摘要
本文基于日本KiK-Net,K-Net台网及汶川地震中6000余条浅壳与上地幔地震竖向加速度记录,分场地类型建立了包含阻尼比和谱周期两个参数的阻尼修正系数模型,用以调整无震级和震源距信息的设计谱。基于固定效应模型对阻尼修正系数对数值及阻尼比对数值的三次多项式进行线性回归得到模型表达式,基于随机效应模型对阻尼修正系数模型的残差及残差的标准差进行分离,分析震源、路径以及场地效应对模型误差的影响。研究结果表明:高阻尼长周期震源效应引起的残差大于其他效应引起的残差;场地效应引起的残差小于其他效应引起的残差。对模型残差的标准差及阻尼比对数值进行线性回归,回归方程易于计算模型残差的标准差。
In this article,based on over 6000 vertical ground-motion records from shallow crustal and upper-mantle earthquakes in Japan obtained by two strong-motion networks,the KiK-Net and K-Net,and the Wenchuan earthquake.A damping modification factor(DMF)model is presented for four site classes,accounting for the damping ratio and spectral periods.This model can be used to adjust a design spectrum that is not associated with a particular earthquake magnitude and distance.The effect of damping ratios was modeled by a third order polynomials for the logarithms of damping ratio in the DMF model.The model coefficients were derived by a fixed effects regression analysis.Based on a random effects model,the total residuals and total standard deviation were separated into between-event and within-event parts.These residuals were used to examine the influences of source,path and site effects on the model standard deviations.These analysis results show that the between-event standard deviations from the source effect at long periods with larger damping ratios are larger than those associated with the path and site effects.The between-site standard deviations are smaller than those for source and path effects.All types of standard deviations can be described by a linear function of damping ratios in the logarithm scale and this equation can be used to calculate the standard deviation of model residuals.
作者
张洪博
刘名吉
康莉莉
ZHANG Hongbo;LIU Mingji;KANG Lili(School of Civil Engineering,Shandong Jianzhu University,Jinan 250100,China)
出处
《世界地震工程》
CSCD
北大核心
2021年第3期155-169,共15页
World Earthquake Engineering
基金
国家自然科学基金资助项目(51878396)。
关键词
阻尼比修正系数
浅壳和上地幔地震
竖向加速度谱
随机效应模型
damping modification factor
shallow crustal and upper-mantle earthquakes
vertical acceleration response spectrum
random effects model
