A convolutional neural network-based model for identifying earthquakes and blasting：Preliminary application in Guangdong region
-
摘要:
本文基于AlexNet卷积神经网络模型，提出了一种数据处理简单、准确率高的人工爆破波形识别方法。利用广东省地震台网记录，选取人工分析入库的天然地震和人工爆破事件数据源对模型进行训练和测试，搭建了一个适用于广东地区的爆破自动识别器，并对广东地区540个波形进行测试。结果显示，运用该模型所得天然地震事件的精确率、召回率以及F1分数均大于0.98，而人工爆破事件的识别精确率、召回率以及F1分数均大于0.90。表明该模型可以高效准确地判别广东地区天然地震与人工爆破波形，比人工识别方法更稳定、准确和高效。
Abstract:This paper proposes an efficient method for identifying artificial blasting waveforms based on the AlexNet convolutional neural network, which is designed earlier and still widely used today. This method directly uses event records as input data, which can simplify the data preprocessing, shorten the event determination time, and achieve a simple and fast effect. From the records of the Guangdong seismic network, this paper selects 312 artificial blasting events with ML＞1.8 and 526 natural earthquake events with ML＞1.4 that were manually analyzed and entered into the database. To achieve the best identification results, waveform within 30 km of the epicenter were selected, and after waveform preprocessing, such as trimming waveforms to a uniform length, waveform normalization, and removing abnormal information like calibration, square waves, sudden jumps, interference, and instrument faults, a total of 1840 valid waveforms were obtained. Among them,
1000 natural earthquake waveforms and 300 blasting waveforms were used to train the model, building an automatic blasting events classifier suitable for the Guangdong region. Additionally, this paper also studied the effect of changing the training set and validation set ratio during training and the training sizes on the accuracy. The results show that in this model, the accuracy reaches its optimum when the training set to validation set ratio is 8∶2, and when the number of training samples exceeds 600, the accuracy is higher than 95%. Finally, the well trained classifier was then tested by 540 waveforms from the Guangdong region, and it correctly identified 526 waveforms in less than 2 seconds, with an accuracy rate of 97.41%. Its precision, recall rate, and F1-score for natural earthquake events are all greater than 0.98, while its precision, recall rate, and F1-score of artificial blasting events are all greater than 0.90. All of these indicate that, on one hand, the size of the training sample needed for the model to achieve high accuracy is small, showing that this method is quite efficient. On the other hand, the AlexNet convolutional neural network model demonstrates higher adaptability to natural earthquake events with more training samples. With the input of more blasting events in the future, the model's recognition rate of artificial blastings will be further improved.In conclusion, the AlexNet convolutional neural network can efficiently and accurately distinguish between natural earthquakes and artificial blasting in the Guangdong region. It can meet the requirement of timely and accurately removing artificial blasting events from the natural earthquake catalog to ensure the completeness and accuracy of the catalog, which is beneficial for regional strong earthquake prediction and seismic hazard assessment. Compared to manual work, this approach is more stable, accurate, and efficient. So the blasting classifier based on this model will save a significant amount of time and manpower for the work of Guangdong seismic network and provide support for the results of post-earthquake emergency response. Future research will focus on the practical application of this classifer, based on the data recorded by the Guangdong seismic network, continuously improving the accuracy and robustness of the classifer through constant testing, and applying it to practical earthquake early warning and daily seismic monitoring.
-
-
![]()
图 4 基于AlexNet卷积神经网络模型搭建的人工爆破识别器流程图
Figure 4. AlexNet-based blasting classifier flow chart
![]()
图 1 所选事件空间分布图及广东省内台站分布图
Figure 1. Spatial distribution of events uesd in this paper and seismic stations in Guangdong
![]()
图 2 天然地震（a）和人工爆破（b）的有效波形示例
Figure 2. Examples of effective waveforms for earthquake （a） and blasting （b）
![]()
图 3 本文采用的卷积神经网络结构
Figure 3. The architecture of the proposed convolutional neural network
![]()
图 5 训练过程中训练集与验证集比例（a）及天然地震训练数量（b）与验证集准确率的对应关系
Figure 5. Validation accuracies verus ratios of training and validation （a） and validation accuracies verus training sizes of natural earthquakes （b）
![]()
图 6 AlexNet卷积神经网络模型训练结果
（a） 准确率；（b） 代价函数损失
Figure 6. Training performance of convolutional neural network of AlexNet
（a） Accuracy；（b） Cost function loss
表 1 测试得到的地震与爆破的精确率、召回率和F1分数
Table 1 Precision，recall and F1-score for two classification categories
事件类型 精确率 召回率 F1分数 天然地震 0.989 0.980 0.984 人工爆破 0.908 0.947 0.927 
下载: 导出CSV
表 2 测试识别结果的混淆矩阵
Table 2 Confusion matrix of waveform classification
天然地震预测值 人工爆破预测值 天然地震真实值 437 9 人工爆破真实值 5 89
下载: 导出CSV
-
陈润航,黄汉明,柴慧敏. 2018. 地震和爆破事件源波形信号的卷积神经网络分类研究[J]. 地球物理学进展,33(4):1331–1338. doi: 10.6038/pg2018BB0326 Chen R H,Huang H M,Chai H M. 2018. Study on the discrimination of seismic waveform signals between earthquake and explosion events by convolutional neural network[J]. Progress in Geophysics,33(4):1331–1338 (in Chinese).
黄汉明,边银菊,卢世军,蒋正锋,李锐. 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. doi: 10.3969/j.issn.0253-3782.2010.03.002 Huang H M,Bian Y J,Lu S J,Jiang Z F,Li R. 2010. A wavelet feature research on seismic waveforms of earthquakes and explosions[J]. Acta Seismologica Sinica,32(3):270–276 (in Chinese).
隗永刚,杨千里,王婷婷,蒋长胜,边银菊. 2019. 基于深度学习残差网络模型的地震和爆破识别[J]. 地震学报,41(5):646–657. doi: 10.11939/jass.20190030 Wei Y G,Yang Q L,Wang T T,Jiang C S,Bian Y J. 2019. Earthquake and explosion identification based on deep learning residual network model[J]. Acta Seismologica Sinica,41(5):646–657 (in Chinese).
赵明,陈石,Yuen D. 2019. 基于深度学习卷积神经网络的地震波形自动分类与识别[J]. 地球物理学报,62(1):374–382. doi: 10.6038/cjg2019M0151 Zhao M,Chen S,Yuen D. 2019. Waveform classification and seismic recognition by convolution neural network[J]. Chinese Journal of Geophysics,62(1):374–382 (in Chinese).
赵永,刘卫红,高艳玲. 1995. 北京地区地震、爆破和矿震的记录图识别[J]. 地震地磁观测与研究,16(4):48–54. Zhao Y,Liu W H,Gao Y L. 1995. Distinguishing earthquake,explosion and mine earthquake in Beijing area[J]. Seismological and Geomagnetic Observation and Research,16(4):48–54 (in Chinese).
郑周,林彬华,金星,韦永祥,丁炳火,陈辉. 2023. 基于卷积神经网络的地震波形智能识别研究[J]. 世界地震工程,39(2):148–157. Zheng Z,Lin B H,Jin X,Wei Y X,Ding B H,Chen H. 2023. Intelligent recognition of seismic waveform based on convolutional neural network[J]. World Earthquake Engineering,39(2):148–157 (in Chinese).
周少辉,蒋海昆,李健,曲均浩,郑晨晨,李亚军,张志慧,郭宗斌. 2021. 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,43(3):663–676. doi: 10.3969/j.issn.0253-4967.2021.03.012 Zhou S H,Jiang H K,Li J,Qu J H,Zheng C C,Li Y J,Zhang Z H,Guo Z B. 2021. Research on identification of seismic events based on deep learning:Taking the records of Shandong seismic network as an example[J]. Seismology and Geology,43(3):663–676 (in Chinese).
Bengio Y,Courville A,Vincent P. 2013. Representation learning:A review and new perspectives[J]. IEEE Trans Pattern Anal Mach Intell,35(8):1798–1828. doi: 10.1109/TPAMI.2013.50
Chen Y K,Zhang G Y,Bai M,Zu S H,Guan Z,Zhang M. 2019. Automatic waveform classification and arrival picking based on convolutional neural network[J]. Earth Space Sci,6(7):1244–1261. doi: 10.1029/2018EA000466
Gao H,Zhang J. 2019. Simultaneous denoising and interpolation of seismic data via the deep learning method[J]. Earthquake Research Advances,33(1):37–51.
Glorot X,Bengio Y. 2010. Understanding the difficulty of training deep feedforward neural networks[C]//Proceedings of the 13th International Conference on Artificial Intelligence and Statistics. Sardinia:JMLR:249−256.
Han J,Moraga C. 1995. The influence of the sigmoid function parameters on the speed of backpropagation learning[C]//International Workshop on Artificial Neural Networks. Malaga-Torremolinos:Springer:195−201.
Hinton G E,Osindero S,Teh Y W. 2006. A fast learning algorithm for deep belief nets[J]. Neural Comput,18(7):1527–1554. doi: 10.1162/neco.2006.18.7.1527
Krizhevsky A,Sutskever I,Hinton G. 2012. ImageNet classification with deep convolutional neural networks[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems. Lake Tahoe Nevada:Curran Associates Inc:1097−1105.
Kuang W H,Yuan C C,Zhang J. 2021. Real-time determination of earthquake focal mechanism via deep learning[J]. Nat Commun,12(1):1432. doi: 10.1038/s41467-021-21670-x
LeCun Y,Bengio Y,Hinton G. 2015. Deep learning[J]. Nature,521(7553):436–444. doi: 10.1038/nature14539
Li Z F,Meier M A,Hauksson E,Zhan Z W,Andrews J. 2018. Machine learning seismic wave discrimination:Application to earthquake early warning[J]. Geophys Res Lett,45(10):4773–4779. doi: 10.1029/2018GL077870
Linville L,Pankow K,Draelos T. 2019. Deep learning models augment analyst decisions for event discrimination[J]. Geophys Res Lett,46(7):3643–3651. doi: 10.1029/2018GL081119
Men-Andrin M,Ross Z E,Anshul R,Ashwin B,Suraj N,Peter K,Li Z F,Jennifer A,Egill H,Yue Y S. 2019. Reliable real-time seismic signal/noise discrimination with machine learning[J]. J Geophys Res:Solid Earth,124(1):788–800. doi: 10.1029/2018JB016661
Nair V,Hinton G E. 2010. Rectified linear units improve restricted Boltzmann machines[C]//Proceedings of the 27th International Conference on Machine Learning. Haifa,Israel:Omnipress:807−814.
Zhang H,Ma C C,Pazzi V,Li T B,Casagli N. 2020a. Deep convolutional neural network for microseismic signal detection and classification[J]. Pure Appl Geophys,177(12):5781–5797. doi: 10.1007/s00024-020-02617-7
Zhang G Y,Lin C Y,Chen Y K. 2020b. Convolutional neural networks for microseismic waveform classification and arrival picking[J]. Geophysics,85(4):WA227–WA240. doi: 10.1190/geo2019-0267.1
Zhang Q S,Zhang W,Wu X M,Zhang J,Kuang W H,Si X. 2022. Deep learning for efficient microseismic location using source migration-based imaging[J]. J Geophys Res:Solid Earth,127(3):e2021JB022649. doi: 10.1029/2021JB022649
Zhou Y T,Chellappa R. 1988. Computation of optical flow using a neural network[C]//IEEE 1988 International Conference on Neural Networks. San Diego,CA:IEEE: 2 :71−78.
-
期刊类型引用(5)
1. 奉建州，廖力，李平恩，吴庆举. 2001年昆仑山口西M_S8.1地震自发破裂过程数值模拟研究. 地球物理学报. 2024(06): 2141-2158 . 
百度学术
2. 杜书恒，赵晔，庞姗，师永民. 岩石水力压裂微观破裂机制. 天然气地球科学. 2016(12): 2237-2245 . 百度学术
3. 陈培善. 全球大震和中国及邻区中强震地震活动(2013年3—4月). 地震学报. 2013(04): 616-618 . 本站查看
4. 张丽芬，姚运生. 震源动力学破裂过程数值模拟研究. 地震学报. 2013(04): 604-615+2 . 本站查看
5. 徐恒垒，靳平，倪四道，刘文学，王红春，李欣. 谱元法数值模拟地表起伏对补偿线性矢量偶极源Rg波低谷点的影响. 地球物理学报. 2011(11): 2831-2837 . 百度学术
其他类型引用(3)
计量
- 文章访问数: 120
- HTML全文浏览量: 28
- PDF下载量: 45
- 被引次数: 8
