Microseismic source location using deep learning:A coal mine case study in China

作     者：Yue Song Enyuan Wang Hengze Yang Chengfei Liu Baolin Li Dong Chen Yue Song;Enyuan Wang;Hengze Yang;Chengfei Liu;Baolin Li;Dong Chen

作者机构：School of Safety EngineeringChina University of Mining and TechnologyXuzhou221116China State Key Laboratory of Coal Mine Disaster Prevention and ControlChina University of Mining and TechnologyXuzhou221116China School of Environment and Safety EngineeringNorth University of ChinaTaiyuan030051China State Key Laboratory for Geomechanics&Deep Underground EngineeringChina University of Mining TechnologyXuzhou221116China 

出 版 物：《Journal of Rock Mechanics and Geotechnical Engineering》 (岩石力学与岩土工程学报（英文版）)

年 卷 期：2024年第16卷第9期

页      面：3407-3418页

核心收录：

学科分类：070801[理学-固体地球物理学] 07[理学] 0708[理学-地球物理学] 

基　　金：financial support of the Fundamental Research Funds for the Central Universities(Grant No.2022XSCX35) the National Natural Science Foundation of China(Grant Nos.51934007 and 52104230) 

主　　题：Microseismic source location Rockburst Deep learning Intelligent early warning 

摘      要：Microseismic source location is crucial for the early warning of rockburst ***,the conventional methods face challenges in terms of the microseismic wave velocity and arrival time *** techniques,such as the full convolutional neural network(FCNN),can capture spatial information but struggle with complex microseismic *** the FCNN with the long shortterm memory(LSTM)network enables better time-series signal classification by integrating multiscale information and is therefore suitable for waveform *** LSTM-FCNN model does not require extensive data preprocessing and it simplifies the microseismic source location through feature *** this study,we utilized the LSTM-FCNN as a regression learning model to locate the seismic ***,the method of short-time-average/long-time-average(STA/LTA)arrival time picking was employed to augment spatiotemporal ***,oversampling the on-site data was performed to address the issue of data imbalance,and finally,the performance of LSTM-FCNN was ***,we compared the LSTM-FCNN model with previous deep-learning *** results demonstrated remarkable location capabilities with a mean absolute error(MAE)of only 7.16 *** model can realize swift training and high accuracy,thereby significantly improving risk warning of rockbursts.