detection of soil-structure-interaction effect by system identification€¦ · ·...

Detection of Soil-Structure-Interaction Effect by System Identification

○Hajime OkanoKobori Research Complex Inc. Tokyo, Japan

Tatsuya AzuhataNILIM, Tsukuba, Ibaraki, Japan

Seismic Network for Building and Ground

Seismic Networks for Ground K-NET,KiK-net

Facilitated by NIED after 1996 1000Site,20km average distance

Seismic Network for Building Has not facilitated yet, and its construction has been

desired. Because, there is the mismatch between the observed ground

motion and the damage of structure.

NILIM* has started to construct the seismic network for building and surrounding ground in 2010

1

* National Institute of Land and Infrastructure Management

Mismatch between Ground Motion and Damage

Large ground motion versus smaller structure damage than expected

2

1995 Kobe Earthquake(Suzuki, Okano et. al 2007)

0 20 40 60 80 1000.0

0.5

1.0C

(1) (Bf.1971)　 (Af.1972年以降)　：Observed ：Simulation

c=1.0

DDamage

Str

engt

h

Possible Factors Designed strength and real strength Evaluation of damage from response Soil Structure Interaction (SSI)

How to know SSI-effect from Seismograph ?

SSI effect is included in the seismograph, and it is difficult to disaggregate SSI effect by simple

manipulation of seismograph.

3

System Identification

System Identification for SSI

4

Soil Structure InteractionInertial Interaction (II) Kinematic Interaction (KI)

yFIM ｙＩ

z0

yFIM θFIMH1 θＩH1 x1

z1

ｙＩ

H

kHH,cHH

kRR,cRR

k1,c1

m1

m0 I0

θ0=θFIM+θＩ

0 1 2 3 4 50.0

0.5

1.0

1.5

Founda

tion Input

Mtion/Fre

e F

ield

YFIM/Y

FF

Frequency(Hz)

η=0.05

Sway-Rocking Model

sin

FF

FIMFIM Y

Y

Diffraction Function

NIT

6F RC-Frame and RC

Bearing Wall Foundation: RC Pile Surface Layer :Vs ≒

200m/s Occupancy

:Education Location :South Part

of Saitama Pref. (near Tokyo)

5

Tokyo

NIT

3/09/2014 Fore-Shock (NIT, Ridge-dir.)

6

6F／1F

6F／Ground

1F／Ground

Foundation Input

Motion／Ground

0 1 2 3 4 5

100

101

Frequency(Hz)

Am

plitud

e R

atio

k1=3.49e+06(kN/m)

h1=0.049

観測：bw=0.20Hz同定対象観測値同定モデル

0 1 2 3 4 510

-1

100

Frequency(Hz)

Am

plitud

e R

atio

観測：bw=0.20Hz同定対象観測値同定モデル入力損失（dif型）

0 1 2 3 4 5

10-1

100

Frequency(Hz)

Am

plitud

e R

atio

kH=6.71e+06(kN/m)

cH=3.97e+05(kN*s/m)

=0.071


0 1 2 3 4 5

10-1

100

Frequency(Hz)

Am

plitud

e R

atio

YFR

-YI：bw=0.20(Hz)

dif型(同定適合値)

Foundation Response- Inertial Response

Diff. Func.

Observed

Idetified

Observed

Idetified

Observed

Idetified

Diff. Func.

0 1 2 3 4 5

100

101

Frequency(Hz)

Am

plitud

e R

atio

k1=2.49e+06(kN/m)

h1=0.081


0 1 2 3 4 510

-1

100

Frequency(Hz)

Am

plitud

e R

atio


0 1 2 3 4 5

10-1

100

Frequency(Hz)

Am

plitud

e R

atio

kH=4.63e+06(kN/m)

cH=2.30e+05(kN*s/m)

=0.069


0 1 2 3 4 5

10-1

100

Frequency(Hz)

Am

plitud

e R

atio

YFR

-YI：bw=0.20(Hz)


3/11/2014 Main Shock (NIT, Ridge-dir.)

7

6F／1F

6F／Ground

1F／Ground

Foundation Input

Motion／Ground Foundation Response

- Inertial Response

Diff. Func.

Observed

Idetified

Observed

Idetified

Observed

Idetified

Diff. Func.

0 1 2 3 4 510

-1

100

101

Frequency(Hz)

Am

plitud

e R

atio

k1=2.06e+06(kN/m)

h1=0.055


0 1 2 3 4 510

-1

100

101

Frequency(Hz)

Am

plitud

e R

atio


0 1 2 3 4 5

10-1

100

Frequency(Hz)

Am

plitud

e R

atio

kH=6.77e+06(kN/m)

cH=1.47e+05(kN*s/m)

=0.073


0 1 2 3 4 5

10-1

100

Frequency(Hz)

Am

plitud

e R

atio

YFR

-YI：bw=0.20(Hz)


4/11/2014 After-Shock (NIT, Ridge-dir.)

8

6F／1F

6F／Ground

1F／Ground

Foundation Input

Motion／Ground Foundation Response

- Inertial Response

Diff. Func.

Observed

Idetified

Observed

Idetified

Observed

Idetified

Diff. Func.

Transition of Stiffness and Damping (NIT, Ridge-dir.)

9

Stiffness Damping Coefficients

0.0E+00

2.0E+06

4.0E+06

6.0E+06

8.0E+06

1.0E+07

3/09(Fore) 3.11(Main) 3.11(After1)4.11(After2)

Stiffne

ss(kN/m

)

Building k1

Soil‐Foundation kHH

0.0E+00

2.0E+05

4.0E+05

6.0E+05

3/09(Fore) 3.11(Main) 3.11(After1) 4.11(After2)

Dam

ping

Coe

fficient

（ｋN*s/m

） Building c1

Soil‐Foundation cHHSoil-FoundationBuilding

Stiffness of building: has not recovered in after main-shock Stiffness of soil-spring: has recovered in after-shock Damping coefficient of soil-spring: has not recovered after main

shock

Soil-FoundationBuilding

Effect on Maximum Building Response

Adopt r.m.s. response (standard deviation) as substitute of maximum response Assumptions

Response is stationary random Peak factor is constant

Benefit Directory calculated by identified transfer functions and

parameters

10

SSI Effect on R.M.S. Building Response

11

Transfer Function r.m.s. Building Resp.

Building vs. Foundation Input Motion (II)

Building vs. Free Filed ground motion (II+KI)

Building vs. Foundation Response(Fixed Base Response)

FIM

obsYX Y

HZZH

FIM 2001

1

FF

obsYX Y

HZZH

FF 2001

1

obs

obsX HZ

HZZH

fix00

2001

1

0 0

2

112 dSH

FIMFIM YXYX

0 0

2

112 dSH

FFFF YXYX

0 0

2

112 dSH

fixfix XX

cnst.Vp S 0S

Input Motion

approximately

Transition of SSI Effects (NIT, Ridge-dir.)

12

3/9(Fore) 3/11(Main) 3/11(After1)4/11(After2)0.0

0.2

0.4

0.6

0.8

1.0

Redu

ctio R

ate o

f r.m

.s. B

uild

ing

Resp

onse

Eq.(9): Inertial Interaction (II) Eq.(10): Kinematic Interaction (KI) Eq.(11): Inertial and Kinematic Interaction (II+KI)

Eq. Reduction by …

Kinematic Interaction (KI)

Inertial Interaction (II)

Kinematic Interaction+ Inertial Interaction (KI+II)

fixX

FIMX

1

1

fixX

FFX

1

1

FIMX

FFX

1

1

Reduction by KI is almost constant Reduction by II decreased after main shock

Conclusions

The stiffness of RC building has not recovered, but the soil spring stiffness has recovered gradually in after-shocks.

The response reductions by kinematic interaction (KI) have been almost constant in fore/main/after-shocks.

Contrary the response reductions by inertial interaction (II) have decreased after main shock.

13

detection of soil-structure-interaction effect by system identification€¦ · ·...

Documents