Research on automatic picking method of effective events of microseismic P-wave in NARX neural network
Received date: 2023-06-25
Online published: 2025-01-14
Copyright
Effective event picking is the basic and critical link in microseismic data processing, but the picking effect of traditional picking methods depends heavily on the selection of picking parameters, and it is easily affected by signal characteristics and signal-to-noise ratio, which is difficult to meet the requirements of real-time processing of massive monitoring data. Based on this, in order to effectively meet the current processing needs of large-area, wide-azimuth and high-density monitoring data, based on the structural characteristics of NARX neural network, a network model that meets the needs is constructed, and the preprocessed synthetic and measured microseismic signals are fed into its Series-Parallel feedback structure for training to fully learn the waveform characteristics of the signals. Then, the monitoring signal to be picked is simply processed and inputted into the NARX model with good performance to output the characteristic curve, and the effective event is picked by the threshold. The results of picking of synthetic and measured microseismic signals of the test database and the comparison with the STA/LTA method show that the NARX method can complete the picking of effective events of microseismic signals, and it shows strong noise resistance and adaptability, and it has certain advantages over STA/LTA method, so it has the potential to become one of the effective tools for picking the events of microseismic monitoring signals.
Hao LUO , DaiPing PENG , Peng HUANG , Chi ZHAO . Research on automatic picking method of effective events of microseismic P-wave in NARX neural network[J]. Progress in Geophysics, 2024 , 39(6) : 2345 -2356 . DOI: 10.6038/pg2024HH0064
图8 合成微地震信号有效事件拾取(不含背景噪声)(a)NARX法与STA/LTA法提取特征曲线;(b)(c)NARX法拾取结果.左:单道信号,右:阵列信号,下同.红色线段:理论有效事件,蓝色线段:NARX拾取有效事件,下同. Fig 8 Effective event picking of synthetic microseismic signal (without background noise) (a)Extracted characteristic curves of NARX method and STA/LTA method; (b)(c)Picked result of NARX method.Left: Single channel signal, Right: Array signals, Same below.Red line segment: Theoretical valid events, Blue line segment: Picked valid events of NARX, Same below. |
图9 合成微地震信号有效事件拾取(SNR=10 dB)(a)NARX法与STA/LTA法提取特征曲线;(b)(c)NARX法拾取结果. Fig 9 Effective event picking of synthetic microseismic signal (SNR=10 dB) (a)Extracted characteristic curves of NARX method and STA/LTA method; (b)(c)Picked result of NARX method. |
图10 合成微地震信号有效事件拾取(SNR=5 dB)(a)NARX法与STA/LTA法提取特征曲线;(b)(c)NARX法拾取结果. Fig 10 Effective event picking of synthetic microseismic signal (SNR=5 dB) (a)Extracted characteristic curves of NARX method and STA/LTA method; (b)(c)Picked result of NARX method. |
图11 合成微地震信号有效事件拾取(SNR=0 dB)(a)NARX法与STA/LTA法提取特征曲线;(b)(c)NARX法拾取结果. Fig 11 Effective event picking of synthetic microseismic signal (SNR=0 dB) (a)Extracted characteristic curves of NARX method and STA/LTA method; (b)(c)Picked result of NARX method. |
表1 STA/LTA法不同信噪比合成数据拾取结果评价Table 1 Evaluation of picked results of synthetic data with different signal-to-noise ratios by STA/LTA method |
| 指标 | 不含噪声 | SNR=10 dB | SNR=5 dB | SNR=0 dB | SNR=-5 dB |
| 拾取率 | 1 | 0.6965 | 0.4925 | 0.3881 | 0.3433 |
| 错拾率 | 0 | 0.3035 | 0.5075 | 0.6119 | 0.6567 |
| 平均相对误差 | 0.0125 | 0.0144 | 0.0135 | 0.0162 | 0.0133 |
表2 NARX法不同信噪比合成数据拾取结果评价Table 2 Evaluation of picked results of synthetic data with different signal-to-noise ratios by NARX method |
| 指标 | 不含噪声 | SNR=10 dB | SNR=5 dB | SNR=0 dB | SNR=-5 dB |
| 拾取率 | 0.9055 | 0.9104 | 0.8408 | 0.7214 | 0.7065 |
| 错拾率 | 0.0945 | 0.0896 | 0.1592 | 0.2786 | 0.2935 |
| 平均相对误差 | 0.0119 | 0.0156 | 0.0134 | 0.0146 | 0.0155 |
图14 实测微地震信号有效事件拾取(No=01)(a)NARX法与STA/LTA法提取特征曲线;(b)(c)NARX法拾取结果.左:单道信号,右:阵列信号,下同.红色线段:理论有效事件,蓝色线段:NARX拾取有效事件,下同. Fig 14 Effective event picking of measured microseismic signal (No=01) (a)Extracted characteristic curves of NARX method and STA/LTA method; (b)(c)Picked result of NARX method.Left: Single channel signal, Right: Array signals, Same below.Red line segment: Theoretical valid events, Blue line segment: Picked valid events of NARX, Same below. |
图15 实测微地震信号有效事件拾取(No=02)(a)NARX法与STA/LTA法提取特征曲线;(b)(c)NARX法拾取结果. Fig 15 Effective event picking of measured microseismic signal (No=02) (a)Extracted characteristic curves of NARX method and STA/LTA method; (b)(c)Picked result of NARX method. |
表3 STA/LTA法实测数据拾取结果评价Table 3 Evaluation of picked results of measured data by STA/LTA method |
| 指标 | No=01 | No=02 | No=03 | Mean |
| 拾取率 | 0.55 | 0.6471 | 0.6667 | 0.6213 |
| 错拾率 | 0.45 | 0.3529 | 0.1429 | 0.3153 |
| 平均相对误差 | 0.0772 | 0.0326 | 0.0305 | 0.0468 |
表4 NARX法实测数据拾取结果评价Table 4 Evaluation of picked results of measured data by NARX method |
| 指标 | No=01 | No=02 | No=03 | Mean |
| 拾取率 | 0.8 | 0.7059 | 0.8095 | 0.7718 |
| 错拾率 | 0.1 | 0.2353 | 0.1429 | 0.1594 |
| 平均相对误差 | 0.0649 | 0.0277 | 0.0286 | 0.0404 |
感谢审稿专家提出的修改意见和编辑部的大力支持!
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/
| 〈 |
|
〉 |