PEER Seismic Simulation: Advances with OpenSees Gregory L. Fenves University of California, Berkeley PEER Annual Meeting January 18, 2002.

PPEEEERR

Seismic Simulation:Advances with OpenSees

Gregory L. FenvesUniversity of California, Berkeley

PEER Annual MeetingJanuary 18, 2002

Simulation in PBEE

• Performance Based Engineering depends on evaluation seismic demands (EDP) and damage (DM).

• Rational, validated models of behavior of structural and geotechnical materials, components and systems are needed for simulating performance.

• Simulation applications:– Assessment– Design using parameterized models– Reliability-based design

• Improved software needed for PBEE methodologies.• OpenSees is PEER’s software framework for seismic

simulation to support these applications.

Status of OpenSees

• All information available on website at opensees.berkeley.edu– Documentation– Source code browser– Downloads (source code, Win32 exe)

• Version 1.2 is the current release• Development versions ahead of release, but most

developments available in CVS (source code versioning system) for checkout.

OpenSees in 2001

• Nonlinear static and dynamic analysis of 3D structural and soil-foundation systems.

• Increase robustness and performance.• Begin to support simulation needs for testbeds.• Continue as research platform for modeling of

non-ductile components, particularly degrading shear behavior.

• Work on documentation and user/developer workshops.

Open-Source CommunitySimulation Framework

Conceptual Approach for Simulation

InformationTechnology Software framework,

Databases, Visualization,Internet/grid computation

ComputationAlgorithms,Solvers,Parallel/distributedcomputing

ModelsMaterial, component, system models

Simulation models,Performance models,Limit state models

Summary of Models

• Elements– Beam-column elements for 2D and 3D based on force

formulation– Zero-length elements– Plate elements (new)– 2D and 3D continuum elements (including mixed

formulation for incompressibility)• Materials

– Library of 1D uniaxial models• Fairly general hysteretic models• New p-y models

– Constitutive models (e.g. J2 plasticity, soil models)

Elementup

Basic System

v q

q1,v1q3,v3

q2,v2

GeometricTran

u

p=afT q

p

u=afu

LinearLinearPDCorotational (Filippou)

Beam-Column Geometry

Section

e s

e, s

s = asTσ

A∫ dA

)(εσ=σ

ε =ase

Material

ε σ

No assumptions are made on section or material behavior; each level in the hierarchy can be defined independently of other levels

σ

ε

Beam-Column Models

s =bq

v = bTedx0

L

∫e=av

q= aTsdx0

L

∫Basic System

v qq1,v1q3,v3

q2,v2

Displacement Force

Aggregation of Section Model

e=

ε0

κ

γ

⎡

⎣

⎢ ⎢ ⎢

⎤

⎦

⎥ ⎥ ⎥

s =

P

M

V

⎡

⎣

⎢ ⎢ ⎢

⎤

⎦

⎥ ⎥ ⎥

ks =ks−sec 0

0∂V∂γ

⎡

⎣

⎢ ⎢

⎤

⎦

⎥ ⎥

yieldultimate

residual

M

residual

ultimate

cracking

V

0,1

ForceDeformation

SectionAggregator

UniaxialMaterial

Decorator Pattern Options for coupling: New ForceDeformation Class Inter-object communication

Models under Development

• Generalized hinge models for beam-columns (Deierlein)

• Beam-column joint models (Lowes)• Coupled shear-axial-flexure models (Filippou)• Continuing work on soil models

Computational Procedures

• Nonlinear solution algorithms:– Newton-Raphson– BFGS and Broyden’s– Other quasi-newton methods (e.g. using Krylov subspace

updates)– Line search options– Converge options and “remediation”

• Load stepping procedures:– Variety of nonlinear static, arc-length, displacement control…– Dynamic including Newmark, HHT

• Equation solvers– Several equation solvers available depending on problem

topology and computer hardware.

Simulation Services Architecture

Law, Peng, Stanford

Application: Performance of Cable-Stayed Bridge

Charles ChadwellUC Berkeley

OpenSees Model

Response to Pulse Motion

Response to Recorded Motion

TCU079 TCU129

Issues for 2002

• Models– Degradation, shear-flexure, bond slip, joints– Soil-pore fluid models– Pile-foundation models– Validation protocols– Input motions and coupled simulations

• PBEE Design and Simulation– Parameterization of models, sensitivity– Reliability

• Computing and Information Technology– Parallel and distributed computing– Visualization– Internet and “Grid Aware” applications (NEES)

PE

ER

Tes

tbed

s