基于深度学习的地震事件分类识别——以山东地震台网记录为例

doi:10.3969/j.issn.0253-4967.2021.03.012

地震地质 ›› 2021, Vol. 43 ›› Issue (3): 663-676.DOI: 10.3969/j.issn.0253-4967.2021.03.012

基于深度学习的地震事件分类识别——以山东地震台网记录为例

周少辉¹⁾, 蒋海昆^2),*, 李健³⁾, 曲均浩¹⁾, 郑晨晨¹⁾, 李亚军¹⁾, 张志慧¹⁾, 郭宗斌¹⁾

1)山东省地震局, 济南 250014;
2)中国地震台网中心, 北京 100045;
3)海南省地震局, 海口 570203

收稿日期:2020-04-03 修回日期:2020-06-20 出版日期:2021-06-20 发布日期:2021-07-20
通讯作者: *蒋海昆, 男, 1964年生, 博士, 研究员, 主要从事余震序列及相关研究, E-mail: jianghaikun@sohu.com。
作者简介:周少辉, 男, 1991年生, 工程师, 2017年于中国地震局地震预测研究所获地球物理学硕士学位, 主要从事数字地震学及人工智能的地震学应用研究, E-mail: 674891062@qq.com。
基金资助:
山东省自然科学基金重点项目(ZR2020KF003)、中国地震局监测、预报、科研三结合项目(3JH-201901064)、山东省地震局一般科研项目(YB-2003, CGZH2001)和中国地震局地震科技星火计划项目(XH19027)共同资助

RESEARCH ON IDENTIFICATION OF SEISMIC EVENTS BASED ON DEEP LEARNING: TAKING THE RECORDS OF SHANDONG SEISMIC NETWORK AS AN EXAMPLE

ZHOU Shao-hui¹⁾, JIANG Hai-kun²⁾, LI Jian³⁾, QU Jun-hao¹⁾, ZHENG Chen-chen¹⁾, LI Ya-jun¹⁾, ZHANG Zhi-hui¹⁾, GUO Zong-bin¹⁾

1)Shandong Earthquake Agency, Jinan 250014, China;
2)China Earthquake Networks Center, Beijing 100045, China;
3)Hainan Earthquake Agency, Haikou 570203, China

Received:2020-04-03 Revised:2020-06-20 Online:2021-06-20 Published:2021-07-20

摘要/Abstract

摘要： 为实现天然地震与爆破、塌陷事件类型的快速高效识别, 文中应用深度学习技术中的卷积神经网络模型, 设计了基于单个事件单个台站波形记录的深度学习训练模块和基于单个事件多个台站波形记录的实时测试模块。以每个事件P波到时最早的5个台站记录到的原始三分向波形为输入, 分别采用目前主流的AlexNet、 VGG16、 VGG19、 GoogLeNet 4种卷积神经网络结构进行学习训练, 结果显示各类卷积神经网络结构对训练集与测试集的识别准确率均达93%以上, 且各个网络在训练过程中的训练集与测试集的准确率及代价函数的走势曲线基本一致。其中, AlexNet网络结构的识别准确率最高, 测试集为98.51%, 且未发生过拟合现象; VGG16、 VGG19网络结构的准确率次之; GoogLeNet网络结构的识别准确率相对较低。为检验深度学习卷积神经网络在数字地震台网实时运行过程中的事件判别效能, 选取训练好的AlexNet卷积神经网络开展基于单个事件多个台站波形记录的事件类型判定检验。最终结果显示, 在山东台网实时触发的110个M≥0.7事件中, 共有89个事件的类型被准确识别, 准确率约为80.9%。具体到各个类型事件中, 天然地震的准确率约为74.6%; 爆破的准确率约为90.9%; 塌陷事件的准确率为100%。若删除其中由于波形失真而造成的类型识别错误事件, 则天然地震的识别准确率将提高至91.4%, 而所有事件的整体识别准确率也将由80.9%提高至91.7%, 与目前地震台网日常工作中人工判定的识别准确率基本相当。这表明, 深度学习技术可以快速高效地实现天然地震与爆破、塌陷的事件类型识别。

关键词: 深度学习, 非天然地震, 自动识别

Abstract: In order to realize the rapid and efficient identification of earthquakes, blasting and collapse events, this paper applies the Convolutional Neural Network(CNN)in deep learning technology to design a deep learning training module based on single station waveform recording of single event and a real-time test module based on multiple stations waveform recording of single event.
On the basis of ensuring that the data is comprehensive, objective and original, the three-component waveforms of the first five stations that recorded the P-wave arrival time of each event are input, and the current mainstream convolutional neural network structures are used for learning test. The four main convolutional neural network structures of AlexNet, VGG16, VGG19 and GoogLeNet are used for learning training, and the learning effects of different network structures are compared and analyzed. The results show that in the training process of various convolutional neural network structures, the accuracy rate and the cost function curve of the training set and the test set of each network are basically the same. The accuracy rate increases gradually with the increase of the training times and exceeds 90%, and finally stabilizes around a certain value. The cost function curve decreases rapidly with the increase of the training times, and eventually the stability does not change near a relatively small value. At the same time, over-fitting occurred in all convolutional neural network structures during training, except for AlexNet. In the end, the cost function of each type of structural training set and test set is finally lower than 0.194, and the recognition accuracy of each type of structure for training sets and test sets is over 93%. Among them, the recognition accuracy of AlexNet network structure is the highest, the accuracy of the training set of AlexNet network structure is as high as 100%, the test set is 98.51%, and no overfitting occurred; the accuracy of VGG16 and VGG19 network structure comes second, and the recognition accuracy of GoogLeNet network structure is relatively low, and the trend curves of the accuracy and cost function in training and test set of each network in the training process are basically the same. Subsequently, in order to test the event discrimination efficiency of the CNN in deep learning in the real-time operation of the digital seismic network, we select the trained AlexNet convolutional neural network to perform event type determination test based on the waveform recording of multiple stations of a single event. The final result shows that the types of a total of 89 events are accurately identified in the 110 events with M ≥0.7 recorded by Shandong seismic network, and the accuracy rate is about 80.9%. Among them, the accuracy rate of natural earthquake is about 74.6%, that of explosion is about 90.9%, and that of collapse is 100%. The recognition accuracy of collapse and explosion events is relatively high, and it basically reaches or exceeds the recognition accuracy of manual determination in the daily work of the seismic network. The accuracy of natural earthquake identification is relatively low. Among the 18 misidentified natural earthquakes, up to 13 events were judged as blasting or difficult to identify due to distortion of waveforms recorded by some stations(They are determined to be explosion and earthquake each by the records of two of the five stations). If sloughing off the recognition type error events caused by waveform distortion due to the background noise interference that overwhelms the real event waveform or waveform drift, the recognition accuracy of earthquake will become 91.4%, and the recognition accuracy of all events will increase from 80.9%to 91.7%, which is basically equivalent to the recognition accuracy of manual judgment in the daily work of the seismic network. This indicates that deep learning can quickly and efficiently realize the type identification of earthquake, blasting and collapse events.

Key words: deep learning, non-natural earthquake, automatic identification

中图分类号:

P315.61

周少辉, 蒋海昆, 李健, 曲均浩, 郑晨晨, 李亚军, 张志慧, 郭宗斌. 基于深度学习的地震事件分类识别——以山东地震台网记录为例[J]. 地震地质, 2021, 43(3): 663-676.

ZHOU Shao-hui, JIANG Hai-kun, LI Jian, QU Jun-hao, ZHENG Chen-chen, LI Ya-jun, ZHANG Zhi-hui, GUO Zong-bin. RESEARCH ON IDENTIFICATION OF SEISMIC EVENTS BASED ON DEEP LEARNING: TAKING THE RECORDS OF SHANDONG SEISMIC NETWORK AS AN EXAMPLE[J]. SEISMOLOGY AND GEOLOGY, 2021, 43(3): 663-676.

参考文献

[1] 边银菊, 邹立晔. 2002. 学习向量量化在地震和爆破识别中的应用[J]. 地震地磁观测与研究, 23(1): 10—15.
BIAN Yin-ju, ZOU Li-ye. 2002. Application of Learning Vector Quantization(LVQ)to discriminating earthquakes and explosions[J]. Seismological and Geomagnetic Observation and Research, 23(1): 10—15(in Chinese).
[2] 陈超, 齐峰. 2019. 卷积神经网络的发展及其在计算机视觉领域中的应用综述[J]. 计算机科学, 46(3): 63—73.
CHEN Chao, QI Feng. 2019. Review on development of convolutional neural network and its application in computer vision[J]. Computer Science, 46(3): 63—73(in Chinese).
[3] 陈润航, 黄汉明, 施佳朋, 等. 2019. 天然地震与人工爆破地震波形的实时分类研究[J]. 地球物理学进展, 34(1): 1—13.
CHEN Run-hang, HUANG Han-ming, SHI Jia-peng, et al. 2019. Study on real-time identification of natural earthquake and artificial blasting seismic waveforms[J]. Progress in Geophysics, 34(1): 1—13(in Chinese).
[4] 付超, 林年添, 张栋, 等. 2018. 多波地震深度学习的油气储层分布预测案例[J]. 地球物理学报, 61(1): 293—303.
FU Chao, LIN Nian-tian, ZHANG Dong, et al. 2018. Prediction of reservoirs using multi-component seismic data and the deep learning method[J]. Chinese Journal of Geophysics, 61(1): 293—303(in Chinese).
[5] 黄汉明, 边银菊, 卢世军, 等. 2010. 天然地震与人工爆破的波形小波特征研究[J]. 地震学报, 32(3): 270—276.
HUANG Han-ming, BIAN Yin-ju, LU Shi-jun, et al. 2010. A wavelet feature research on seismic waveforms of earthquakes and explosions[J]. Acta Seismologica Sinica, 32(3): 270—276(in Chinese).
[6] 贾佳, 王夫运, 吴庆举. 2019. 机器学习在地震检测与震相识别的应用综述[J]. 地震工程学报, 41(6): 1419—1425.
JIA Jia, WANG Fu-yun, WU Qing-ju. 2019. Review of the application of machine learning in seismic detection and phase identification[J]. China Earthquake Engineering Journal, 41(6): 1419—1425(in Chinese).
[7] 蒋一然, 宁杰远. 2019. 基于支持向量机的地震体波震相自动识别及到时自动拾取[J]. 地球物理学报, 62(1): 361—373.
JIANG Yi-ran, NING Jie-yuan. 2019. Automatic detection of seismic body-wave phases and determination of their arrival times based on support vector machine[J]. Chinese Journal of Geophysics, 62(1): 361—373(in Chinese).
[8] 孙瑜阳. 2018. 深度学习及其在图像分类识别中的研究综述[J]. 信息技术与信息化, 23(1): 138—140.
SUN Yu-yang. 2018. A review of deep learning and its research in image classification and recognition[J]. Information Technology & Informatization, 23(1): 138—140(in Chinese).
[9] 孙志军, 薛磊, 许阳明, 等. 2012. 深度学习研究综述[J]. 计算机应用研究, 29(8): 2806—2810.
SUN Zhi-jun, XUE Lei, XU Yang-ming, et al. 2012. Overview of deep learning[J]. Application Research of Computers, 29(8): 2806—2810(in Chinese).
[10] 田野. 2015. 不同震源类型的可视化识别研究 [D]. 桂林: 广西师范大学.
TIAN Ye. 2015. Research on visual identifying distinct seismic sources [D]. Guangxi Normal University, Guilin(in Chinese).
[11] 王军. 2018. 支持向量机在地震与爆破识别中的应用[J]. 地震地磁观测研究, 39(3): 181—188.
WANG Jun. 2018. Application of SVM to discriminating earthquakes and explosions[J]. Seismological and Geomagnetic Observation and Research, 39(3): 181—188(in Chinese).
[12] 张博. 2013. 爆炸和地震的识别研究 [D]. 北京: 中国地震局地球物理研究所.
ZHANG Bo. 2013. Research on earthquake and explosions identification [D]. Institute of Geophysics, China Earthquake Administration, Beijing(in Chinese).
[13] 赵明, 陈石, 房立华, 等. 2019. 基于U形卷积神经网络的震相识别与到时拾取方法研究[J]. 地球物理学报, 62(8): 3034—3042.
ZHAO Ming, CHEN Shi, FANG Li-hua, et al. 2019. Earthquake phase arrival auto-picking based on U-shaped convolutional neural network[J]. Chinese Journal of Geophysics, 62(8): 3034—3042(in Chinese).
[14] Bengio Y, Courville A, Vincent P. 2013. Representation learning: A review and new perspectives[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(8): 1798—1828.
[15] Fah D, Koch K. 2002. Discrimination between earthquakes and chemical explosions by multivariate statistical analysis: A case study for Switzerland[J]. Bulletin of the Seismological Society of America, 92(5): 1795—1805.
[16] Krizhevsky A, Sutskever I, Hinton G. 2012. ImageNet classification with deep convolutional neural networks[J]. Advances in Neural Information Processing Systems, 25(2): 1097—1105.
[17] LeCun Y, Bottou L, Bengio Y, et al. 1998. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 86(11): 2278—2324.
[18] LeCun Y, Boser B, Denker J S, et al. 2014. Backpropagation applied to handwritten zip code recognition[J]. Neural Computation, 1(4): 541—551.
[19] Moyano L G. 2017. Learning network representations[J]. European Physical Journal Special Topics, 226(3): 499—518.
[20] Perol T, Gharbi M, Denolle M. 2018. Convolutional neural network for earthquake detection and location[J]. Science Advances, 4(2): e1700578.
[21] Ross Z E, Meier M A, Hauksson E. 2018. P-wave arrival picking and first-motion polarity determination with deep learning[J]. Journal of Geophysical Research: Solid Earth, 123(6): 5120—5129.
[22] Simonyan K, Zisserman A. 2014. Very deep convolutional networks for large-scale image recognition[J]. Computer Science, 9(4): 1542—1556.
[23] Szegedy C, Liu W, Jia Y, et al. 2015. Going deeper with convolutions[J]. Access IEEE, 2(1): 1—9.

基于深度学习的地震事件分类识别——以山东地震台网记录为例

RESEARCH ON IDENTIFICATION OF SEISMIC EVENTS BASED ON DEEP LEARNING: TAKING THE RECORDS OF SHANDONG SEISMIC NETWORK AS AN EXAMPLE

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价