PAST, PRESENT AND FUTURE

OF
EARTHQUAKE ANALYSIS OF STRUCTURES
By Ed Wilson

September 24, 2014

SEAONC Lecture #2

Terminology in nonlinear analysis

Which does not have a unique definition
1. Equal Displacement Rule
2. Pushover Analysis
3. Equivalent Linear Damping
4. Equivalent Static Analysis
5. Nonlinear Spectrum Analysis
6. Onerous Response History Analysis

Summary of Lecture Topics

Field measurements of frequencies and mode shapes
Exact Eigenvectors or Approximate Ritz Vectors
The Load Dependent Ritz Vectors LDR Vectors
The Fast Nonlinear Analysis Method FNA Method
Error Estimation Conservation of Energy
Foundation Structure Interaction
The Retrofit of the Richmond-San Rafael Bridge
Recommendations

FIELD MEASUREMENTS
REQUIRED TO VERIFY
1. MODELING ASSUMPTIONS
2. SOIL-STRUCTURE MODEL
3. COMPUTER PROGRAM
Resulted in program modification

CHECK OF RIGID
DIAPHRAGM
APPROXIMATION
MECHANICAL
VIBRATION
DEVICES

FIELD MEASUREMENTS OF
PERIODS AND MODE SHAPES
MODE

TFIELD

TANALYSIS

Diff. - %

1.77 Sec.

1.78 Sec.

0.5

1.69

1.68

0.6

1.68

1.68

0.0

0.60

0.61

0.9

0.60

0.61

0.9

0.59

0.59

0.8

0.32

0.32

0.2

11

0.23

0.32

2.3

FIRST DIAPHRAGM
MODE SHAPE
15 th Period
TFIELD = 0.16 Sec.

Simple Example of Dynamic Response

With no external load applied
u( 0 )

Initial displacement u( 0 )

u( 0 )

Simple Example of Free Vibration

mu( t )

u( t )

k
u( t )
2

k
u( t )
2

Equilibrium of mass
m u( t ) ku( t ) 0

Columns
in bending
Rigid Foundation

For a Static Initial Condition of

u( 0 ) 1.0

The Solution of the Force Equilibrium Equation is u( t ) cos( t )

for t 0 to

Check Solution:
Since

u( t ) 2 cos( t )

m k 0
2

Therefore

for t 0 to

Where ,

k
m

rad / sec

Or , f
cps
2

Physical Analysis of Free Vibration

Displacement u( t ) cos( t )

Velocity

u ( t ) sin( t )

1
1
2
2
E
(
t
)

ku
(
t
)

k
cos
( t )
Strain Energy s
2
2

1
1
1
2
2
2
2
Kinetic Energy Ek ( t ) mu ( t ) m sin ( t ) k sin ( t )
2
2
2

Total Energy

E( t ) Ek ( t ) E s ( t )

1
1
k [sin 2 ( t ) cos 2 t ] k
2
2

Without Energy Dissipation the System would Vibrate Forever

Energy Pump

Dynamic Mode Shapes are very important

To help you design a better structure

Rigid in-plane stiffness slabs

24 Mass DOF - 24 Modes

Modal Base Shears and Directions

and
Overturning Moments for each Mode

Regular and Irregular Structures

The current code defines an irregular structure as one
that has a certain geometric shape or in which stiffness and
mass discontinuities exist.
A far more rational definition is that a regular structure
is one in has minimum coupling between the lateral
displacements and the torsional rotations for the mode
shapes associated with the lower frequencies of the system.
Therefore, if the model is modified and tuned by studying
the three-dimensional mode shapes during the preliminary
design phase, it may be possible to convert a geometrically
irregular structure to a dynamically regular structure
from an earthquake-resistant design standpoint.

A Dynamic Irregular Structure

Plan View
Geometrically Regular Structure
Damage
Area
Non Structural Element with Large Stiffness
Unreinforced Concrete Brick Wall

Carl Gustav Jacob Jacobi (1804 1851)

was a German who made fundamental contributions to
classical mechanics, dynamics and astronomy. A crater on
the Moon is named after him.
Jacobi first presented the method for the calculation of
mode shapes and frequencies in 1846 (168 year ago).

After using the method for over 50

years, I have concluded it is the most
robust numerical method for the
calculation of mode shapes and
frequencies.

It never fails to produce results.

M=0.05

100

M=0.10

100

(a) Beam Model

I = 1.0

E=10,000

M=0.05

6 DOF System
Only one dynamic mode

(b) Rigid Body Mode

10

T1

(c) Rigid Body Mode

2 0

T2

(d) Dynamic Mode

3 0.995 T3 6.31

(e) Static Mode

T4 0

(f) Static Mode

T5 0

(g) Static Mode

T6 0

Example of the Jacobi Method for the Evaluation of

Mode Shapes and Frequencies

For the Earthquake Analysis of Structures the

Load Dependent Ritz method produces more
Accurate results than the uses of the
Exact Eigenvectors
Fewer LDR mode shapes are required
Less computation time is required
Zero and infinite frequencies are calculated, or,
dynamic, static and rigid body modes are found
The LDR vectors can be used for nonlinear
analysis

Load-Dependent Ritz Vectors

LDR Vectors 1980 2000
14.8 Page 157

MOTAVATION 3D Reactor on Soft Foundation

Dynamic Analysis - 1979
by Bechtel using SAP IV
200 Exact Eigenvalues
were Calculated and all
of the Modes were in the
foundation No Stresses
in the Reactor.
The cost for One analysis
on the CRAY Computer
was

$10,000

3 D Concrete Reactor
3 D Soft Soil Elements
360 degrees

Linear Dynamic Equilibrium Equation

M u( t ) C u ( t ) K u( t )

g j ( t ) F 0G( t )

First , solve for static displacements u0 K F0 V0

where has K LDLT has been factored
.;

The LDR Vectors are calculated by :

Solve by iteration i 1,2 ,........ N blocks
ui K 1MVi 1 Vi (this is an error estimation from previous block)
where Vi

indicates all vectors are made stiffness and

mass orthogonal using the Jacobi Method for each step i

GENERATION
DEPENDENT
1.

LOAD
VECTORS

OF

RITZ

Approximately Three Times Faster Than

The Calculation Of Exact Eigenvectors

2. Results In Improved Accuracy Using A

Smaller Number Of LDR Vectors
3. Computer Storage Requirements
Reduced
4. Can Be Used For Nonlinear Analysis To
Capture Local Static Response

DYNAMIC RESPONSE OF BEAM

100 pounds

10 AT 12" = 120"

FORCE = Step Function

TIME

MAXIMUM DISPLACEMENT
Number of Vectors

Eigen Vectors

Load Dependent Vectors

0.004572

(-2.41)

0.004726

(+0.88)

0.004572

(-2.41)

0.004591

( -2.00)

0.004664

(-0.46)

0.004689

(+0.08)

0.004664

(-0.46)

0.004685

(+0.06)

0.004681

(-0.08)

0.004685

( 0.00)

0.004683

(-0.04)

0.004685

(0.00)
( Error in Percent)

MAXIMUM MOMENT
Number of Vectors

Eigen Vectors

Load Dependent Vectors

4178

( - 22.8 %)

5907

( + 9.2 )

4178

( - 22.8 )

5563

( + 2.8 )

4946

( - 8.5 )

5603

( + 3.5 )

4946

( - 8.5 )

5507

( + 1.8)

5188

( - 4.1 )

5411

( 0.0 )

5304

( - .0 )

5411

( 0.0 )

( Error in Percent )

LDR Vector Summary

After Over 20 Years Experience Using the
LDR Vector Algorithm
We Have Always Obtained More Accurate
Displacements and Stresses
Compared to Using the Same Number of
Exact Dynamic Eigenvectors.
SAP 2000 has Both Options

The Fast Nonlinear Analysis Method

The FNA Method was Named in 1996
Designed for the Dynamic Analysis of
Structures with a Limited Number of Predefined
Nonlinear Elements

FAST NONLINEAR ANALYSIS

1. EVALUATE LDR VECTORS WITH
NONLINEAR ELEMENTS REMOVED AND
DUMMY ELEMENTS ADDED FOR STABILITY
2. SOLVE ALL MODAL EQUATIONS WITH
NONLINEAR FORCES ON THE RIGHT HAND SIDE
3. USE EXACT INTEGRATION WITHIN EACH TIME STEP
4. FORCE AND ENERGY EQUILIBRIUM ARE
STATISFIED AT EACH TIME STEP BY ITERATION

BASE ISOLATION

Base Isolators

Building Impact Analysis

Base Shear Equal Zero

FRICTION
DEVICE
CONCENTRATED
DAMPER

NONLINEAR
ELEMENT

GAP ELEMENT

BRIDGE DECK

ABUTMENT

TENSION ONLY ELEMENT

PLAS TI C
HINGES
2 ROTATIONAL DOF

DEGRADING STIFFNESS ?

Mechanical Dampers

f s ksu s

f p k pu p
f d cu

Mathematical Model

LINEAR VISCOUS DAMPING

Does not exist in normal structures and
foundations
5 or 10 percent modal damping values are
often used to justify energy dissipation due
to nonlinear effects
If energy dissipation devices are used, then 1
percent modal damping should be used for
the elastic part of
the structure - CHECK ENERGY PLOTS

Comparison with Experimental Results

General Pivot Element

Comparison with Test Results

Summary of the FNA Method

See Chapter 18 for details

Nonlinear Equilibrium Equations

M u( t ) C u ( t ) K u( t ) FN ( t )

g j ( t ) F( t )

After moving the nonlinear forces to the right hand side

M u( t ) C u ( t ) K u( t ) F ( t ) FN ( t )
In order to inprove the rate of convergence
(or to make the structure stable)
We can add an effective stiffness to both sides of the equation
M u( t ) C u ( t ) K K E u( t ) F ( t ) [ K E u( t ) FN ]

Nonlinear Equilibrium Equations

M u( t ) C u ( t ) K u( t ) F ( t ) K E u( t ) FN ( t )
Let u( t ) Y ( t ), u Y ( t ) and u( t ) Y( t )
T

After premutipli cation by the modal equations are

IY( t ) cY ( t ) 2Y ( t ) R( t ) R( t )
N

Where R( t )N T K E u( t ) T FN ( t ) must be solved by

iteration at each time " t"
All Modal Equations are integrated at the same time
since the modes are coupled by the nonlinear elements.

The deformations in the nonlinear elements can

be calculated from the following displacement:
transformation
d ( t ) bu( t ) equation:
bY ( t ) BY ( t ) where B b
(i )

Or , in iterative form , d ( t )

( i 1 )

BY ( t )

Where the size of the B array is equal to the

number of nonlinear deformations times the
number of LDR vectors. This array is calculated
only once prior to the start of mode integration.
Also, the modal forces associated with the nonlinear elements are calculated from

(i)

(i )

B f (t )

Calculate error for iteration i , at the end of

each time step, for the N
Nonlinear elements given Tolerance
N

Err =

| f (t

i
)n | -

n=1

| f (t

i -1
)n |

n=1

| f (t

i
)n |

n=1

If
If

Err Tol Continue to iterate with i i 1

Err Tol go to next time step with t t t

103 FEET DIAMETER - 100 FEET HEIGHT

NONLINEAR
DIAGONALS

BASE
ISOLATION

ELEVATED WATER STORAGE TANK

COMPUTER MODEL
92 NODES Over 500 DOF
103 ELASTIC FRAME ELEMENTS
56 NONLINEAR DIAGONAL ELEMENTS

600 TIME STEPS @ 0.02 Seconds

COMPUTER TIME
REQUIREMENTS
PROGRAM
ANSYS

INTEL 486

3 Days

ANSYS

CRAY

3 Hours

SADSAP

INTEL 486

( 4300 Minutes )

( B Array was 56 x 20 )

( 180 Minutes )
(2 Minutes )

FRAME WITH
UPLIFTING
ALLOWED

UPLIFTING ALLOWED

Four Static Load Conditions

Are Used To Start The
Generation of LDR Vectors

EQ

DL

Left

Right

NONLINEAR STATIC ANALYSIS

50 STEPS AT dT = 0.10 SECONDS

DEAD LOAD

LOAD

LATERAL LOAD

1.0

2.0

3.0

4.0

5.0

TIME - Seconds

Axial Forces in Left and Right

Columns as a Function of Time

Results With and Without Uplift

Loma Prieta Earthquake

Results With and Without Uplift

2 Times the Loma Prieta Earthquake

Advantages Of The FNA Method

1. The Method Can Be Used For Both
Static And Dynamic Nonlinear Analyses
2. The Method Is Very Efficient And
Requires A Small Amount Of
Additional Computer Time As
Compared To Linear Analysis
2.

The Method Can Easily Be Incorporated

Into Existing Computer Programs For
LINEAR DYNAMIC ANALYSIS.

FIRST LARGE APPLICTION

OF
THE FNA METHOD
Retrofit of the
RICHMOND - SAN RAFAEL BRIDGE
1997 to 2000
Using SADSAP

S
A
D
S
A
P

TATIC
ND
YNAMIC
TRUCTURAL
NALYSIS
ROGRAM

TYPICAL ANCHOR
PIER

MULTISUPPORT
ANALYSIS
( Displacements )

ANCHOR PIERS

RITZ VECTOR
LOAD
PATTERNS

ECCENTRICALLY BRACED FRAME

Comparison of Relative and Absolute Displacement Seismic Analysis

Properties:

20@15=300

Thickness = 2.0 ft
Width =20.0 ft
I = 27,648,000 in4
E = 4,000 ksi
W = 20 kips /story
Mx = 20/g

Typical Story Load

Mub (t )

= 0.05176 kip-sec2 /in

Myy = 517.6 kip-sec2 -in
Total Mass = 400 /g
Typical Story Height
h = 15 ft = 180 in.
First Story Load

x ub (t )
A. 20 Story Shear Wall
With Story Mass

B. Base Acceleration Loads

Relative Formulation

12 EI
ub (t )
3
h

First Story Moment

6 EI
ub (t )
2
h

C. Displacement Loads
Absolute Formulation

Idealized Near-Field Earthquake Motions

u(t ) g

0.50 g 0.50 g

Time

ACCELERATION

1.00 g

6 @ 0.1 Sec.

u (t ) g
VELOCITY

19.32 in./sec

u (t ) g
3.22 inches

DISPLACEMENT

Shear at Second Level Vs. Time With Zero Damping

Time Step = 0.01

140
120

Linear Acceleration Loads, or Cubic Displacement Loads - Zero Damping - 40 Modes

100

Linear Displacement Loads - Zero Damping - 40 Modes

80

SHEAR - Kips

60
40
20
0
-20
-40
-60
-80
-100
0.00

0.20

0.40

0.60

0.80
1.00
1.20
TIME - Seconds

1.40

1.60

1.80

2.00

Illustration of Mass-Proportional Component in Classical Damping .

us

ur

x
mu

Css I x u x 0

Css I x u x 0

ux
us u x ur
RELATIVE DISPLACEMENT FORMULATION

ABSOLUTE DISPLACEMENT FORMULATION

Comparison of Mode Superposition of LDR Vectors with the

Step-By-Step Solution using the Trapezoidal Rule

A Fluid is a Solid with a

Very Low Shear Modulus

Recommendations
Use realistic physical approximations to create the
mathematical model . Do parameter studies to
investigate the sensitivity when it is possible. Find out
if it matters.
Structural and Geotechnical Engineers must work
together to produce a family of realistic Earthquake
records. Do not accept a site Spectra only .
After the model is created and the loading selected,
use the most accurate solution methods available.
Computer time is free.

Eds Simple Consideration

After you understand the behavior of the structure,
take time to consider several different options to
improve the earthquake resistance of the structure.
It is my feeling there are two fundamental approaches:
.

Make the structure stiffer to move as a rigid body.

Make the structure more flexible at the base

to
. isolate it without causing irreparable
damage.
The solution must prevent collapse and be cost

You and SAP

2000
Can do your own
Research