measurable signal can be considered
an input, including measurable disturbances.

Complicated Nonlinearities
If the Model Output plot and Residual Analysis plot shows a poor fit, consider
if nonlinear effects are present in the system.

You can model the nonlinearities by performing a simple transformation

on the signals to make the problem linear in the new variables. For
example, if electrical power is the driving stimulus in a heating process and
temperature is the output, you can form a simple product of voltage and
current measurements.

If your problem is sufficiently complex and you do not have physical insight
into the problem, you might try fitting nonlinear black-box models. For more
information, see Chapter 4, “Nonlinear Black-Box Model Identification”.

8-68
 Next Steps After Getting an Accurate Model

Next Steps After Getting an Accurate Model

For linear parametric models (idmodel objects), you can perform the following
operations:

• Transform between continuous-time and discrete-time representation.

See “Transforming Between Discrete-Time and Continuous-Time
Representations” on page 3-113.
• Transform between linear model representations, such as between
polynomial, state-space, and zero-pole representations.
See “Transforming Between Linear Model Representations” on page 3-118.
• Extract numerical data from transfer functions, pole-zero models, and
state-space matrices.
See “Extracting Parameter Values from Linear Models” on page 3-109.

For nonlinear black-box models (idnlarx and idnlhwobjects), you can

compute a linear approximation of the nonlinear model. See “Linear
Approximation of Nonlinear Black-Box Models” on page 4-81.

System Identification Toolbox models in the MATLAB workspace are

immediately available to other MathWorks® products. However, if you used
the System Identification Tool GUI to estimate models, you must first export
the models to the MATLAB workspace.

Tip To export a model from the GUI, drag the model icon to the To
Workspace rectangle. For more information about working with the GUI,
see Chapter 11, “System Identification Tool GUI”.

If you have the Control System Toolbox software installed, you can import
your linear plant model for control-system design. For more information, see
“Using Models with Control System Toolbox Software” on page 9-2.

Finally, if you have Simulink software installed, you can exchange data
between the System Identification Toolbox software and the Simulink

8-69
 8 Model Analysis

environment. For more information, see Chapter 10, “System Identification

Toolbox Blocks”.

8-70
 9

Control Design Applications

 9 Control Design Applications

Using Models with Control System Toolbox Software

In this section...
“How Control System Toolbox Software Works with Identified Models”
on page 9-2
“Using balred to Reduce Model Order” on page 9-3
“Compensator Design Using Control System Toolbox Software” on page 9-3
“Converting Models to LTI Objects” on page 9-4
“Viewing Model Response Using the LTI Viewer” on page 9-5
“Combining Model Objects” on page 9-6
“Example – Using System Identification Toolbox Software with Control
System Toolbox Software” on page 9-6

How Control System Toolbox Software Works with

Identified Models
System Identification Toolbox software integrates with Control System
Toolbox software by providing a plant for control design.

Control System Toolbox software also provides the LTI Viewer GUI to extend
System Identification Toolbox functionality for linear model analysis.

Control System Toolbox software supports only linear models. If you

identified a nonlinear plant model using System Identification Toolbox
software, you must linearize it before you can work with this model in the
Control System Toolbox software. For more information, see the linapp,
linearize(idnlarx), or linearize(idnlhw) reference page.

Note You can only use the System Identification Toolbox software to linearize
nonlinear ARX (idnlarx) and Hammerstein-Wiener (idnlhw) models.
Linearization of nonlinear grey-box (idnlgrey) models is not supported.

For information about using the Control System Toolbox software, see the
Control System Toolbox documentation.

9-2
 Using Models with Control System Toolbox™ Software

Using balred to Reduce Model Order

In some cases, the order of your identified model might be higher than
necessary to capture the dynamics. If you have the Control System Toolbox
software, you can use balred to compute a state-spate model approximation
with a reduced model order for any idmodel object, including idarx, idpoly,
idss, and idgrey.

For more information about using balred, see the corresponding reference
page. To learn how you can reduce model order using pole-zero plots, see
“Reducing Model Order Using Pole-Zero Plots” on page 8-57.

Compensator Design Using Control System Toolbox

Software
After you estimate a plant model using System Identification Toolbox
software, you can use Control System Toolbox software to design a controller
for this plant.

System Identification Toolbox models in the MATLAB workspace are

immediately available to Control System Toolbox commands. However, if you
used the System Identification Tool GUI to estimate models, you must first
export the models to the MATLAB workspace. To export a model from the
GUI, drag the model icon to the To Workspace rectangle.

Control System Toolbox software provides both the SISO Design Tool GUI
and commands for working at the command line. You can import polynomial
and state-space models directly into SISO Design Tool using the following
command:

sisotool(model('measured'))

where you use only the dynamic model and not the noise model. For more
information about subreferencing the dynamic or the noise model, see
“Subreferencing Measured and Noise Models” on page 3-121. To design a
controller using Control System Toolbox commands and methods at the
command line, you must convert the plant model to an LTI object. For more
information, see “Converting Models to LTI Objects” on page 9-4.

9-3
 9 Control Design Applications

Note The syntax sisotool(model('m')) is equivalent to

sisotool(model('measured')).

For more information about controller design using SISO Design Tool
and Control System Toolbox commands, see the Control System Toolbox
documentation.

Converting Models to LTI Objects

Control System Toolbox commands operate on LTI objects. To design a
controller for a plant model, you must first convert the System Identification
Toolbox model object to an LTI object.

You can convert linear polynomial, state-space, and grey-box model objects,
including idarx, idpoly, idproc, idss, or idgrey, to LTI objects.

The following table summarizes the commands for transforming linear

state-space and polynomial models to an LTI object.

Commands for Converting Models to LTI Objects

Command Description Example

frd Convert to
frequency-response ss_sys = frd(model)
representation.
ss Convert to state-space
representation. ss_sys = ss(model)

tf Convert to
transfer-function form. tf_sys = tf(model)

zpk Convert to zero-pole form.

zpk_sys = zpk(model)

The following code transforms an idmodel object to an LTI state-space object:

% Extract the measured model

% and ignore the noise model

9-4
 Using Models with Control System Toolbox™ Software

model = model('measured')
% Convert to LTI object
LTI_sys = idss(model)

The LTI object includes only the dynamic model and not the noise model,
which is estimated for every linear model in the System Identification Toolbox
software.

Note To include noise channels in the LTI models, first use noisecnv to
convert the noise in the idmodel object to measured channels, and then
convert to an LTI object.

For more information about subreferencing the dynamic or the noise model,
see “Subreferencing Measured and Noise Models” on page 3-121.

Viewing Model Response Using the LTI Viewer

• “What Is the LTI Viewer?” on page 9-5
• “Displaying Identified Models in the LTI Viewer” on page 9-6

What Is the LTI Viewer?

If you have the Control System Toolbox software, you can plot models in the
LTI Viewer from either the System Identification Tool GUI or the MATLAB
Command Window.

The LTI Viewer is a graphical user interface for viewing and manipulating
the response plots of linear models.

Note The LTI Viewer does not display model uncertainty.

For more information about working with plots in the LTI Viewer, see the
Control System Toolbox documentation.

9-5
 9 Control Design Applications

Displaying Identified Models in the LTI Viewer

When the MATLAB software is installed, the System Identification Tool GUI
contains the To LTI Viewer rectangle. To plot models in the LTI Viewer,
drag and drop the corresponding icon to the To LTI Viewer rectangle in
the System Identification Tool GUI.

Alternatively, use the following syntax when working at the command line
to view a model in the LTI Viewer:

view(model)

Combining Model Objects

If you have the Control System Toolbox software, you can combine linear
model objects, such as idarx, idgrey, idpoly, idproc, and idss model
objects, similar to the way you combine LTI objects.

For example, you can perform the following operations on identified models:

• G1+G2
• G1*G2
• append(G1,G2)
• feedback(G1,G2)

Note These operations lose covariance information.

Example – Using System Identification Toolbox

Software with Control System Toolbox Software
This example demonstrates how to use both System Identification Toolbox
commands and Control System Toolbox commands to create and plot models:

% Construct model using Control System Toolbox

m0 = drss(4,3,2)
% Convert model to an idss object
m0 = idss(m0,'NoiseVar',0.1*eye(3))
% Generate input data for simulating output

9-6
 Using Models with Control System Toolbox™ Software

u = iddata([], idinput([800 2],'rbs'));

% Simulate model output using System Identification Toolbox
% with added noise
y = sim(m0,u,'noise')
% Form an input-output iddata object
Data = [y u];
% Estimate state-space model from the generated data
% using System Identification Toolbox command pem
m = pem(Data(1:400))
% Convert the model to a Control System Toolbox transfer function
tf(m)
% Plot model output for model m using System Identification Toolbox
compare(Data(401:800),m)
% Display identified model m in LTI Viewer
view(m)

9-7
 9 Control Design Applications

9-8
 10

System Identification
Toolbox Blocks

• “System Identification Toolbox Block Library” on page 10-2

• “Opening the System Identification Toolbox Block Library” on page 10-3
• “Preparing Data” on page 10-4
• “Identifying Linear Models” on page 10-5
• “Simulating Model Output” on page 10-6
• “Example – Simulating a Model Using Simulink Software” on page 10-9
 10 System Identification Toolbox™ Blocks

System Identification Toolbox Block Library

System Identification Toolbox provides blocks for sharing information
between the MATLAB and Simulink environments.

You can use the System Identification Toolbox block library to perform the
following tasks:

• Stream time-domain data source (iddata object) into a Simulink model.

• Export data from a simulation in Simulink software as a System
Identification Toolbox data object (iddata object).
• Import estimated models into a Simulink model, and simulate the models
with or without noise.
The model you import might be a component of a larger system modeled in
Simulink. For example, if you identified a plant model using the System
Identification Toolbox software, you can import this plant into a Simulink
model for control design.
• Estimate parameters of linear polynomial models during simulation from
single-output data.

10-2
 Opening the System Identification Toolbox™ Block Library

Opening the System Identification Toolbox Block Library

To open the System Identification Toolbox block library, select
Start > Simulink > Library Browser. In the Library Browser, select
System Identification Toolbox.

You can also open the System Identification Toolbox block library directly by
typing the following command at the MATLAB prompt:

slident

For more information about blocks, see “Block Reference” in the System
Identification Toolbox Reference. To get help on a specific block, right-click
the block in the Library Browser, and select Help.

10-3
 10 System Identification Toolbox™ Blocks

Preparing Data
The following table summarizes the blocks you use to transfer data between
the MATLAB and Simulink environments.

After you add a block to the Simulink model, double-click the block to specify
block parameters. For an example of bringing data into a Simulink model,
see the tutorial on estimating process models in the System Identification
Toolbox Getting Started Guide.

Block Description
Iddata Sink Export input and output signals to the MATLAB
workspace as an iddata object.
Iddata Source Import iddata object from the MATLAB
workspace.
Input and output ports of the block correspond
to input and output signals of the data. These
inputs and outputs provide signals to blocks that
are connected to this data block.

For information about configuring each block, see the corresponding reference
pages.

10-4
 Identifying Linear Models

Identifying Linear Models

The following table summarizes the blocks you use to estimate model
parameters in a Simulink model during simulation and export the results to
the MATLAB environment.

After you add a block to the model, double-click the block to specify block
parameters.

Block Description
AutoRegressive model Estimate AR model parameters from time-series
estimator data, which has one output and no input.
AutoRegressive Moving Estimate ARMAX model parameters from
Average with eXternal input/output data.
input model estimator
AutoRegressive with Estimate ARX model parameters from
eXternal input model input/output data.
estimator
Box-Jenkins model Estimate BJ model parameters from input/output
estimator data.
Output-error model Estimate OE model parameters from
estimator input/output data.

General model Estimate ARX, ARMAX, Box-Jenkins, and

estimator using Output-Error models (idpoly objects) from
Predictive Error Method single-input and single output data using general
prediction-error method.

For information about configuring each block, see the corresponding reference
pages.

10-5
 10 System Identification Toolbox™ Blocks

Simulating Model Output

In this section...
“When to Use Simulation Blocks” on page 10-6
“Summary of Simulation Blocks” on page 10-6
“Specifying Initial Conditions for Simulation” on page 10-7

When to Use Simulation Blocks

Add model simulation blocks to your Simulink model from the System
Identification Toolbox block library when you want to:

• Represent the dynamics of a physical component in a Simulink model using

a data-based nonlinear model.
• Replace a complex Simulink subsystem with a simpler data-based
nonlinear model.

You use the model simulation blocks to import the models you identified using
System Identification Toolbox software from the MATLAB workspace into the
Simulink environment. For a list of System Identification Toolbox simulation
blocks, see “Summary of Simulation Blocks” on page 10-6.

Summary of Simulation Blocks

The following table summarizes the blocks you use to import models from
the MATLAB environment into a Simulink model for simulation. Importing
a model corresponds to entering the model variable name in the block
parameter dialog box.

Block Description
Idmodel Simulate idmodel model in Simulink, including
low-order transfer function (idproc), linear
polynomial (idpoly), state-space (idss), and
grey-box (idgrey) models. Also simulates idarx
model objects.
Nonlinear ARX Model Simulate idnlarx model in Simulink.

10-6
 Simulating Model Output

Block Description
Hammerstein-Wiener Simulate idnlhw model in Simulink.
Model
Nonlinear Grey-Box Simulate nonlinear ODE (idnlgrey model object)
Model in Simulink.

After you import the model into Simulink software, use the block parameter
dialog box to specify the initial conditions for simulating that block. (See
“Specifying Initial Conditions for Simulation” on page 10-7.) For information
about configuring each block, see the corresponding reference pages.

Specifying Initial Conditions for Simulation

For accurate simulation of a linear or a nonlinear model, you can use default
initial conditions or specify the initial conditions for simulation using the
block parameters dialog box.

For more information on specifying initial conditions, see the following topics:

• “Specifying Initial States of Linear Models” on page 10-7

• “Specifying Initial States of Nonlinear ARX Models” on page 10-8
• “Specifying Initial States of Hammerstein-Wiener Models” on page 10-8

Specifying Initial States of Linear Models

For idss and idgrey models, specify the initial states for simulation in the
Initial state field of the Function Block Parameters: Idmodel dialog box:

• To specify the initial states values as zero, use 'z'.

• To use the initial states values stored in the X0 property of the model, use
'm'.
• To match the simulated response of the model to a certain input/output
data set, first use findstates to estimate initial states values that maximize
the fit. Specify the estimated initial states values in the Initial state field.

For idpoly and idarx models, the default initial states values are zero. If
you want to specify different values, such as maximize fit to a given output

10-7
 10 System Identification Toolbox™ Blocks

data, convert the model to idss object, and then specify its initial states as
described previously. For example, for the following idpoly model:

m1=idpoly([1 2 1],[2 2]);

the initial states correspond to those of the equivalent state-space model:

m2=idss(m1);

For more information about specifying initial conditions for simulation, see
the IDMODEL Model reference page.

Specifying Initial States of Nonlinear ARX Models

The states of a nonlinear ARX model correspond to the dynamic elements
of the nonlinear ARX model structure, which are the model regressors.
Regressors can be the delayed input/output variables (standard regressors)
or user-defined transformations of delayed input/output variables (custom
regressors). For more information about the states of a nonlinear ARX model,
see the idnlarx reference page.

For simulating nonlinear ARX models, you can specify the initial conditions
as input/output values, or as a vector. For more information about specifying
initial conditions for simulation, see the IDNLARX Model reference page.

Specifying Initial States of Hammerstein-Wiener Models

The states of a Hammerstein-Wiener model correspond to the states of the
embedded linear (idpoly or idss) model. For more information about the
states of a Hammerstein-Wiener model, see the idnlhw reference page.

The default initial state for simulating a Hammerstein-Wiener model is 0.

For more information about specifying initial conditions for simulation, see
the IDNLHW Model reference page.

10-8
 Example – Simulating a Model Using Simulink® Software

Example – Simulating a Model Using Simulink Software

In this example, you set the initial states for simulating a model such that the
simulation provides a best fit to measured input-output data.

Prerequisites

Estimate a three-state model M using a multiple-experiment data set Z, which

contains data from three experiments — z1, z2, and z3:

% Load multi-experiment data.

load(fullfile(matlabroot, 'toolbox', 'ident', 'iddemos',...
'data', 'twobodiesdata'));

% Create an iddata object to store the multi-experiment data.

z1=iddata(y1, u1, 0.005,'Tstart',0);
z2=iddata(y2, u2, 0.005,'Tstart',0);
z3=iddata(y3, u3, 0.005,'Tstart',0);
Z = merge(z1,z2,z3);

% Estimate a 5th order state-space model.

M = n4sid(Z,5,'Focus','Simulation');

When you estimate a model using multiple data sets, the initial-states
property, X0, of the model, M, stores only the estimated states corresponding to
the last data set. In this example:

• Values of M.X0 are the estimated state values, corresponding to the last
experiment in Z.
• M.X0 is a vector of length 5, corresponding to the five states of the model.

To compute initial states that maximizes the fit to the corresponding output
y2, and simulate the model using the second experiment:

1 Estimate the initial states using the second experiment:

X0est = findstates(M,z2);

2 Open the System Identification Toolbox library by typing the following

command at the MATLAB prompt:

10-9
 10 System Identification Toolbox™ Blocks

slident

3 Open a new Simulink model window. Then, drag and drop an Idmodel
block from the library into the model window.

4 Open the Function Block Parameters dialog box by double-clicking the

idmodel block. Specify the following block parameters:
a In the idmodel variable field, type M to specify the estimated model.

b In the Initial states field, type X0est to specify the estimated initial
states.
c Click OK.

5 Drag and drop an Iddata Source block into the model window. Then,
configure the model, as shown in the following figure.

6 Simulate the model for 2 seconds, and compare the simulated output ysim
with the measured output ymeasured using the Scope block.

10-10
 11

System Identification Tool

GUI

• “Steps for Using the System Identification Tool GUI” on page 11-2
• “Starting and Managing GUI Sessions” on page 11-3
• “Managing Models in the GUI” on page 11-9
• “Working with Plots in the System Identification Tool GUI” on page 11-15
• “Customizing the System Identification Tool GUI” on page 11-20
 11 System Identification Tool GUI

Steps for Using the System Identification Tool GUI

A typical workflow in the System Identification Tool GUI includes the
following steps:

1 Import your data into the MATLAB workspace, as described in “Importing

Data into the MATLAB Workspace” on page 2-7.

2 Start a new session in the System Identification Tool GUI, or open a saved
session. For more information, see “Starting a New Session in the GUI”
on page 11-4.

3 Import data into the GUI from the MATLAB workspace. For more
information, see “Importing Data into the GUI” on page 2-14.

4 Plot and preprocess data to prepare it for system identification. For

example, you can remove constant offsets or linear trends (for linear models
only), filter data, or select data regions of interest. For more information,
see Chapter 2, “Data Import and Processing”.

5 Specify the data for estimation and validation. For more information, see
“Specifying Estimation and Validation Data” on page 2-31.

6 Select the model to estimating using the Estimate menu. For more
information, see Chapter 1, “Choosing Your System Identification
Approach”.

7 Validate models. For more information, see Chapter 8, “Model Analysis”.

8 Export models to the MATLAB workspace for further analysis. For

more information, see “Exporting Models from the GUI to the MATLAB
Workspace” on page 11-13.

11-2
 Starting and Managing GUI Sessions

Starting and Managing GUI Sessions

In this section...
“What Is a System Identification Tool Session?” on page 11-3
“Starting a New Session in the GUI” on page 11-4
“Description of the System Identification Tool Window” on page 11-5
“Opening a Saved Session” on page 11-6
“Saving, Merging, and Closing Sessions” on page 11-6
“Deleting a Session” on page 11-7
“Getting Help in the GUI” on page 11-7
“Exiting the System Identification Tool GUI” on page 11-8

What Is a System Identification Tool Session?

A session represents the total progress of your identification process, including
any data sets and models in the System Identification Tool GUI.

You can save a session to a file with a .sid extension. For example, you can
save different stages of your progress as different sessions so that you can
revert to any stage by simply opening the corresponding session.

To start a new session, see “Starting a New Session in the GUI” on page 11-4.

For more information about the steps for using the System Identification Tool
GUI, see “Steps for Using the System Identification Tool GUI” on page 11-2.

11-3
 11 System Identification Tool GUI

Starting a New Session in the GUI

To start a new session in the System Identification Tool GUI, type the
following command in the MATLAB Command Window:

ident

Alternatively, you can start a new session by selecting

Start > Toolboxes > System Identification > System Identification
Tool GUI in the MATLAB desktop. This action opens the System
Identification Tool GUI.

Note Only one session can be open at a time.

You can also start a new session by closing the current session using
File > Close session. This toolbox prompts you to save your current session
if it is not already saved.

11-4
 Starting and Managing GUI Sessions

Description of the System Identification Tool Window

The following figure describes the different areas in the System Identification
Tool GUI.

Data Board Model Board

Select check Select check

boxes to display boxes to display
data plots. model plots.

The layout of the window organizes tasks and information from left to right.
This organization follows a typical workflow, where you start in the top-left
corner by importing data into the System Identification Tool GUI using

11-5
 11 System Identification Tool GUI

the Import data menu and end in the bottom-right corner by plotting the
characteristics of your estimated model on model plots. For more information
about using the System Identification Tool GUI, see “Steps for Using the
System Identification Tool GUI” on page 11-2.

The Data Board area, located below the Import data menu in the System
Identification Tool GUI, contains rectangular icons that represent the data
you imported into the GUI.

The Model Board, located to the right of the <--Preprocess menu in the
System Identification Tool GUI, contains rectangular icons that represent the
models you estimated or imported into the GUI. You can drag and drop model
icons in the Model Board into open dialog boxes.

Opening a Saved Session

You can open a previously saved session using the following syntax:

ident(session,path)

session is the file name of the session you want to open and path is the
location of the session file. Session files have the extension .sid. When the
session file in on the matlabpath, you can omit the path argument.

If the System Identification Tool GUI is already open, you can open a session
by selecting File > Open session.

Note If there is data in the System Identification Tool GUI, you must
close the current session before you can open a new session by selecting
File > Close session.

Saving, Merging, and Closing Sessions

The following table summarizes the menu commands for saving, merging, and
closing sessions in the System Identification Tool GUI.

11-6
 Starting and Managing GUI Sessions

Task Command Comment

Close the File > Close session You are prompted to save the
current current session before closing
session and it.
start a new
session.
Merge the File > Merge session You must start a new session
current and import data or models
session with before you can select to merge
a previously it with a previously saved
saved session. session. You are prompted
to select the session file to
merge with the current. This
operation combines the data
and the models of both sessions
in the current session.
Save the File > Save Useful for saving the session
current repeatedly after you have
session. already saved the session once.
Save the File > Save As Useful when you want to save
current your work incrementally. This
session under command lets you revert to a
a new name. previous stage, if necessary.

Deleting a Session
To delete a saved session, you must delete the corresponding session file.

Getting Help in the GUI

System Identification Tool GUI provides online help topics that you can
access from the Help menu.

Contextual help is available from each dialog box by clicking the Help button
in the dialog box.

11-7
 11 System Identification Tool GUI

Exiting the System Identification Tool GUI

To exit the System Identification Tool GUI, click Exit in the bottom-left
corner of the window.

Tip Alternatively, select File > Exit System Identification Tool GUI.

11-8
 Managing Models in the GUI

Managing Models in the GUI

In this section...
“Importing Models into the GUI” on page 11-9
“Viewing Model Properties” on page 11-10
“Renaming Models and Changing Display Color” on page 11-11
“Organizing Model Icons” on page 11-11
“Deleting Models in the GUI” on page 11-12
“Exporting Models from the GUI to the MATLAB Workspace” on page 11-13

Importing Models into the GUI

You can import System Identification Toolbox models from the MATLAB
workspace into the System Identification Tool GUI. If you have Control
System Toolbox software, you can also import any models (LTI objects) you
created using this toolbox.

The following procedure assumes that you begin with the System
Identification Tool GUI already open. If this window is not open, type the
following command at the prompt:

ident

To import models into the System Identification Tool GUI:

1 In the System Identification Tool GUI, select Import from the Import
models list to open the Import Model Object dialog box.

2 In the Enter the name field, type the name of a model object. Press Enter.

3 (Optional) In the Notes field, type any notes you want to store with this
model.

4 Click Import.

5 Click Close to close the Import Model Object dialog box.

11-9
 11 System Identification Tool GUI

Viewing Model Properties

You can get information about each model in the System Identification Tool
GUI by right-clicking the corresponding model icon.

The Data/model Info dialog box opens. This dialog box describes the contents
and the properties of the corresponding model. It also displays any associated
notes and the command-line equivalent of the operations you used to create
this model.

Tip To view or modify properties for several models, keep this window
open and right-click each model in the System Identification Tool GUI. The
Data/model Info dialog box updates when you select each model.

11-10
 Managing Models in the GUI

Renaming Models and Changing Display Color

You can rename a model and change its display color by double-clicking the
model icon in the System Identification Tool GUI.

The Data/model Info dialog box opens. This dialog box describes both the
contents and the properties of the model. The object description area displays
the syntax of the operations you used to create the model in the GUI.

To rename the model, enter a new name in the Model name field.

You can also specify a new display color using three RGB values in the Color
field. Each value is between 0 to 1 and indicates the relative presence of
red, green, and blue, respectively. For more information about specifying
default data color, see “Customizing the System Identification Tool GUI” on
page 11-20.

Tip As an alternative to using three RGB values, you can enter any one of
the following letters in single quotes:

'y' 'r' 'b' 'c' 'g' 'm' 'k'

These strings represent yellow, red, blue, cyan, green, magenta, and black,
respectively.

Finally, you can enter comments about the origin and state of the model in
the Diary And Notes area.

To view model properties in the MATLAB Command Window, click Present.

Organizing Model Icons

You can rearrange model icons in the System Identification Tool GUI by
dragging and dropping the icons to empty Model Board rectangles.

Note You cannot drag and drop a model icon into the data area on the left.

11-11
 11 System Identification Tool GUI

When you need additional space for organizing model icons, select
Options > Extra model/data board in the System Identification Tool GUI.
This action opens an extra session window with blank rectangles. The new
window is an extension of the current session and does not represent a new
session.

Tip When you import or estimate models and there is insufficient space for
the icons, an additional session window opens automatically.

You can drag and drop model icons between the main System Identification
Tool GUI and any extra session windows.

Type comments in the Notes field to describe the models. When you save a
session, as described in “Saving, Merging, and Closing Sessions” on page 11-6,
all additional windows and notes are also saved.

Deleting Models in the GUI

To delete models in the System Identification Tool GUI, drag and drop the
corresponding icon into Trash. Moving items to Trash does not permanently
delete these items.

11-12
 Managing Models in the GUI

To restore a model from Trash, drag its icon from Trash to the Model Board
in the System Identification Tool GUI. You can view the Trash contents
by double-clicking the Trash icon.

Note You must restore a model to the Model Board; you cannot drag model
icons to the Data Board.

To permanently delete all items in Trash, select Options > Empty trash.

Exiting a session empties Trash automatically.

Exporting Models from the GUI to the MATLAB

Workspace
The models you create in the System Identification Tool GUI are not available
in the MATLAB workspace until you export them. Exporting is necessary
when you need to perform an operation on the model that is only available
at the command line. Exporting models to the MATLAB workspace also
makes them available to the Simulink software or another toolbox, such as
the Control System Toolbox product.

To export a model to the MATLAB workspace, drag and drop the

corresponding icon to the To Workspace rectangle.

11-13
 11 System Identification Tool GUI

When you export models to the MATLAB workspace, the resulting variables
have the same name as in the System Identification Tool GUI.

11-14
 Working with Plots in the System Identification Tool GUI

Working with Plots in the System Identification Tool GUI

In this section...
“Identifying Data Sets and Models on Plots” on page 11-15
“Changing and Restoring Default Axis Limits” on page 11-16
“Selecting Measured and Noise Channels in Plots” on page 11-18
“Grid and Line Styles in Plots” on page 11-19
“Opening a Plot in a MATLAB Figure Window” on page 11-19
“Printing Plots” on page 11-19

Identifying Data Sets and Models on Plots

You can identify data sets and models on a plot by color: the color of the line
in the data or model icon in the System Identification Tool GUI matches the
line color on the plots.

You can also display data tips for each line on the plot by clicking a plot curve
and holding down the mouse button.

Note You must disable zoom by selecting Style > Zoom before you can
display data tips. For more information about enabling zoom, see “Magnifying
Plots” on page 11-16.

The following figure shows an example of a data tip, which contains the name
of the data set and the coordinates of the data point.

11-15
 11 System Identification Tool GUI

Data Tip on a Plot

Changing and Restoring Default Axis Limits

There are two ways to change which portion of the plot is currently in view:

• “Magnifying Plots” on page 11-16

• “Setting Axis Limits” on page 11-17

Magnifying Plots
Enable zoom by selecting Style > Zoom in the plot window. To disable zoom,
select Style > Zoom again.

Tip To verify that zoom is active, click the Style menu. A check mark should
appear next to Zoom.

You can adjust magnification in the following ways:

11-16
 Working with Plots in the System Identification Tool GUI

• To zoom in default increments, left-click the portion of the plot you want to
center in the plot window.
• To zoom in on a specific region, click and drag a rectangle that identifies
the region for magnification. When you release the mouse button, the
selected region is displayed.
• To zoom out, right-click on the plot.

Note To restore the full range of the data in view, select

Options > Autorange in the plot window.

Setting Axis Limits

You can change axis limits for the vertical and the horizontal axes of the input
and output channels that are currently displayed on the plot.

1 Select Options > Set axes limits to open the Limits dialog box.

2 Specify a new range for each axis by editing its lower and upper limits. The
limits must be entered using the format [LowerLimit UpperLimit]. Click
Apply. For example:

[0.1 100]

Note To restore full axis limits, select the Auto check box to the right
of the axis name, and click Apply.

3 To plot data on a linear scale, clear the Log check box to the right of the
axis name, and click Apply.

Note To revert to base-10 logarithmic scale, select the Log check box to
the right of the axis name, and click Apply.

4 Click Close.

11-17
 11 System Identification Tool GUI

Note To view the entire data range, select Options > Autorange in the
plot window.

Selecting Measured and Noise Channels in Plots

Model inputs and outputs are called channels. When you create a plot of
a multivariable input-output data set or model, the plot only shows one
input-output channel pair at a time. The selected channel names are
displayed in the title bar of the plot window.

Note When you select to plot multiple data sets, and each data set contains
several input and output channels, the Channel menu lists channel pairs
from all data sets.

You can select a different input-output channel pair from the Channel menu
in any System Identification Toolbox plot window.

The Channel menu uses the following notation for channels: u1->y2 means
that the plot displays a transfer function from input channel u1 to output
channel y2. System Identification Toolbox estimates as many noise sources as
there are output channels. In general, e@ynam indicates that the noise source
corresponds to the output with name ynam.

For example, e@y3->y1 means that the transfer function from the noise
channel (associated with y3) to output channel y2 is displayed. For more
information about noise channels, see “Subreferencing Measured and Noise
Models” on page 3-121.

Tip When you import data into the System Identification Tool GUI, it is
helpful to assign meaningful channel names in the Import Data dialog box.
For more information about importing data, see “Importing Data into the
GUI” on page 2-14.

11-18
 Working with Plots in the System Identification Tool GUI

Grid and Line Styles in Plots

There are several Style options that are common to all plot types. These
include the following:

• “Grid Lines” on page 11-19

• “Solid or Dashed Lines” on page 11-19

Grid Lines
To toggle showing or hiding grid lines, select Style > Grid.

Solid or Dashed Lines

To display currently visible lines as a combination of solid, dashed, dotted,
and dash-dotted line style, select Style > Separate linestyles.

To display all solid lines, select Style > All solid lines. This choice is the
default.

All line styles match the color of the corresponding data or model icon in
the System Identification Tool GUI.

Opening a Plot in a MATLAB Figure Window

The MATLAB Figure window provides editing and printing commands for
plots that are not available in the System Identification Toolbox plot window.
To take advantage of this functionality, you can first create a plot in the
System Identification Tool GUI, and then open it in a MATLAB Figure
window to fine-tune the display.

After you create the plot, as described in “Plotting Models in the GUI” on page
8-5, select File > Copy figure in the plot window. This command opens the
plot in a MATLAB Figure window.

Printing Plots
To print a System Identification Toolbox plot, select File > Print in the plot
window. In the Print dialog box, select the printing options and click OK.

11-19
 11 System Identification Tool GUI

Customizing the System Identification Tool GUI

In this section...
“Types of GUI Customization” on page 11-20
“Displaying Warnings While You Work” on page 11-20
“Saving Session Preferences” on page 11-20
“Modifying idlayout.m” on page 11-21

Types of GUI Customization

The System Identification Tool GUI lets you customize the window behavior
and appearance. For example, you can set the size and position of specific
dialog boxes and modify the appearance of plots.

You can save the session to save the customized GUI state.

Advanced users might choose to edit the file that controls default settings, as
described in “Modifying idlayout.m” on page 11-21.

Displaying Warnings While You Work

In the System Identification Tool GUI, select Options > Warnings to display
informational dialog boxes while you work. Verify that a check mark appears
to the right of Warnings.

To stop warnings from being displayed during your session, select

Options > Warnings and clear the check mark.

Saving Session Preferences

Use Options > Save preferences to save the current state of the System
Identification Tool GUI. This command saves the following settings to a
preferences file, idprefs.mat:

• Size and position of the System Identification Tool GUI

• Sizes and positions of dialog boxes

11-20
 Customizing the System Identification Tool GUI

• Four recently used sessions

• Plot options, such as line styles, zoom, grid, and whether the input is
plotted using zero-order hold or first-order hold between samples

You can only edit idprefs.mat by changing preferences in the GUI.

The idprefs.mat file is located in the same folder as startup.m, by default.

To change the location where your preferences are saved, use the midprefs
command with the new path as the argument. For example:

midprefs('c:\matlab\toolbox\local\')

You can also type midprefs and browse to the desired folder.

To restore the default preferences, select Options > Default preferences.

Modifying idlayout.m
Advanced users might want to customize the default plot options by editing
idlayout.m.

To customize idlayout.m defaults, save a copy of idlayout.m to a folder in

your matlabpath just above the ident folder level.

Caution Do not edit the original file to avoid overwriting the idlayout.m
defaults shipped with the product.

You can customize the following plot options in idlayout.m:

• Order in which colors are assigned to data and model icons

• Line colors on plots
• Axis limits and tick marks
• Plot options, set in the plot menus
• Font size

11-21
 11 System Identification Tool GUI

Note When you save preferences using Options > Save preferences
to idprefs.mat, these preferences override the defaults in idlayout.m.
To give idlayout.m precedence every time you start a new session, select
Options > Default preferences.

11-22
 Index

A
Index confidence interval
active impulse response plot 8-35
model in GUI 8-5 model output plot 8-15
advice noise spectrum plot 8-47
for data 2-87 residual plot 8-23
for models 8-7 step response plot 8-35
AIC 8-58 confidence interval on plots 8-62
definition 8-59 constructor 1-15
Akaike’s Final Prediction Error (FPE) 8-58 continuous-time models
Akaike’s Information Criterion (AIC) 8-58 supported 1-10
Algorithm property 1-18 continuous-time process models 3-20
algorithms for estimation Control System Toolbox
recursive 7-6 combining model objects 9-6
spectral models 3-5 converting models to LTI objects 9-4
aliasing effects 2-105 for compensator design 9-3
AR 6-7 LTI Viewer 9-5
ARMA 6-7 reducing model order 9-3
ARMAX 3-42 correlation analysis 3-11
ARX 3-42 covariance 8-61
ARX Model Structure Selection window 3-54 CovarianceMatrix 8-61
cra 3-14
cross-validation 8-3
B
best fit
D
definition 8-12
negative value 8-13 D matrix 3-90
BJ model. See Box-Jenkins model d2c 3-113
Bode plot 8-40 d2d 3-113
Box-Jenkins model 3-42 data
Burg’s method 6-11 creating iddata object 2-49
creating idfrd object 2-69
creating subsets 2-33
C detrending 2-97
c2d 3-113 exporting to MATLAB workspace 2-47
canonical parameterization 3-92 filter 2-109
complex data 2-133 frequency-domain 2-9
concatenating frequency-response 2-11
iddata objects 2-66 importing into System Identification Tool
idfrd objects 2-73 GUI 2-14
models 3-125 managing in GUI 2-14

Index-1
 Index

merging 2-35 time-series 6-1

missing data 2-93 transient response 3-11
multiexperiment data 2-35 uncertainty 8-61
outliers 2-94 EstimationInfo property 1-19
plotting 2-78 etfe
renaming in GUI 2-43 algorithm 3-5
resampling 2-102 export
sampling interval 2-30 data to MATLAB workspace 2-47
segmentation 7-14 model to MATLAB workspace 11-13
selecting 2-89
simulating 2-117
F
supported types 2-2
time-domain 2-7 filtering data 2-109
time-series 2-8 forgetting factor algorithm 7-10
transforming domain 2-121 FPE 8-58
viewing properties in GUI 2-42 free parameterization 3-84
Data Board 11-6 frequency resolution 3-6
arranging icons 2-45 frequency response
deleting icons 2-46 estimating in the GUI 3-3
data tip 11-15 etfe 3-5
dead time 3-39 spa 3-5
delay spafdr 3-5
estimating for polynomial models 3-48 frequency-domain data 2-9
detrending data 2-97 frequency-response data 2-11
discrete-time models frequency-response plot 8-38
supported 1-11 Bode plot 8-41
Nyquist plot 8-43

E
H
estimating models
black-box polynomial 3-39 Hammerstein-Wiener models 4-49
commands 1-12
frequency response 3-2 I
Hammerstein-Wiener 4-49
idarx 1-16
linear grey-box 5-6
iddata
nonlinear ARX 4-8
concatenating 2-66
nonlinear grey-box 5-16
creating 2-49
process models 3-20
subreferencing 2-57
recursive estimation 7-2
ident 11-4
state-space 3-73
idfrd

Index-2
 Index

concatenating 2-73 M
creating 2-69 MDL 3-53
model 1-16 merging
subreferencing 2-72 data 2-35
idgrey 1-16 models 3-129
idlayout.m 11-21 methods 1-14
idnlarx 1-16 missing data 2-93
idnlgrey 1-16 model
idnlhw 1-16 black-box polynomial 3-39
idpoly 1-16 estimating frequency response 3-2
idproc 1-16 estimating process model 3-20
idss 1-16 estimating transient response 3-11
importing exporting to MATLAB workspace 11-13
data into System Identification Tool grey-box estimation 5-1
GUI 2-14 Hammerstein-Wiener estimation 4-49
impulse response importing into GUI 11-9
computing values 3-15 linear grey-box estimation 5-6
confidence interval 8-35 managing in GUI 11-9
definition 3-11 nonlinear ARX estimation 4-8
estimating in the GUI 3-12 nonlinear black-box estimation 4-1
impulse 3-14 nonlinear grey-box estimation 5-16
impulse-response plot 8-30 ordinary difference equation 5-1
independence test 8-21 ordinary differential equation 5-1
plotting 8-4
K properties 1-17
recursive estimation 7-2
K matrix 3-90
reducing order using balred 9-3
Kalman filter algorithm 7-8
reducing order using pole-zero plot 8-57
refining linear parametric 3-104
L renaming in GUI 11-11
linear grey-box models 5-6 state-space 3-73
linear models time-series 6-1
extracting numerical data 3-109 uncertainty 8-61
transforming between continuous and validating 8-2
discrete time 3-113 viewing properties in GUI 11-10
transforming between structures 3-118 Model Board 11-6
LTI Viewer 9-5 arranging icons 11-11
deleting icons 11-12
model object
concatenating 3-125

Index-3
 Index

definition 1-14 offset levels 2-97

instantiating 1-15 order. See model order
merging 3-129 outliers 2-94
methods 1-14 Output-Error model 3-42
properties 1-17
types of 1-16
P
model order
definition 3-39 pem
estimating for polynomial models 3-48 for polynomial models 3-61
estimating for state-space 3-79 for process models 3-27
too high or too low 8-64 for state-space models 3-87
Model Order Selection window 3-83 periodogram
model output etfe for time series 6-5
confidence interval 8-15 physical equilibrium 2-97
model output plot 8-8 plot
model properties copy to MATLAB Figure window 11-19
accessing 1-20 data 2-78
help on 1-22 data tip 11-15
specifying 1-19 in LTI Viewer 9-5
multiexperiment data 2-35 models 8-4
models in the GUI 8-5
print 11-19
N selecting noise channels 11-18
noise pole-zero cancelation 8-57
converting to measured channels 3-123 pole-zero plot 8-53
evidence in estimated model 8-66 polynomial models 3-39
subreferencing 3-121 estimating order 3-48
noise spectrum for time-series 6-7
confidence interval 8-47 print plot 11-19
noise spectrum plot 8-46 process model 3-20
nonlinear ARX models 4-8 definition 3-20
nonlinear grey-box models 5-16 properties
nonlinear models 4-1 for models 1-17
nonlinearity estimators
troubleshooting 8-65
R
normalized gradient algorithm 7-11
recursive estimation 7-2
reducing model order
O using balred 9-3
OE model. See Output-Error model using pole-zero plot 8-57

Index-4
 Index

refining models canonical parameterization 3-92

linear parametric 3-104 estimating order 3-79
renaming data 2-43 for time series 6-12
resampling data 2-102 free parameterization 3-84
avoiding aliasing 2-105 structured parameterization 3-94
residual analysis supported parameterization 3-78
confidence interval 8-23 step response
plot 8-21 computing values 3-15
residuals confidence interval 8-35
plotting using the System Identification definition 3-11
Tool 8-26 estimating in the GUI 3-12
Rissanen’s Minimum Description Length step 3-14
(MDL) 3-53 step-response plot 8-30
robust criterion structured parameterization 3-94
for outliers 2-94 subreferencing
iddata objects 2-57
idfrd objects 2-72
S
model channels 3-120
sampling interval 2-30 model noise channels 3-121
saving session preferences 11-20 models 3-120
segmentation of data 7-14 System Identification Tool GUI
selecting data 2-89 customizing 11-20
session exiting 11-8
definition 11-3 help 11-7
managing in GUI 11-3 open 11-4
preferences 11-20 plots 11-15
starting 11-4 window 11-5
simulating data 2-117 workflow 11-2
Simulink 10-2 System Identification Toolbox blocks 10-2
slident 10-3 for data 10-4
spa for model identification 10-5
algorithm 3-5 for simulating models 10-6
spafdr open 10-3
algorithm 3-5
spectral analysis 3-2
algorithm 3-5 T
frequency resolution 3-6 time-domain data 2-7
spectrum normalization 3-8 time-series data 2-8
spectrum normalization 3-8 time-series models 6-1
state-space models 3-73 transforming data domain 2-121

Index-5
 Index

troubleshooting models 8-64 comparing model output 8-8

complicated nonlinearities 8-68 residual analysis 8-21
high noise content 8-66 troubleshooting 8-64
missing inputs 8-68 Validation Data 2-31
model order 8-64
nonlinearity estimators 8-65
W
unstable models 8-66
warnings 11-20
whiteness test 8-21
U Working Data 2-31
uncertainty of models 8-61
confidence interval on plots 8-62
X
covariance 8-61
unnormalized gradient algorithm 7-11 X0 matrix 3-90
unstable models 8-66
Y
V Yule-Walker approach 6-11
validating models 8-2

Index-6