在地脉动数据上应用分类算法的地震预测实验

doi:10.12196/j.issn.1000-3274.2020.01.013

摘要/Abstract

摘要： 以“红肿”假说为基础, 在由地脉动数据统计量和过往震例构成的样本集上应用数据挖掘中的分类算法开展地震预测实验。筛选符合震级、震中距、发震时间间隔以及未受台风影响等要求的地震对, 并以其尾地震作为预测对象。计算地震对时间范围内各时间窗中地脉动数据的标准差, 并采用z-score标准化方法对标准差数据进行标准化处理。然后, 选取距震中最近三个台的最后一组标准化数据的中位数作为正样本数据, 选取各台站平静期数据的中位数作为负样本数据, 最后将上述正负样本数据构成样本集。使用CART算法、 GBDT算法和SVM算法在此样本集上分别构建预测模型, 采用5折交叉验证方法对预测模型进行评估。实验结果表明: ① 地震与地脉动变化存在一定的关系, 且地脉动异常现象更多地出现在6.0级以上地震发生前; ② 6.0级以上地震构成的正样本对预测模型的构建影响较大; ③ SVM算法更适用于小样本数据环境。

关键词: 地脉动, 分类, 地震预测

Abstract: Based on the “redness and swelling” hypothesis, the classification algorithm of data mining is applied to carry out earthquake prediction experiments on the sample set which is composed of microtremor data statistics and past earthquake cases. We chose seismic pairs that meet the requirements of magnitude, epicenter distance, occurrence time interval, and unaffected by typhoons conditions, and use the tail earthquakes as prediction targets. The standard deviation of microtremor data in time windows is calculated, and the standard deviation data is standardized by Z-score standardization method. Then, the median of the last group of standardized data from the nearest three stations to the epicenter is selected as the positive sample data, and the median of the seismic quiet period data of each station is selected as the negative sample data. Finally, the positive and negative sample data are constructed into the sample set. CART, GBDT and SVM methods are used to construct the prediction model on this sample set, and 5 fold cross validation method is used to evaluate the prediction model. The results show that: ① there is a relationship between earthquakes and microtremor changes, and there are microtremor anomalies before earthquakes (M≥6.0) occurred. ② The positive samples that magnitude over 6 have a greater influence on the model construct. ③ The SVM algorithm is more suitable for the small sample data environment of this paper.

Key words: Microtremor, Classification, Earthquake forecast

中图分类号:

P315.7

马士振, 刘宏志, 牟磊育. 在地脉动数据上应用分类算法的地震预测实验[J]. 地震, 2020, 40(1): 159-171.

MA Shi-zhen, LIU Hong-zhi, MU Lei-yu. Earthquake Forecast Experiment with Classification Algorithm Based on Microtremor Data[J]. EARTHQUAKE, 2020, 40(1): 159-171.

参考文献

[1] 强祖基.怀念中国地震泰斗重温地震前兆研究先驱傅承义院士的学术成果[C]. 北京: 天灾预测学术研讨会, 2009.
QIANG Zu-ji. Recalling academician Fu Chengyi’s academic achievements in the study of earthquake precursors[C]. Beijing: Symposium on Natural Disaster Prediction, 2009 (in Chinese).
[2] 高金哲, 吕郑, 张洪艳, 等. 地震前驱波观测与研究进展[J]. 华南地震, 2005, 25(1): 53-58.
GAO Jin-zhe, LV Zheng, ZHANG Hong-yan, et al. Progress in the observation and research on earthquake precursor wave[J]. South China Journal of Seismology, 2005, 25(1): 53-58 (in Chinese).
[3] 陈德福. 潮汐形变前驱波的时空特征[J]. 大地测量与地球动力学, 2006, 26(2): 24-30.
CHEN De-fu. Space-time characteristics of tidal deformation precursor[J]. Journal of Geodesy and Geodynamics, 2006, 26(2): 24-30 (in Chinese).
[4] 张淑亮, 李丽, 李惠玲. 多种观测仪器记录地震前驱波的证据[C]. 国际地震动态, 2017, (8): 105-106.
ZHANG Shu-liang, LI Li, LI Hui-ling. The evidence for earthquake precursor waves recorded by various observing instruments[C]. Recent Developments in World Seismology, 2017, (8): 105-106 (in Chinese).
[5] 张淑亮, 李斌, 吕芳. 低频前驱波传播机理的研究现状[J]. 国际地震动态, 2012, (2): 13-19.
ZHANG Shu-liang, LI Bin, LV Fang. Review of the research on the propagation mechanism of low-frequency precursory waves[J]. Recent Developments in World Seismology, 2012, (2): 13-19 (in Chinese).
[6] 朱传镇, 房明山, 安镇文, 等. 脉动与地震关系的初步探讨[J]. 地球物理学报, 1977, 20(1): 20-32.
ZHU Chuan-zhen, Fang Ming-shan, AN Zhen-wen, et al. Preliminary study of the microseisms in relation to earthquakes[J]. Chinese Journal of Geophysics, 1977, 20(1): 20-32 (in Chinese).
[7] 牛志仁, 朱传镇. 脉动与地震关系的初步探讨(二)[J]. 地球物理学报, 1978, 21(4): 325-331.
NIU Zhi-ren, ZHU Chuan-zhen. Preliminary study of the microseisms in relation to earthquakes[J]. Chinese Journal of Geophysics, 1978, 21(4): 325-331 (in Chinese).
[8] 赵根模, 杨港生, 陈化然. 寂静的前震与地震预测[J]. 地震, 2001, 21(1): 69-77.
ZHAO Gen-mo, YANG Gang-sheng, CHEN Hua-ran. Quiescent foreshock and earthquake prediction[J]. Earthquake, 2001, 21(1): 69-77 (in Chinese).
[9] 杨立明, 王建军, 冯建刚, 等. 汶川地震前地脉动低频波动现象及其应用的初步研究[J]. 中国地震, 2009, 25(4): 356-366
YANG Li-ming, WANG Jian-jun, FENG Jian-gang, et al. Preliminary study and application on the tremors with lower frequency by seismograph before the Wenchuan earthquake[J]. Earthquake Research in China, 2009, 25(4): 356-366 (in Chinese).
[10] 董蕾. 大震前地脉动及其频谱特征研究[D]. 兰州: 中国地震局兰州地震研究所, 2010.
DONG Lei. Research on displacement and frequency spectrum characteristics of microtremors before strong earthquakes[D]. Lanzhou: Lanzhou Institute of Seismology, 2010 (in Chinese).
[11] 王梅德, 韩艳杰, 于仁宝, 等. 地脉动在大震前的异常变化研究[J]. 地震研究, 2014, 37(1): 73-78.
WANG Mei-de, HAN Yan-jie, YU Ren-bao, et al. Research on anomaly changes of microtremor before strong earthquake[J]. Journal of Seismological Research, 2014, 37(1): 73-78 (in Chinese).
[12] Tan Pang-ning, Steinbach Michael, Kumar Vipin. 数据挖掘导论[M]. 范明, 范宏建, 译. 2版. 北京: 人民邮电出版社, 2011.
Tan Pang-ning, Steinbach Michael, Kumar Vipin. Introduction to data mining[M]. FAN Ming, FAN Hong-Jian, Transl, Second Edition. Beijing: Posts & Telecom Press Co. LTD, 2011 (in Chinese).
[13] 吴绍春. 地震预报中的数据挖掘方法研究[D]. 上海: 上海大学, 2005.
WU Shao-chun. Study on novel approaches of data mining for earthquake prediction[D]. Shanghai: Shanghai University, 2005 (in Chinese).
[14] 刘悦, 刘辉, 李远, 等. 基于冲量权值的ART神经网络及其在地震预报中的应用[J]. 计算机工程与应用, 2005, (5): 229-232.
LIU Yue, LIU Hui, LI Yuan, et al. Impulse force based ART network with GA optimization and its application to earthquake prediction[J]. Computer Engineering and Applications, 2005, (5): 229-232 (in Chinese).
[15] 王炜, 刘悦, 李国正, 等. 我国大陆强震预测的支持向量机方法[J]. 地震学报, 2006, 28(1): 29-36.
WANG Wei, LIU Yue, LI Guo-zheng, et al. Support vector machine method for forecasting future strong earthquakes in Chinese mainland[J]. Acta Seismological Sinica, 2006, 28(1): 29-36 (in Chinese).
[16] Dehbozorgi L, Farokhi F. Effective feature selection for short-term earthquake prediction using Neuro-Fuzzy classifier[C]. In 2010 2nd IITA, 2010.
[17] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.
ZHOU Zhi-hua. Machine Learning[M]. Beijing: Tsinghua University Press, 2016 (in Chinese).
[18] Friedman J H. Greedy function approximation: a gradient boosting machine[J]. Annals of Statistics, 2001 , 29(5): 1189-1232.
[19] Han Jiawei, Kamber Micheline, Pei Jian. 数据挖掘概念与技术[M]. 范明, 孟小峰, 译. 3版. 北京: 机械工业出版社, 2016.
Han Jiawei, Kamber Micheline, Pei Jian. Data mining concepts and techniques[M]. FAN Ming, MENG Xiao-feng, Transl, Third Edition. Beijing: China Machine Press, 2016 (in Chinese).
[20] 郑秀芬, 欧阳飚, 张东宁, 等. “国家数字测震台网数据备份中心”技术系统建设及对汶川大地震研究的数据支撑[J]. 地球物理学报, 2009, 52(5): 1412-1417.
ZHENG Xiu-fen, OUYANG Biao, ZHANG Dong-ning, et al. Technical system construction of data backup centre for China seismograph network and the data support to researches on the Wenchuan earthquake[J]. Chinese Journal Geophysics, 2009, 52(5): 1412-1417 (in Chinese).
[21] 胡小刚, 郝晓光. 汶川大地震宽带地震仪短临异常及成因初探[J]. 地球物理学报, 2008, 51(6): 1726-1737.
HU Xiao-gang, HAO Xiao-guang. The short-term anomalies detected by broadband seismographs before the May 12 Wenchuan earthquake, Sichuan, China[J]. Chinese Journal of Geophysics, 2008, 51(6): 1726-1737 (in Chinese).
[22] 傅容珊, 万柯松, 崇加军, 等. 地震前兆还是其他因素？与“汶川大地震宽带地震仪短临异常及成因初探”作者商榷[J]. 地球物理学报, 2009, 52(2): 584-589.
FU Rong-shan, WAN Ke-song, CHONG Jia-jun, et al. Earthquake auspice or other factor?-Discuss with authors of the paper “The short-term anomalies detected by broadband seismographs before the May 12 Wenchuan earthquake, Sichuan, China”[J]. Chinese Journal of Geophysics, 2009, 52(2): 584-589 (in Chinese).
[23] 夏英杰, 倪四道, 曾祥方. 汶川地震前地脉动信号的单台法研究[J]. 地球物理学报, 2011, 54(10): 2590-2596.
XIA Ying-jie, NI Si-dao, ZENG Xiang-fang. Polarization research on seismic noise before Wenchuan earthquake[J]. Chinese Journal of Geophysics, 2011, 54(10): 2590-2596 (in Chinese).
[24] 郝晓光, 胡小刚. 强震短临前兆探索: “震前扰动”现象研究[M]. 北京: 测绘出版社, 2012.
HAO Xiao-guang, HU Xiao-gang. Study of the short-term anomalous tremors before large earthquakes[M]. Beijing: Surveying & Mapping Publishing House, 2012 (in Chinese).
[25] 冯志生, 杨从杰, 梅卫萍, 等. 新沂台地电阻率整点值的远大震前驱波变化[J]. 地震地磁观测与研究, 2003, 24(4): 9-13.
FENG Zhi-sheng, YANG Cong-jie, MEI Wei-ping, et al. The preseismic wave variation of resistivity hour value about the distant and strong earthquakes at Xinyi seismic station[J]. Seismological and Geomagnetic Observation and Research, 2003, 24(4): 9-13 (in Chinese).
[26] 许昭永, 杨润海, 胡毅力, 等. 慢地震慢前兆的机制研究[J]. 地震, 2003, 23(2): 12-20.
XU Zhao-yong, YANG Run-hai, HU Yi-li, et al. A study of mechanism for slow earthquakes and precursors[J]. Earthquake, 2003, 23(2): 12-20 (in Chinese).
[27] 杨又陵, 赵根模, 高国英.等. 2001年11月14日昆仑山口西M8.1地震前的缓慢地震事件[J]. 国际地震动态, 2003, (9): 1-4.
YANG You-ling, ZHAO Gen-mo, GAO Guo-ying, et al. The slow earthquake event occuring before west to Kunlun Mountain Pass Earthquake of M_S8.1 on November 14, 2001[J]. Recent Development in World Seismology, 2003, (9): 1-4 (in Chinese).
[28] David Freedman, Robert Pisani, Roger Purves, 等. 统计学[M]. 魏宗舒, 施锡铨, 林举干, 等, 译. 1版. 北京: 中国统计出版社, 1997.
David Freedman, Robert Pisani, Roger Purves, et al. Statistics[M]. WEI Zong-shu, SHI Xi-quan, LIN Ju-gan, et al, Transl, First Edition. Beijing: China Statistics Press, 1997 (in Chinese).