场地条件校正的地震动快速评估方法研究

地表峰值加速度PGA是工程地震预警、烈度速报及结构抗震设计的重要指标参数。搜集整理日本Kik-net台网的地震记录,以我国抗震规范中采用的场地覆盖层厚度D和等效剪切波速Vse为场地特征参数,以基岩加速度峰值PBA为输入地震动强度指标,基于分类回归树CART算法,建立了以地表PGA阈值分别为40 gal、80 gal及120 gal的场地条件修正的地震动预警方法,并给出了参数指标的取值范围。通过对搜集的数据进行回判检验,3种地震强度下总体预警成功率分别为84.7%、90.1%和93.6%。采用我国川滇地区强震数据验证该方法的可靠性,结果显示总体成功率分别为88.4%、92.3%和93.4%。

Fast assessment method for ground motion considering site condition correction

Peak ground acceleration (PGA) is an important index for earthquake early warning, intensity quick report and structural aseismic design.Here, earthquake records of KiK-net in Japan were collected and arranged.The site cover thickness D and equivalent shear wave velocity Vse adopted in our country’s aseismic code were taken as site characteristic parameters, and peak bedrock acceleration (PBA) was taken as input earthquake dynamic strength index, based on the classification and regression tree (CART) method, a ground motion early warning method considering site condition correction under PGA’s thresholds of 40 gal, 80 gal and 120 gal, respectively was proposed, and ranges of parameter indexes were given.Through check-back tests for the data collected from KiK-net in Japan, it was shown that the total earthquake early warning success rates are 84.7%, 90.1% and 93.6% corresponding to PGA values of 40 gal, 80 gal or 120 gal, respectively.The strong earthquake data collected in our country’s Sichuan and Yunnan areas were used to verify the reliability of the proposed method, the results showed that the total earthquake early warning success rates are 88.4%, 92.3% and 93.4%, respectively under 3 seismic intensities mentioned above.