Non-Linear Analysis Untar 040206

1

Non-Linear Analysis & Performance-based Design: Current State-of-practice in Earthquake EngineeringShort SeminarUniversitas Tarumanagara, Jakarta 6 February 2004Leonardi KawidjajaSenior AssociatePT Rekacipta KinematikaConsulting Engineerswww.kinematika.com

Current Global Trends• More complex structures• Higher performance requirements• More complete earthquake records• Better understanding of structural non-linear

behaviour• Lessons learned from major earthquakes• Advanced earthquake protection technology• Advanced material technology• Public awareness on earthquake hazard

Earthquake Engineering in 21st Century

towards “Performance-based” Design

2

Earthquake Engineering in 21st Century

Traditional Design• Force-based• Force Reduction Factor (R)• Linear Analysis• “Extrapolated” Structural Behaviour

Earthquake Engineering in 21st Century

Nature is Non Linear

Traditional Design• Force-based• Force Reduction Factor (R)• Linear Analysis• “Extrapolated” Structural Behaviour

Modern Design Methodology• Displacement-based• Performance Design• Non-linear Analysis• “Real” Structural Behaviour

3

Earthquake Engineering in 21st Century

Modern Design Methodology• Displacement-based• Performance Design• Non-linear Analysis• “Real” Structural Behaviour

Traditional Design• Force-based• Force Reduction Factor (R)• Linear Analysis• “Extrapolated” Structural Behaviour

Performance-based Design

• Performance Objectives– Earthquake Risk– Acceptable Damage

• Code Provisions– SEAOC “Blue Book”– ATC-40– FEMA 356

• Japanese Practice• Acceptance Criteria

– Plastic deformation– Drift ratio

CPLS

IO

OP

4

Non-Linear Analysis

• Analysis Procedure:– Non-linear Static:

pushover analysis– Non-linear Dynamic:

time history analysis• Geometric Non-linearity:

– P-delta effect– tension structures

• Material Non-linearity:– Elastoplastic material– Viscoelastic material– Hysteretic damping

Non-Linear Analysis: Static Pushover

Pushover Forces

Plastic Hinges Formation

5

Non-Linear Analysis: Static Pushover

Performance Point :• Approximate displacement of the

building in the desired Performance Objective

• Bilinear Coefficient Method– Graphical & Analytical– δT = C0 C1 C2 C3 (4π2)/(Te

2) Sa W

• ADRS Method– Acceleration-Displacement

Response Spectrum– Analytical

Performance Point

Non-Linear Analysis: Static Pushover

Acceptance Criteria?• Inelastic Deformation

– plastic hinge rotation, plastic strain, etc• Total Deformation:

– drift, total strain, etc

@ Performance Point !

6

Non-Linear Analysis: Dynamic Time History

Earthquake time-history record

Plastic hinge formation in “real time”

Non-Linear Analysis: Dynamic Time History

Scaled Response Spectra

Peak Ground Acceleration

Scaled Time History

7

Non-Linear Analysis: Dynamic Time History

Acceptance Criteria?• Inelastic Deformation

– plastic hinge rotation, plastic strain, etc• Total Deformation:

– drift, total strain, etc

Analysis Methodology

Static Pushover• Less complex• Similar to traditional method• “Pushover” force• No dynamic behaviour• No energy dissipation• No cyclic stiffness

degradation• Conventional structure

Dynamic Time History• Very complex• Numerical simulation• Real earthquake input• Non-linear dynamic behaviour• Hysteretic energy dissipation• Cyclic strength & stiffness

degradation• Any structure

8

Non-Linear Analysis ToolsStatic Pushover Analysis• DRAIN 2DX• RAM X-Linea• SAP2000 NL-Push• etcDynamic Time History Analysis• SAP2000 NL-Push• LS-DYNA• ADINA• ABAQUS• LUSAS• IDARC• etc

SAP2000 NL-PUSH

• Non-Linear Static & Dynamic Implicit FEA Code• By Computer & Structures Inc (CSI), CA• Features:

– Integrated ATC-40 Static Pushover protocol– Non-linear Dynamic Time History analysis– Non-linear frame elements for Static

Pushover– Non-linear “link” elements for Time History

• Platform: Windows • Practical Seismic Analysis & Design

9

SAP2000 NL-PUSH

Ford Otosan Golcük, Turkey: RC Moment FrameNon-Linear Static Pushover Analysis

SAP2000 NL-PUSH

Maison Hermés, Tokyo: stepping columns with viscoelastic dampersNon-Linear Dynamic Time History Analysis

10

LS-DYNA

• Non-Linear DYNAMIC Time-Domain Explicit FEA Code

• Source code by Livermore Software, CA• Impact, Metal Forming, Blast, Vibration, Earthquake,

Wave, Wind & other dynamic loadings• Platform: Unix (64-bit) or WindowsNT (32-bit)• “Heavy-duty” dynamic analysis for complex structures

Non

-line

ar

Sol

ver

LS-DYNA

Gra

phic

al P

re-p

roce

ssor

s

Gra

phic

al P

ost-p

roce

ssor

s

D3PLOTPRIMER

THISOASYS SHELL

11

LS-DYNA

Maison Hermés, Tokyo: stepping columns with viscoelastic dampersNon-Linear Dynamic Time History Analysis

LS-DYNA

Maison Hermés, Tokyo: stepping columns with viscoelastic dampersNon-Linear “Pseudo-Static” Pushover Analysis

12

LS-DYNA: Performance Verification

Menara Jakarta, Indonesia• 588 m Telecommunications Tower• To be the tallest free standing structure in

the world• Construction halted in 1997• Long period• RC mega-columns & spandrels• RC perforated shear walls• Response to long period ground motion

from strong distant earthquake

LS-DYNA: Performance Verification

• 2000 yrs EQ• 100 sec EQ• Seismic Beam• Elastoplastic Shell

13

LS-DYNA: Performance Based Design

Toyota Football Stadium, Toyota, Japan• 40,000 seat stadium• Retractable Roof• Kisho Kurokawa Architects• Cable-stayed Grand Stand Roof• Unbonded Braced Frame• 2-level performance design:

– Level 1: pgv = 25 kine– Level 2: pgv = 50 kine

• Hyotei Special Building Permit Process

LS-DYNA: Performance Based Design

• Tension Cables• Cable Pretension• Unbonded Brace• Steel Moment

Frame• Inelastic Spring• Seismic Beam

14

LS-DYNA: Design of Seismic Isolation

San Francisco Civic Center Sculpture, San Francisco, CA• 13 m tall• Light-weight tension wire “wraps”• Slender compression column• UBC Zone 4, EPA = 0.4g• Friction Pendulum base isolation

LS-DYNA: Design of Seismic Isolation

• Wire pretension• Base isolation• Inelastic Springs

15

LS-DYNA: Dynamic Soil-Structure Interaction Analysis

Core Pacific City, Taipei, Taiwan• Multi-storey commercial

complex• Multi-level basement• Piled foundation• Soil-pile interaction

modelling• Reduced pile design!

LS-DYNA: Dynamic Soil-Structure Interaction Analysis

• Rock ground motion

• Hysteretic soil model

• Seismic beam piles

• Seismic beam frames

16

LS-DYNA: Dynamic Soil-Structure Interaction Analysis

• Rock ground motion

• Hysteretic soil model

• Seismic beam piles

• Seismic beam frames

LS-DYNA: Pounding Analysis

P&G Sanipak Production Building, Gebze, Turkey

• Single-storey cantilever column building• Near fault strong ground motion• Six separate compartments• Insufficient gap to avoid pounding• Mid-level and Roof-level pounding

17

LS-DYNA: Pounding Analysis

• Gap element: compression only spring

• Seismic beam• Elastoplastic shell

LS-DYNA: Pounding Analysis

18

LS-DYNA: Rocking AnalysisSpent Nuclear Fuel Flask Storage Platform

• Dynamic overturning• Gravity stabilised

structures• Two overturning

bodies:– Fuel Flask– Platform

• Contact surfaces

LS-DYNA: Rocking AnalysisSpent Nuclear Fuel Flask Storage Platform

• Dynamic overturning• Gravity stabilised

structures• Two overturning

bodies:– Fuel Flask– Platform

• Contact surfaces

19

LS-DYNA: Virtual Testing

P&G Takasaki Building 51, Takasaki, Japan

• Seven-storey steel frame building• FEMA 273 seismic retrofit• Latticed beams & columns• Numerical simulations of cyclic and

monotonic testing of latticed beam & column sub-assemblies

• Ductility parameters

LS-DYNA: Virtual Testing

• Repeated incremental cyclic loading

• Bilinear elastoplastic shell elements

• Non-linear springs for rivet shear & tension

• Contact surface

20

CURRENT STATE-OF-PRACTICE

• Not yet required by standard design codes• ALTERNATIVE design methodology• High-Performance or High-Tech structures• Evaluation of Existing Structures• Risk Assessment• Expensive: manhours & software• Limited knowledge on non linear concepts• Limited skills on non linear methodologies

Performance-based Design Exotica

THE FUTURE

• Performance-Based Design Codes• Taller, Bigger, Crooker, Better • “Active” Seismic Protection Device• Advanced Material Technology• Advanced Structural Optimisation• Powerful computing power• Lower computing cost• Advanced, affordable analysis codes• Man (woman) behind the machines• Designer or Operator?

21

www.tarumanagara.ac.idwww.kinematika.com

© 2004