[1]靳超越,胡进军,胡磊,等.基于机器学习的地震动特征提取与模拟以2021年云南漾濞6.4级地震为例[J].世界地震工程,2021,(04):073-80.
 JIN Chaoyue,HU Jinjun,HU Lei,et al.Machine learning-based seismic feature extraction and simulation a case study of the 2021 Yangbi M6.4 earthquake[J].,2021,(04):073-80.
点击复制

基于机器学习的地震动特征提取与模拟以2021年云南漾濞6.4级地震为例
分享到:

《世界地震工程》[ISSN:/CN:]

卷:
期数:
2021年04期
页码:
073-80
栏目:
出版日期:
2021-11-06

文章信息/Info

Title:
Machine learning-based seismic feature extraction and simulation a case study of the 2021 Yangbi M6.4 earthquake
作者:
靳超越 胡进军 胡磊 王中伟
中国地震局工程力学研究所, 中国地震局地震工程与工程振动重点实验室, 黑龙江 哈尔滨 150080
Author(s):
JIN Chaoyue HU Jinjun HU Lei WANG Zhongwei
Key Laboratory of Earthquake Engineering and Engineering Vibration, Institute of Engineering Mechanics, China Earthquake Administration, Harbin 150080, China
关键词:
漾濞MS6.4地震地震动模拟机器学习特征提取序列型地震
Keywords:
Yangbi MS6.4 earthquakesimulation ground motionmachine learningfeature extractionsequence-type ground motions
分类号:
P315
摘要:
随着地震动数据数量的增长和质量的提高,将基于数据驱动的机器学习方法应用到地震动模拟中有重要意义。以2021年5月21日云南漾濞MS6.4地震为例,利用主成分析方法从前震及余震地震动记录中提取特征母波时程,将地震动三要素作为模拟误差约束,在求解母波的线性组合系数时使用多目标优化算法寻优,最终找到帕累托最优解作为模拟目标台站记录时的组合系数,得到模拟地震动时程。结果表明:主成分析法在对实际地震动记录进行特征提取后,得到的特征母波时程可以在一定程度上保留原始数据的主要信息;考虑幅值、频谱和持时这三要素的角度去控制模拟误差,可以使得模拟的地震动时程更加接近真实记录。提出的基于特征提取的地震动模拟方法可以为基于小震数据合成大震地震动提供参考。
Abstract:
With the increase of the quantity and quality of ground motion data, it is of great significance to apply data-driven machine learning method to ground motion simulation for the development of simulation methods. Taking the Yangbi MS6.4 earthquake on May 21, 2021 as an example, the principal component analysis method is used to extract the characteristic mother wave time histories from the foreshock-aftershock records. The three elements of ground motion are taken as the simulation error constraints, and the multi-objective optimization algorithm is used to solve the linear combination coefficients of the mother waves. Finally, the pareto optimal solution is found as the combined coefficients of the simulated target stations, and the simulated ground motion time histories are obtained. From the simulation results, it can be seen that: After feature extraction of actual ground motion records by principal component analysis, the characteristic mother waves obtained can retain the main information of original data; To control the simulation errors from the perspective of amplitude, spectrum and duration can make the simulated time histories of ground motion more close to the real records; The whole process of ground motion simulation can provide reference for the synthesis of large earthquake ground motion records based on small earthquake data.

参考文献/References:

[1] 王振宇, 赵培培, 薄景山. 地震动随机模拟方法主要影响参数分析[J].世界地震工程, 2017, 33(3):34-41.WANG Zhengyu, ZHAO Peipei, BO Jingshan. Analysis for the effects of main parameters on ground motions by stochastic simulation method[J]. World Earthquake Engineering, 2017, 33(3):34-41. (in Chinese)
[2] 刘中宪, 苗岳云, 陈頔.点震源作用下三维沉积盆地地震动谱元模拟[J].世界地震工程, 2020, 36(2):200-208.LIU Zhongxian, MIAO Yue yun, CHEN Di. Seismic spectral element simulation of three-dimensional sedimentary basin under the action of point seismic source[J]. World Earthquake Engineering, 2020, 36(2):200-208. (in Chinese)
[3] 朱景宝, 宋晋东, 李山有.基于支持向量机的2021年2月13日日本福岛近海Mj7.3级地震震级估算[J].世界地震工程, 2021, 37(2):74-81.ZHU Jingbao, SONG Jindong, LI Shanyou. Magnitude estimation for the February 13, 2021Mj7.3 earthquake near the coast of Fukushima Japan based on support vector machine.[J]. World Earthquake Engineering, 2021, 37(2):74-81. (in Chinese)
[4] KONG Q, TRUGMAN D T, ROSS Z E, et al. Machine learning in seismology:turning data into insights[J]. Seismological Research Letters, 2019, 90(1):3-14.
[5] ALIMORADI A, BECKJ L. Machine-learning methods for earthquake ground motion analysis and simulation[J]. Journal of Engineering Mechanics, 2015, 141(4):04014147.
[6] 胡进军, 张辉, 靳超越, 等.基于PCA及PSO智能算法的地震动合成方法-以中国西部中强地震为例[J].工程力学, 2021, 38(03):159-168.HU Jinjun, ZHANG Hui, JIN Chaoyue, et al. A method to simulate ground motion based on PCA and PSO intelligent algorithms- a case study of moderate magnitude earthquakes in western China[J]. Engineering Mechanics, 2021, 38(03):159-168. (in Chinese)
[7] 杨建文, 叶泵, 高琼, 等. 2021年云南漾濞Ms 6.4地震前后主动源观测走时数据的变化[J].地震工程学报, 2021(04):767-776.YANG Jianwen, YE Beng, GAO Qiong, et al. Travel-time variations before and after the Yangbi Ms6.4 earthquake in 2021 derived from active-source seismic data[J]. China Earthquake Engineering Journal, 2021(04):767-776. (in Chinese)
[8] 宋晋东, 白琳娟, 李山有, 冯继威, 马强. 2017年3月27日云南漾濞5.1级地震强震动记录特征初步分析[J].地震工程与工程振动, 2017, 37(02):197-204.SONG Jindong, BAI Linjuan, LI Shanyou, et al. The characteristics of strong motion records of Yunnan Yangbi Ms5.1 earthquake on March 27, 2017[J]. Earthquake Engineering and Engineering Dynamics, 2017, 37(2):197-204. (in Chinese)
[9] 潘章容, 李同林, 崔建文, 等.2021年5月21日云南漾濞M6.4及相关地震强震动记录特征分析[J].地震工程学报, 2021(04):791-798.PAN Zhangrong, LI Tonglin, CUI Jianwen, et al. Characteristics of strong motion records of the Yangbi M 6.4 and related earthquakes on May 21, 2021[J]. China Earthquake Engineering Journal, 2021(04):791-798. (in Chinese)
[10] 徐培彬, 温瑞智.基于我国强震动数据的地震动持时预测方程[J].地震学报, 2018, 40(06):809-819, 832.XU Peibin, WEN Ruizhi. The prediction equations for the significant duration of strong motion in Chinese mainland[J]. Acta Seismologica Sinica, 2018, 40(06):809-819, 832. (in Chinese)
[11] JOLLIFFE I T. Principal component analysis[J]. Springer, New York.2002.
[12] ALIMORADI A. Earthquake Ground Motion Simulation using Novel Machine Learning Tools[J]. earthquake engineering research laboratory, 2011.
[13] DEB K, PRATAP A, AGARWAL S, MEYARIVANT. A fast and elitist multiobjective genetic algorithm:NSGA-Ⅱ[J]. IEEE Transactions on Evolutionary Computation, 2002, 6(2):182-197.

相似文献/References:

[1]张翠然,陈厚群.工程地震动模拟研究综述[J].世界地震工程,2008,(2):150.
 ZHANG Cuiran,CHEN Houqun.Review and prospects on the simulation research of engineering earthquake ground motion[J].,2008,(04):150.
[2]廖旭,赵伯明,黄河,等.沈阳地区地下速度结构模型建立及优化[J].世界地震工程,2008,(4):064.
 LIAO Xu,ZHAO Boming,HUANG He,et al.Establishment and optimization of underground velocity structure model in Shenyang area[J].,2008,(04):064.
[3]戴必辉,陶忠,徐国林,等.云南漾濞MS6.4级地震震中区农房震害调查[J].世界地震工程,2021,(03):009.
 DAI Bihui,TAO Zhong,XU Guolin,et al.Investigation of seismic damage to rural houses in the epicenter area of Yangbi MS 6.4 earthquake in Yunnan[J].,2021,(04):009.
[4]黄勇,宋光有,李国昌,等.玛多地震中大野马岭大桥震害机理分析[J].世界地震工程,2021,(04):012.
 HUANG Yong,SONG Guangyou,LI Guochang,et al.Analysis of damage mechanism of Dayemaling bridge during the Maduo earthquake[J].,2021,(04):012.

备注/Memo

备注/Memo:
收稿日期:2021-08-05;改回日期:2021-09-06。
基金项目:国家自然科学基金重点项目(U1939210;52078470);国家重点研发计划(2017YFC1500403)
作者简介:靳超越(1996-),男,硕士生,主要从事地震工程研究.E-mail:1697092641@qq.com
通讯作者:胡进军(1978-),男,研究员,博士,主要从事地震动模型和强度指标研究.E-mail:hujinjun@iem.ac.cn
更新日期/Last Update: 1900-01-01