摘要
挑选合适的地震动记录是开展结构体抗震性能评估的重要前提。常用的地震动选择方法一般用于单次地震情景下的地震动挑选,而不适用于主余震序列情景。为此,基于2 268条余震记录分析了余震谱加速度相关性,并基于分析结果,提出了余震谱加速度相关系数预测模型。在此基础上,根据条件谱匹配法建立了主余震序列型地震动选择方法。研究结果表明:所建立的相关系数预测模型的预测误差较低,可以很好地表征余震谱加速度相关性;所提方法选出的地震动数据集在均值和标准差等多个指标上满足目标反应谱要求,表明该地震动选择方法准确性较好。基于条件谱匹配构建的主余震序列型地震动选择方法进一步完善了现有选波体系,可为开展主余震序列作用下结构的动态时程分析提供指导。
Selecting suitable ground motion records is crucial for seismic structural performance assessment.Existing ground motion selection methods are primarily applicable to single earthquake scenarios and cannot address mainshock-aftershock sequencescenarios.To address this issue,2268aftershock records were analyzed to investigate the spectral acceleration correlations of aftershocks.Based on the analysis,a predictive model for aftershock spectralacceleration correlation coefficients was proposed.A ground motion selection method for mainshock-aftershock sequences was then developed using the conditional spectrum matching approach.The results demonstrate that the proposed predictive model exhibits low prediction errors,effectivelycharacterizing the correlation trends of aftershock spectral accelerations.The selected ground motion suites satisfy the target response spectrum requirements in terms of mean values,standard deviations,and other metrics,indicating the high accuracy of the proposed method.The ground-motion selection method based on conditional spectrum matching further improves the existing selection framework and provides guidance for dynamic time-history analysis of structures subjected to mainshock-aftershock sequences.
作者
白治朋
文勇波
贺秀儒
钟洪星
张晨阳
杜文琪
BAI Zhipeng;WEN Yongbo;HE Xiuru;ZHONG Hongxing;ZHANG Chenyang;DU Wenqi(Three Gorges Jinsha River Yunchuan Hydmpower Development Co.,Lid.,Kunming 650021,China;State Key Laboratory of Water Reaources Engineering and Management,Wuhan Univensity,Wuhan 430072,China)
出处
《世界地震工程》
北大核心
2025年第2期21-31,共11页
World Earthquake Engineering
基金
三峡金沙江云川水电开发有限公司禄劝乌东德电厂资助项目(Z522302014)。
关键词
主余震序列
地震动选择
条件谱
相关性分析
预测模型
mainshock-aftershock sequences
ground motion selection
conditional spectrum
correlation analysis
predictive model
