Validation of SCEC seismogram simulations: Validation Metrics and Recorded Events

Size: px

Start display at page:

Download "Validation of SCEC seismogram simulations: Validation Metrics and Recorded Events"

Claribel Miranda O’Connor’
5 years ago
Views:

1 Validation of SCEC seismogram simulations: Validation Metrics and Recorded Events Farzin Zareian, Huda Munjy, Peng Zhong, Carmine Galasso (UC Irvine) Nicolas Luco, Sanaz Rezaeian, Rob Graves (USGS) Gregory Deierlein, Kuanshi Zhong, Wen-Yi Yen (Stanford) Ting Lin (Marquette), Fabio Silva, Phil Maechling, Christine Goulet (USC/SCEC)

2 Ground Motion Simulation Validation History Ground Motion Validation Exercises ) Comparison between waveform shapes. (comparing wiggles) a) Spectral Acceleration and EDPs of recorded data from past earthquakes. (Sa rec to Sa sim, EDP rec to EDP sim ) b) Spectral Acceleration of simulated motions to empirical ground motion models. (Sa sim to Sa GMPE ) c) Estimation of average of maxidr using Building Code (ASCE 7) Nonlinear Response History Procedure. (conditioned on similarity of Sa rec to Sa sim ) d) Estimation of probabilistic distribution function of maxidr and PFA for loss estimation based on FEMA P-58. (conditioned on similarity of Sa rec to Sa sim )

3 Ground Motion Simulation Validation a) Spectral Acceleration and EDPs of recorded data from past earthquakes. For GP 21 C. Galasso, F. Zareian, I. Iervolino, R. W. Graves; Validation of Ground Motion Simulations for Historical Events Using SDoF Systems. Bulletin of the Seismological Society of America ; 12 (6): doi:

Ground Motion Simulation Validation a) Spectral Acceleration and EDPs of recorded data from past earthquakes. For GP 21 Galasso, C., Zhong, P., Zareian, F., Iervolino, I.

4 Ground Motion Simulation Validation a) Spectral Acceleration and EDPs of recorded data from past earthquakes. For GP 21 Galasso, C., Zhong, P., Zareian, F., Iervolino, I. and Graves, R. W. (213), Validation of ground-motion simulations for historical events using MDoF systems. Earthquake Engng Struct. Dyn., 42: doi:1.12/eqe.2278

5 Ground Motion Simulation Validation b) Spectral Acceleration of empirical ground motion models. Lisa M. Star, S.M.EERI, Jonathan P. Stewart, M.EERI, and Robert W. Graves, M.EERI (211) Comparison of Ground Motions from Hybrid Simulations to NGA Prediction Equations. Earthquake Spectra: May 211, Vol. 27, No. 2, pp

6 Ground Motion Simulation Validation c) Estimation of average of maxidr using Building Code (ASCE 7) Nonlinear Response History Procedure. W24X94 W24X131 W21X68 W24X84 W27X94 W27X94 W24X94 W24X131 W21X68 W24X84 W27X94 W27X94 Story number MAX IDR of 4 storeys Models(DBE Type B) Mean of Record Mean of GP Sim Story number MAX IDR of 8 storeys Models(DBE Type B) Mean of Record Mean of GP Sim W24X131 W27X94 W27X94 W3X116 W24X162 W3X116 W3X116 W24X131 W24X162 W3X18 W3X116 W3X18 8-Story Story number EDP(max(IDR)) MAX IDR of 4 storeys Models(MCE Type B) Mean of Record Mean of GP Sim EDP(max(IDR)) Story number EDP(max(IDR)) MAX IDR of 8 storeys Models(MCE Type B) Mean of Record Mean of GP Sim EDP(max(IDR)) P. Zhong, F. Zareian; Applicability of Simulated Ground Motion Waveforms for Building-Code Applications. for Earthquake Spectra;

GM Waveform Parameters & Metrics c) Estimation of average of maxidr using Building Code (ASCE 7) Nonlinear Response History Procedure. Northridge Event Baker, J. W. (27).

7 GM Waveform Parameters & Metrics c) Estimation of average of maxidr using Building Code (ASCE 7) Nonlinear Response History Procedure. Northridge Event Baker, J. W. (27). Quantitative Classification of Near-Fault Ground Motions Using Wavelet Analysis. Bulletin of the Seismological Society of America, 97(5), Shahi, S. (213). A probabilistic framework to include the effects of near-fault directivity in seismic hazard assessment. Ph.D. Thesis, Stanford University, Stanford, CA.

8 GM Waveform Parameters & Metrics c) Estimation of average of maxidr using Building Code (ASCE 7) Nonlinear Response History Procedure.

9 Ground Motion Simulation Validation d) Estimation of probabilistic distribution function of maxidr and PFA for loss estimation based on FEMA P C.O.V. Loss.15.1 Conditioned Median (EDP) 3-story Office Return Period

Ground Motion Simulation Validation Bed Rock Motions At Site Level Motion SDOF Response MDOF Response IMs: IMs: IMs: EDPs: Duration. Frequency Content. Energy Measures. Evolutionary Measures.

10 Ground Motion Simulation Validation Bed Rock Motions At Site Level Motion SDOF Response MDOF Response IMs: IMs: IMs: EDPs: Duration. Frequency Content. Energy Measures. Evolutionary Measures. Scalar or Vector. Duration. Frequency Content. Energy Measures. Evolutionary Measures. Scalar or Vector. Energy Demands. Deformation Demands. Evolutionary Measures. Scalar or Vector. Energy Measures. Deformation Demands. Force Demands. Evolutionary Measures. Scalar or Vector.

com/explore/science%2clipart%2professor/

11 Ground Motion Simulation Validation Green's Function RZZ parameters Images:

GM Waveform Parameters & Metrics Metrics Cumulative energy Cumulative number of zero-level up crossings Cumulative number of pos. minima and neg.

12 GM Waveform Parameters & Metrics Metrics Cumulative energy Cumulative number of zero-level up crossings Cumulative number of pos. minima and neg. maxima Parameters Green's Function RZZ parameters P = D I, ω ω a 5 95, I a,, tmid mid, D5 95 Sanaz Rezaeian, Peng Zhong, Stephen Hartzell, Farzin Zareian; Validation of Simulated Earthquake Ground Motions Based on Evolution of Intensity and Frequency Content. Bulletin of the Seismological Society of America ; 15 (6): doi:

13 GM Waveform Parameters & Metrics.1 I a = π 2g t n a 2 dt : Arias intensity a(t) D Four Parameters: I a D 5 95 I a : Effective duration (from 5% to 95% I a ) : Input energy t 2 a( τ ) dτ i a.45i a.95i a I a : Rate of input energy D 5 95 t mid : Middle of strong shaking (~ at 45% I a ) Sanaz Rezaeian, Peng Zhong, Stephen Hartzell, Farzin Zareian; Validation of Simulated Earthquake Ground Motions Based on Evolution of Intensity and Frequency Content. Bulletin of the Seismological Society of America ; 15 (6): doi:

14 GM Waveform Parameters & Metrics A measure of predominant frequency: Cumulative number of zero-level up crossings Recorded 3 Fitted points Count 2 Second degree polynomial 1 ω acceleration 1 Two Parameters: Time ω mid : Frequency of strong shaking phase Time, s ω' : Rate of change of frequency Sanaz Rezaeian, Peng Zhong, Stephen Hartzell, Farzin Zareian; Validation of Simulated Earthquake Ground Motions Based on Evolution of Intensity and Frequency Content. Bulletin of the Seismological Society of America ; 15 (6): doi:

15 GM Waveform Parameters & Metrics A measure of bandwidth: Cumulative number of positive minima and negative maxima Count Simulation using.3 damping ratio Recorded Time, s acceleration Time y(t) (a) narrow-band y(t) (b) wide-band Time, s Sanaz Rezaeian, Peng Zhong, Stephen Hartzell, Farzin Zareian; Validation of Simulated Earthquake Ground Motions Based on Evolution of Intensity and Frequency Content. Bulletin of the Seismological Society of America ; 15 (6): doi:

16 GM Waveform Parameters & Metrics For each evolutionary characteristic: Recorded Simulated Average Error =ε = Shape Error =ν = Area Recorded - Simulated Area (Recorded) Area(Recorded - Simulated) Area Recorded - Simulated Comparison: Two component error vector: E(ε,ν) No difference between Sim. & Rec. E(,a) Sim. & Rec. follow the same trend with a shift. Sim. & Rec. follow different trends with no shift. E(b,±1) E(c,) Sanaz Rezaeian, Peng Zhong, Stephen Hartzell, Farzin Zareian; Validation of Simulated Earthquake Ground Motions Based on Evolution of Intensity and Frequency Content. Bulletin of the Seismological Society of America ; 15 (6): doi:

17 GM Waveform Parameters & Metrics 2 Non-synched Acc time series 2 Synched Acc time series 25 Fourier Spectra 1.5 Recorded Simulated 1.5 Recorded Simulated Acceleration, g Acceleration, g Frequency Amplitude Time, s Time, s Frequency, Hz.35 Metric 1 8 Metric 2 1 Metric Evolution of Intensity Northridge Event AvgError:.13.5 ShapeError:-.5 Evolution of Frequency AvgError:.2 ShapeError:.93 Evolution of Bandwidth AvgError:.19 ShapeError: Time, s Time, s Time, s

18 GM Waveform Validation Northridge Event Rec: 1/17/1994 Sim: 3/ Stations, 5 realizations

19 GM Waveform Validation I a = t 2 a( τ ) dτ RZZ Arias Intensity (Realization1) ln I a Recorded GP SONG SDSU EXSIM

20 GM Waveform Validation I a = t 2 a( τ ) dτ Arias Intensity-GP Method-Realization 1 1 Simulated I a Recorded I a

21 GM Waveform Validation Arias Intensity-GP Method-Realization 1 1 Arias Intensity-SONG Method-Realization Simulated I a.5.25 Simulated I a Recorded I a Recorded I a Arias Intensity-SDSU Method-Realization 1 1 Arias Intensity-EXSIM Method-Realization Simulated I a.5.25 Simulated I a Recorded I a Recorded I a

22 GM Waveform Validation 5% to 95% Duration 3 25 D Recorded GP SONG SDSU EXSIM

23 GM Waveform Validation 4 D GP Method-Realization 1 Simulated D Recorded D 5-95

24 GM Waveform Validation 4 D GP Method-Realization 1 4 D SONG Method-Realization 1 Simulated D Simulated D Recorded Recorded 4 D SDSU Method-Realization 1 4 D EXSIM Method-Realization 1 Simulated D Simulated D Recorded Recorded

25 GM Waveform Validation ε = ν = Area Recorded - Simulated Area (Recorded) Area(Recorded - Simulated) Area Recorded - Simulated

26 GM Waveform Validation How do these IMs affect EDPs? Used SMRFs (2-, 4-, 8-, 12-story) Focus on maxidr and PFA. W16X31 W16X31 Multiple regression analysis to investigate the relationship between GM parameters and EDPs. W24X131 W3X132 W24X162 W3X132 The aim is to: a) Identify if Sim and Rec motions generate the same trend in EDP response. b) Are the differences statistically significant.

27 GM Waveform Validation Ln(IDR)=b +b 1 (I a )+b 2 (D 5-95 )+b 3 (ω mid )+b 4 (ω mid )+ε Story Recorded EXSIM GP SDSU Song 1 I a I a I a D 5-95 I a 2 I a I a I a D 5-95 I a

28 GP, 2-story GM Waveform Validation W16X31 W16X31 W24X131 W3X132 W24X162 W3X132

29 GP, 2-story GM Waveform Validation W16X31 W16X31 W24X131 W3X132 W24X162 W3X132

30 GP, 2-story GM Waveform Validation W16X31 W16X31 W24X131 W3X132 W24X162 W3X132

31 GP, 12-story GM Waveform Validation W24X84 W24X84 W24X84 W24X84 W24X94 W24X84 W27X94 W27X94 W24X131 W27X94 W24X131 W27X94 W3X116 W3X116 W24X131 W3X116 W24X162 W3X116 W3X116 W3X116 W24X146 W3X116 W24X176 W3X116 W3X132 W3X132 W24X162 W3X132 W24X27 W3X132 W24X27 W3X132 W3X124 W24X27 W3X132 W3X124

32 GP, 12-story GM Waveform Validation W24X84 W24X84 W24X84 W24X84 W24X94 W24X84 W27X94 W27X94 W24X131 W27X94 W24X131 W27X94 W3X116 W3X116 W24X131 W3X116 W24X162 W3X116 W3X116 W3X116 W24X146 W3X116 W24X176 W3X116 W3X132 W3X132 W24X162 W3X132 W24X27 W3X132 W24X27 W3X132 W3X124 W24X27 W3X132 W3X124

33 GP, 12-story GM Waveform Validation W24X84 W24X84 W24X84 W24X84 W24X94 W24X84 W27X94 W27X94 W24X131 W27X94 W24X131 W27X94 W3X116 W3X116 W24X131 W3X116 W24X162 W3X116 W3X116 W3X116 W24X146 W3X116 W24X176 W3X116 W3X132 W3X132 W24X162 W3X132 W24X27 W3X132 W24X27 W3X132 W3X124 W24X27 W3X132 W3X124

34 GM Waveform Validation GP 2-story SONG SDSU EXSIM

35 GM Waveform Validation GP 12-story SONG SDSU EXSIM

36 GM Waveform Validation Summary and Conclusions based on current information: Parameters and Metrics for waveform validation are proposed and utilized for improvement of simulation models. Arias Intensity (and Spectral acceleration) are parameters that can describe the differences between drift response of building structures to comparable simulated and recorded ground motions.

Quantitative Identification of Near-Fault Ground Motion using Baker s Method; an Application for March 2011 Japan M9.0 Earthquake

Cite as: Tazarv, M., Quantitative Identification of Near-Fault Ground Motion using Baker s Method; an Application for March 2011 Japan M9.0 Earthquake, Available at: http://alum.sharif.ir/~tazarv/ Quantitative