Study On Selection Of Ground Motion Records For Uniform Risk Seismic Design

The primary goal of seismic design is to prevent building structures from collapsing and to ensure safety of human life.Until now,our country has been set the ground motion parameter with exceeding probability of 10% within 50 years as the seismic design level.However,this tradition ignores both the differences of seismic hazard between different areas caused by seismotectonic environment and the differences of collapse resistance of structures,thus can not guarantee collapse risk structures uniformly across the whole country.Therefore,the basic seismic design principle of buildings should be shifted from the tranditinal "uniform hazard" paradigm to more reasonable "uniform risk" paradigm.In addition,the selection and scaling of ground motion records is the basis of nonlinear dynamic time-history analysis and seismic collapse risk checking of structures.However,the ground motion selection and scaling method in Chinese seismic design code of buildings is relatively indistinct and obscure,which makes the operation poor for practicing engineers.This thesis systematically and comprehensively studies the selection and scaling of ground motion records for uniform risk aseismic design.From matching the horizontal seismic influence coefficient curve of the 2010 version of Chinese seismic design code of buildings,to considering the probabilistic seismic hazard analysis results of specific sites and taking site-specific response spectrum as the target spectrum,and to probabilistic seismic performance evaluation based on the new generation of Performance-Based Earthquake Engineering(PBEE),the selection and scaling of ground motion records and the evaluation of structural collapse resistance are more closely linked with seismic hazard of the target site,and can better serve for uniform-collapse-risk seismic design of buildings.The main research contents of this theses are as follows:(1)The main process of ground motion records selection and scaling based on the code design spectrum is summarized.Taking three single degree of freedom(SDOF)systems as case-study examples,the response prediction models of structures are established by using the point-of-comparison(POC)method.Then different methods of ground motion selection and scaling are evaluated by comparing the accuracy and variations of structural responses,thus the relatively reasonable methods are chosen for structures with different natural vibration periods.In addition,an alternative set of ground motion records is constructed under certain conditions with an improved method.For time history analysis in the work of structural seismic design,ground motion records can be directly selected according to natural vibration period of the structure,which will save time and provide convenience for actual engineering projects.(2)Taking uniform risk spectrum of a specific site as the target spectrum,the point-of-comparison method is used to predict structural responses under the specified intensity measure,and the different selection methods of ground motion records are evaluated by the accuracy and variations of structural responses.On this basis,the SDOF systems are taken as the research objects,and different elastoplastic response parameters are selected to invesitage the effects of three different sitespecific spectra,namely,uniform hazard spectrum(UHS),conditional mean spectrum(CMS)and uniform risk spectrum(URS),on the elastoplastic response parameters of the SDOF systems.(3)In the new generation of Performance-Based Earthquake Engineering(PBEE)research framework,the ground motion records selection and scaling method for probabilistic performance assessment method in FEMA-P58 report is combined with the multiple stripes analysis(MSA)method to obtain probabilistic seismic demand hazard analysis results,which is in good agreement with the seismic hazard analysis results.Based on the results of probabilistic seismic hazard analysis in Xi’an region,the seismic demand hazard analysis results using the traditional cloud analysis method(CAM)are compared with those using multiple stripes analysis(MSA)method for SDOF systems.It is shown by the comparison results that the traditional cloud analysis method(CAM)overestimates seismic demand risk of structures.(4)Convolving probabilistic seismic hazard analysis curves results of the buildings site with the collapse fragility curves of structures,the seismic collapse risk of structures in the design reference period 50 years can be evaluated.Taking an eightstoreys reinforced concrete(RC)frame structure as case-study example,the collapse probability of the structure is calculated by different ground motion records selection methods and collapse fragility analysis methods,considering the seismic hazard analysis results of the site.Then the seismic collapse risk of the structure is quantitatively evaluated according to two cirteria: one is the aboslute collapse probability within 50 years should be less than 1%,another is the relative conditional collapse probability of the struture under the maximum considered earthquake(MCE)should be less than 10%.