| Ground motion prediction equations(GMPEs)are used to compute the response spectrum from a given earthquake source type,magnitude,source depth,source distance,and site conditions.GMPEs are used to calculated seismic hazard curves as well as the uniform hazard spectrum through a probabilistic seismic hazard analysis(PSHA),seismic zonation and determine the earthquake load for a given engineering structure.Different earthquake types in a subduction zone produce different ground motions.The ground motions from a plate boundary earthquake differ from those within the subduction plate,because of the complex plate tectonic geological settings in a subduction zone.Different GMPEs are required for different parts of subduction zones for engineering construction or seismic hazard zonation in a subduction zone.However,most GMPEs use 5% damped response spectrum as the ground motion parameters.In reality,not all the structures have a damping ratio of 5% and a damping ratio modification model is used to scale the 5% damped spectrum;the conversion brings more ground-motion uncertainty.In this study,we developed a set of GMPEs with a number of damping ratios for the horizontal component of the strong-motion records from subduction slab earthquakes.This set of GMPEs can be used for a PSHA for structures with a damping ratio other than 5% so that a more accurate estimation of the seismic loads can be achieved.3107 strong-motion records from subduction interface in Japan were used to derive a set of GMPEs for 14 damping ratios,based on the GMPE by Zhao et al.(20016a).In this thesis,a random effects model was used to carry out the regression analysis and the total residual were separated into a number of components corresponding to the between-and within-event parts.The model parameters were obtained through residual analysis,statistical tests of the regression parameters and the smoothing of model parameters.A comparison of source,path and site effect parameters among different damping ratios will be presented for GMPEs with different damping ratios.As an application example,GMPEs from this thesis were used in a PSHA example,and seismic hazard curves calculated by GMPEs with different damping ratios were compared.In order to evaluate the effects of damping ratios on PSHA and seismic resistance design in general,uniform hazard spectrums under different damping ratios were also compared with the response spectra derived by two damping modification models.Main conclusions are as follow:(1)At very short periods up to 0.04 s,parameters in the GMPEs for different damping ratios are very similar,while those parameters vary with damping ratios at spectral periods over 0.05 s;(2)At short periods up to 0.03 s,all standard deviations do not change with spectral periods and are nearly identical for all damping ratios.At the other spectral periods,all standard deviations decrease with increasing damping ratios at nearly all spectral periods.(3)The predicted spectra for different damping ratios at short periods(T<0.04 s)are essentially the same for an event with the same moment magnitude,distance,depth,and site class.At periods between 0.05 s and approximately 1.20 s,the predicted response spectrum decreases with increasing damping ratios.At spectral periods longer than 1.25 s,the predicted response spectra increase with increasing damping ratio at small magnitudes but decrease for larger magnitudes.(4)At periods over 0.04 s,seismic hazard curves and uniform hazard spectra derived by GMPEs under different damping ratios differ significantly.The uniform hazard spectra using GMPE with different damping ratios differ significantly from the response spectra derived by using damping modification factors(DMFs)to scale the 5% response spectrum from a PSHA to obtain the spectrum with an appropriate damping ratio.Seismic hazard may be underestimated in a PSHA when DMFs are used.(5)Seismic hazards for structures with different damping ratios would be more accurate when GMPEs with different damping ratios are used in a PSHA. |
PDF Full Text Request