使用模拟退火算法,以2019年4月6日江苏溧水ML 3.3地震为典型震例,进行地震定位反演试算,确定适用于江苏及邻区地震定位的最佳初始温度、降温策略和扰动函数。设置初始温度为25,降温策略为T = 0.96T,选用效率较高的局部收敛加强型方法产生随机模型,对2019年江苏及周边地区105次ML≥1.8地震进行重新定位,结果表明,76%的事件震中经、纬度定位误差绝对值在0.1°以内,但深度定位误差较大。鉴于该算法对P波到时信息及地震台站空间分布的依赖性,认为模拟退火算法在江苏及邻区陆地地震定位中效果较好,对于海域地震,则定位误差较大。
We took the Lishui ML 3.3 earthquake on April 6, 2019, in Jiangsu Province, as a typical case to determine the optimal initial temperature, cooling schedule, and perturbation function of the Simulated Annealing Algorithm for seismic location in Jiangsu and its adjacent areas. Then, we set the initial temperature to 25, the cooling schedule to T = 0.96T, and chose a local convergence method which is more efficient in generating random models to relocate 105 earthquakes with ML≥1.8 occurred in Jiangsu province and surrounding areas in 2019. The results show that the absolute longitude and latitude location errors of 76% events are within 0.1°, but the depth location errors are large. It is considered that the Simulated Annealing Algorithm is more suitable for the continental seismic location, and the errors are significant for earthquakes in the sea because of the dependence on P-wave arrival time and station spatial distribution.
2021,42(4): 51-56 收稿日期:2020-02-05
DOI:10.3969/j.issn.1003-3246.2021.04.007
基金项目:中国地震局监测、预报、科研三结合课题(项目编号:3JH-202001028);江苏省地震局青年科学基金(项目编号:201909)
作者简介:于悦颖(1993-),女,工程师,主要从事地震监测、数字地震资料分析工作。E-mail:yy_seism@163.com
参考文献:
窦立婷,成诚,李云,等. 利用模拟退火法反演大陆地震位置[J]. 地震地磁观测与研究,2018,39(1):31-35.
师学明,王家映. 地球物理资料非线性反演方法讲座(三)模拟退火法[J]. 工程地球物理学报,2007,4(3):165-174.
田玥,陈晓非. 地震定位研究综述[J]. 地球物理学进展,2002,17(1):147-155.
杨波,张炳,隆爱军,等. 模拟退火算法在地震定位中的应用[J]. 华北地震科学,2019,37(4):73-77.
尹奇峰,潘冬明,郭全仕,等. 基于快速模拟退火算法的井中微地震事件定位反演[J]. 地球物理学进展,2019,34(5):
1 954-1 961.
Billings S D. Simulated annealing for earthquake location[J]. Geophysical Journal International, 1994, 118(3): 680-692.
Gao X, Wang W M, Yao Z X. Hypocentral determination using the simulated annealing method[J]. Chinese Journal of Geophysics, 2002, 45(4): 546-553.
Geiger L. Probability method for the determination of earthquake epicenters from the arrival time only[J]. Bull St Louis Univ, 1912, 8: 60-71.
Ingber L. Very fast simulated re-annealing[J]. Mathematical and Computer Modelling, 1989, 12(8): 967-973.
Pishro-Nik H. Introduction to probability, statistics, and random processes[M]. Kappa Research LLC, 2014: 703-723.
Sen M K, Stoffa P L. Global optimization methods in geophysical inversion[M]. 2nd ed. Cambridge: Cambridge University Press, 2013: 81-117.
Van Laarhoven P J M, Aarts E H L. Simulated annealing: theory and applications[M]. Dordrecht: Springer, 1987: 99-135.