Benefits and Detriments of Soil Foundation Structure Interaction: Simulation Platform...

Hypothesis Computational Platform SFSI Case Study Summary

Benefits and Detriments ofSoil Foundation Structure Interaction:

Simulation Platform and Examples

Boris Jeremic

Department of Civil and Environmental EngineeringUniversity of California, Davis, U.S.A.

Graduate Students: Guanzhou Jie and Matthias Preisig

Funding: NSF–CMS–0324661, NSF–TeraGrid,NSF–EEC–9701568

Jeremic Computational Geomechanics Group

Benefits and Detriments of Soil Foundation Structure Interaction: Simulation Platform and Examples


Outline

HypothesisSFS System Evolution

Computational PlatformSoftware ComponentHardware Component

SFSI Case StudyHigh Fidelity, 3D ModelsBehavior for Short Period MotionsBehavior in Long Period Motions

Summary




SFS System Evolution

SFSI HypothesisI Flexibility (elastic) of foundations and soils modifies

dynamic properties of the SFS system (Gazetas andMylonakis)

I Reduction in stiffness (elasto–plasticity) of the SFS systemmodifies those dynamic properties even more so

I Earthquake period and SFS period might (will) coincideI SFSI response a function of a tightly coupled triad of

dynamic characteristic ofI EarthquakeI SoilI Structure

I Use detailed numerical models to study prototype models




Software Component

Parallel Elastic–Plastic Finite Element Computations

I Current Parallel FEM areI Well developed for elastic FEMI Undeveloped for elastic–plastic FEMI Well developed for homogeneous distributed memory

parallel (DMP) computers,I Undeveloped for multiple performance (multi–generation)

DMPsI Need: dynamic computational load balancing for

I multiple element types,I multiple material modelsI multiple compute node performancesI multiple network performances




Software Component

Plastic Domain Decomposition (PDD) MethodI Multi-objective optimization problem (minimize both the

inter-processor communications, the data redistributioncosts and create balanced partitions)

I Parallel program based on UCD CompGeoMech libraries(elements, mat. models, algorithms, PDD method),ParMETIS (graph repartitioning), PETSc (system solvers)and parts of OpenSees (upgraded analysis model)

I Adaptive domain repartitioning depending on computenode and network performance

I Scalable to a large number of CPUs (2–1024 or more)

I Performance tuning and production runs on my clusterGeoWulf (UCD), LongHorn (TACC) and DataStar (SDSC)




Software Component

Speedup Overview

0 8 16 24 32 40 48 56 64 72 80 88 96 104 112 120 1280

8

16

24

32

40

48

56

64

72

80

88

96

104

112

120

128

Number of CPUs

Par

alle

l Sp

eed

up Theoretic

al Speedup

398,721 DOFs (125,225 Elements)

118,275 DOFs (36,037 Elements)







Hardware Component

Parallel Computer GeoWulfI Distributed memory

parallel computerI Multiple generation

compute nodesand networks

I Very cost effective!I Same architecture as

large parallel supercomputers(SDSC, TACC, EarthSimulator...)

I Local design, construction,available at all times!




High Fidelity, 3D Models

Detailed 3D FEM Model (one of)I Soils:

elastic–plasticsolids(Drucker-Prager,Armstrong-Frederick)

I Structureand piles:non–linearbeam–columnelements(fiber element)





Detailed 3D FEM ModelI Construction processI Two types of soil: stiff soil (UT, UCD), soft soil (Bay Mud)I Deconvolution of given surface ground motionsI Use of the DRM (Prof. Bielak et al.) for seismic inputI Piles→ beam-column elements in soil holesI No artificial damping (only mat. dissipation, radiation)I Structural model: UCB (Fenves), UW (Eberhardt)I Element size issues (filtering of frequencies)

elem. # elem. size fcutoff min. Gep/Gmax γ

12K 1.00 m 10 Hz 1.0 <0.5 %15K 0.90 m >3 Hz 0.08 <1.0 %

150K 0.30 m 10 Hz 0.08 <1.0 %500K 0.15 m 10 Hz 0.02 <5.0 %





Northridge and Kocaeli Input Motions

0 10 20 30 40 50 60

−2

0

2

Time (s)

Acc

eler

atio

n (m

/s2 ) Acceleration Time Series − Input Motion (NORTHRIDGE EARTHQUAKE, 1994)

0 10 20 30 40 50 60−0.2

0

0.2

Time (s)

Dis

plac

emen

t (m

) Displacement Time Series − Input Motion (NORTHRIDGE EARTHQUAKE, 1994)

0 5 10 15 20 25 30 35

−2

0

2

Time (s)

Acc

eler

atio

n (m

/s2 ) Acceleration Time Series − Input Motion (TURKEY KOCAELI EARTHQUAKE, 1999)

0 5 10 15 20 25 30 35−0.4

−0.2

0

0.2

Time (s)

Dis

plac

emen

t (m

) Displacement Time Series − Input Motion (TURKEY KOCAELI EARTHQUAKE, 1999)

1 2 3 4 5 6 7 8 9 100

1

2

Period (s)Fou

rier

Am

plitu

de (

m/s

2 )

Acceleration Frequency Content − Input Motion (NORTHRIDGE EARTHQUAKE, 1994)

1 2 3 4 5 6 7 8 9 100

0.2

0.4

Period (s)Fou

rier

Am

plitu

de (

m)

Displacement Frequency Content − Input Motion (NORTHRIDGE EARTHQUAKE, 1994)

2 4 6 8 10 12 14 16 18 200

2

4

Period (s)Fou

rier

Am

plitu

de (

m/s

2 )

Acceleration Frequency Content − Input Motion (TURKEY KOCAELI EARTHQUAKE, 1999)

2 4 6 8 10 12 14 16 18 200

2

4

Period (s)Fou

rier

Am

plitu

de (

m)

Displacement Frequency Content − Input Motion (TURKEY KOCAELI EARTHQUAKE, 1999)





Simulation Results

0 10 20 30 40 50 60

−0.2

−0.1

0

0.1

0.2

0.3

Time (s)

Dis

plac

emen

t (m

)

SSS CCC SSC CCS SCS CSC SCC CSS Freefield Input Motion

0 10 20 30 40 50 60

−0.2

−0.1

0

0.1

0.2

0.3

Time (s)

Dis

plac

emen

t (m

)

SSS CCC SSC CCS SCS CSC SCC CSS

0 10 20 30 40 50 60

−0.2

−0.1

0

0.1

0.2

0.3

Time (s)

Dis

plac

emen

t (m

)


0 10 20 30 40 50 60

−0.2

−0.1

0

0.1

0.2

0.3

Time (s)

Dis

plac

emen

t (m

)


0 10 20 30 40 50 60

−0.2

−0.1

0

0.1

0.2

0.3

Time (s)

Dis

plac

emen

t (m

)


0 10 20 30 40 50 60

−0.2

−0.1

0

0.1

0.2

0.3

Time (s)

Dis

plac

emen

t (m

)


0 10 20 30 40 50 60−3

−2

−1

0

1

2

3

Time (s)

Acc

eler

atio

n (m

/s2 )


0 10 20 30 40 50 60−3

−2

−1

0

1

2

3

Time (s)

Acc

eler

atio

n (m

/s2 )


0 10 20 30 40 50 60−3

−2

−1

0

1

2

3

Time (s)

Acc

eler

atio

n (m

/s2 )


0 10 20 30 40 50 60−3

−2

−1

0

1

2

3

Time (s)

Acc

eler

atio

n (m

/s2 )


0 10 20 30 40 50 60−3

−2

−1

0

1

2

3

Time (s)

Acc

eler

atio

n (m

/s2 )


0 10 20 30 40 50 60−3

−2

−1

0

1

2

3

Time (s)

Acc

eler

atio

n (m

/s2 )


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.2

0.4

0.6

0.8

1

Period (s)

Fou

rier

Am

plitu

de (

m)


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.5

1

1.5

Period (s)

Fou

rier

Am

plitu

de (

m/s

2 )


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Period (s)

Fou

rier

Am

plitu

de (

m)


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.2

0.4

0.6

0.8

1

1.2

1.4

Period (s)

Fou

rier

Am

plitu

de (

m/s

2 )


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Period (s)

Fou

rier

Am

plitu

de (

m)


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Period (s)

Fou

rier

Am

plitu

de (

m/s

2 )


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

1

2

3

4

5

Period (s)

Fou

rier

Am

plitu

de (

m/s

2 )


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

2

4

6

8

10

Period (s)

Fou

rier

Am

plitu

de (

m/s

2 )


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

1

2

3

4

5

6

Period (s)

Fou

rier

Am

plitu

de (

m/s

2 )


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

1

2

3

4

5

6

7

8

Period (s)

Fou

rier

Am

plitu

de (

m/s

2 )


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

0.5

1

1.5

2

2.5

3

3.5

4

Period (s)

Fou

rier

Am

plitu

de (

m/s

2 )


0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

2

4

6

8

10

Period (s)

Fou

rier

Am

plitu

de (

m/s

2 )


0 10 20 30 40 50 60−4000

−3000

−2000

−1000

0

1000

2000

3000

4000

Time (s)

Mom

ent (

kN*m

)


0 10 20 30 40 50 60−4000

−3000

−2000

−1000

0

1000

2000

3000

4000

Time (s)

Mom

ent (

kN*m

)


0 10 20 30 40 50 60−4000

−3000

−2000

−1000

0

1000

2000

3000

4000

Time (s)

Mom

ent (

kN*m

)


0 10 20 30 40 50 60−4000

−3000

−2000

−1000

0

1000

2000

3000

4000

Time (s)

Mom

ent (

kN*m

)


0 10 20 30 40 50 60−4000

−3000

−2000

−1000

0

1000

2000

3000

4000

Time (s)

Mom

ent (

kN*m

)


0 10 20 30 40 50 60−4000

−3000

−2000

−1000

0

1000

2000

3000

4000

Time (s)

Mom

ent (

kN*m

)





Behavior for Short Period Motions

Short Period E.: Left Bent, Structure and Soil, Disp.

0 10 20 30 40 50 60

−0.2

−0.1

0

0.1

0.2

0.3

Time (s)

Dis

plac

emen

t (m

)


0 10 20 30 40 50 60

−0.2

−0.1

0

0.1

0.2

0.3

Time (s)

Dis

plac

emen

t (m

)






Short Period E.: Left Bent, Bending Moments

0 5 10 15 20 25−4000

−3000

−2000

−1000

0

1000

2000

3000

4000

Time (s)

Mom

ent (

kN*m

)

SSS CCC





Short Period E.: Left Bent, Free Field vs Real Disp.

0 5 10 15 20 25

−0.1

−0.05

0

0.05

0.1

0.15

Time (s)

Dis

plac

emen

t (m

)

SSS CCC Freefield Input Motion




Behavior in Long Period Motions

Long Period E.: Left Bent, Bending Moments.

0 5 10 15 20 25 30 35−4000

−3000

−2000

−1000

0

1000

2000

3000

4000

Time (s)

Mom

ent (

kN*m

)

SSS CCC




Behavior in Long Period Motions

Long Period E.: Left Bent, Structure and Soil, Disp.

0 5 10 15 20 25 30 35−0.8

−0.6

−0.4

−0.2

0

0.2

0.4

0.6

0.8

Time (s)

Dis

plac

emen

t (m

)

SSS CCC Freefield Input Motion




SummaryI High fidelity numerical models of ESS systems

I High performance computational tools (software andhardware) developed and available

I Matching Triad: Earthquake, Soil and Structure (ESS)interaction determines possible benefits or detriments ofSFSI

I Program sources and toolsavailable in public domain (GPL)at Author’s web site



Benefits and Detriments of Soil Foundation Structure Interaction: Simulation Platform...

Documents

Transcript of Benefits and Detriments of Soil Foundation Structure Interaction: Simulation Platform...