Accurately picking P-and S-wave arrivals of microseismic(MS)signals in real-time directly influences the early warning of rock mass failure.A common contradiction between accuracy and computation exists in the current...Accurately picking P-and S-wave arrivals of microseismic(MS)signals in real-time directly influences the early warning of rock mass failure.A common contradiction between accuracy and computation exists in the current arrival picking methods.Thus,a real-time arrival picking method of MS signals is constructed based on a convolutional-recurrent neural network(CRNN).This method fully utilizes the advantages of convolutional layers and gated recurrent units(GRU)in extracting short-and long-term features,in order to create a precise and lightweight arrival picking structure.Then,the synthetic signals with field noises are used to evaluate the hyperparameters of the CRNN model and obtain an optimal CRNN model.The actual operation on various devices indicates that compared with the U-Net method,the CRNN method achieves faster arrival picking with less performance consumption.An application of large underground caverns in the Yebatan hydropower station(YBT)project shows that compared with the short-term average/long-term average(STA/LTA),Akaike information criterion(AIC)and U-Net methods,the CRNN method has the highest accuracy within four sampling points,which is 87.44%for P-wave and 91.29%for S-wave,respectively.The sum of mean absolute errors(MAESUM)of the CRNN method is 4.22 sampling points,which is lower than that of the other methods.Among the four methods,the MS sources location calculated based on the CRNN method shows the best consistency with the actual failure,which occurs at the junction of the shaft and the second gallery.Thus,the proposed method can pick up P-and S-arrival accurately and rapidly,providing a reference for rock failure analysis and evaluation in engineering applications.展开更多
Microseismic event location is one of the core parameters in microseismic monitoring,and the accuracy of localization will directly affect the effectiveness of engineering applications.However,limited by spatial facto...Microseismic event location is one of the core parameters in microseismic monitoring,and the accuracy of localization will directly affect the effectiveness of engineering applications.However,limited by spatial factors,the geometry of the sensor installation will be close to linear,which makes the localization equation suffer from the pathological problem,and the localization accuracy is greatly reduced.To address this problem,the reasons for the pathological problem are analyzed from the perspective of the objective function residuals and coefficient matrix.The pathological problem is caused by the combined effect of the poorer sensor array and data errors,and its residual isosurface shows a conical distribution,and as the residual value decreases,the apex of the isosurface gradually extends to the far side,and the localization results do not converge.For this reason,an improved regularized Newton downhill localization algorithm is proposed.In this method,firstly,the Newtonian downhill method is improved so that the magnitudes of the seismic source parameters are the same,and the condition number of the coefficient matrix is reduced;then,the L-curve method is used to calculate the regularization factor for the pathological equations,and the coefficient matrix is improved;finally,the pathological equations are regularized,and the seismic source coordinates are obtained by the improved Newtonian downhill method.The results of engineering applications show that compared with the traditional algorithm based on automatic of P-arrival picking,the number of effective microseismic events calculated by the proposed localization algorithm is increased by 194.7%,and the localization accuracy is substantially improved.The proposed algorithm reduces the problem of low accuracy of S-arrival picking and allows localization using only P-wave arrival.The method reduces the quality requirements of the data and significantly improves the utilization of microseismic events and positioning accuracy.展开更多
基金We acknowledge the funding support from National Natural Science Foundation of China(Grant No.42077263).
文摘Accurately picking P-and S-wave arrivals of microseismic(MS)signals in real-time directly influences the early warning of rock mass failure.A common contradiction between accuracy and computation exists in the current arrival picking methods.Thus,a real-time arrival picking method of MS signals is constructed based on a convolutional-recurrent neural network(CRNN).This method fully utilizes the advantages of convolutional layers and gated recurrent units(GRU)in extracting short-and long-term features,in order to create a precise and lightweight arrival picking structure.Then,the synthetic signals with field noises are used to evaluate the hyperparameters of the CRNN model and obtain an optimal CRNN model.The actual operation on various devices indicates that compared with the U-Net method,the CRNN method achieves faster arrival picking with less performance consumption.An application of large underground caverns in the Yebatan hydropower station(YBT)project shows that compared with the short-term average/long-term average(STA/LTA),Akaike information criterion(AIC)and U-Net methods,the CRNN method has the highest accuracy within four sampling points,which is 87.44%for P-wave and 91.29%for S-wave,respectively.The sum of mean absolute errors(MAESUM)of the CRNN method is 4.22 sampling points,which is lower than that of the other methods.Among the four methods,the MS sources location calculated based on the CRNN method shows the best consistency with the actual failure,which occurs at the junction of the shaft and the second gallery.Thus,the proposed method can pick up P-and S-arrival accurately and rapidly,providing a reference for rock failure analysis and evaluation in engineering applications.
基金the financial support from the National Natural Science Foundation of China(Grant no.42077263).
文摘Microseismic event location is one of the core parameters in microseismic monitoring,and the accuracy of localization will directly affect the effectiveness of engineering applications.However,limited by spatial factors,the geometry of the sensor installation will be close to linear,which makes the localization equation suffer from the pathological problem,and the localization accuracy is greatly reduced.To address this problem,the reasons for the pathological problem are analyzed from the perspective of the objective function residuals and coefficient matrix.The pathological problem is caused by the combined effect of the poorer sensor array and data errors,and its residual isosurface shows a conical distribution,and as the residual value decreases,the apex of the isosurface gradually extends to the far side,and the localization results do not converge.For this reason,an improved regularized Newton downhill localization algorithm is proposed.In this method,firstly,the Newtonian downhill method is improved so that the magnitudes of the seismic source parameters are the same,and the condition number of the coefficient matrix is reduced;then,the L-curve method is used to calculate the regularization factor for the pathological equations,and the coefficient matrix is improved;finally,the pathological equations are regularized,and the seismic source coordinates are obtained by the improved Newtonian downhill method.The results of engineering applications show that compared with the traditional algorithm based on automatic of P-arrival picking,the number of effective microseismic events calculated by the proposed localization algorithm is increased by 194.7%,and the localization accuracy is substantially improved.The proposed algorithm reduces the problem of low accuracy of S-arrival picking and allows localization using only P-wave arrival.The method reduces the quality requirements of the data and significantly improves the utilization of microseismic events and positioning accuracy.
