Rockburst is a common geological disaster in mines, tunnels, deep underground engineering, and during excavation, mining, and construction. Rockburst frequently occurs as the depth of burial increases, and its early warning technology is in urgent need of further development. At present, the most effective monitoring and analysis method of rockburst is microseismic technology, which detects a large number of rock micro-fracture signals through geophones. The identification of microseismic monitoring data is an essential part of microseismic data processing. It is necessary to identify effective microseismic signals from considerable monitoring data for subsequent early warning. Aiming at the identification of rock micro-fracture signals, this thesis proposes a microseismic data identification method based on the Deep Convolution Neural Network Inception (DCNN-Inception) algorithm. The algorithm uses an existing Convolutional Neural Network (CNN) model, adding Inception structure in the middle of the model to form a DCNN-Inception model. A data set was established depending on the actual measured data of Baihetan Hydropower Station, and CNN and DCNN-Inception were employed to identify effective microseismic signals. The results demonstrate that the DCNN-Inception algorithm is better than CNN in recognition accuracy and can effectively identify effective microseismic signals. It provides an essential foundation for the identification of microseismic abnormal signals of rock microfracture and the early warning of rock rupture precursors and is of practical significance for the study of rockburst warning technology.

Determining the precise arrival time of earthquake recordings in noisy settings is crucial for effective earthquake data processing and analysis. This study introduces a novel U-Net3+-based network specially designed for accurate first arrival picking in noisy data. To reduce information loss in seismic waveform extraction, enhancements were made to the U-Net3+ decoder, featuring a residual structure for better information propagation, a convolutional layer replacing the pooling layer to preserve high-frequency seismic waveform details, and a convolutional block attention module (CBAM) integrated into the deeper layers to prioritize critical information at the onset of the seismic wave. We evaluated our model against U-Net, U-Net2+, U-Net3+, and Phase-Net using noisy earthquake recordings, and carried out comprehensive ablation studies on each module. The model’s generalizability was further validated using the “DiTing” dataset. The proposed network model demonstrated superior accuracy and robustness against noise, with picking rates exceeding 92.42%, 90.82%, and 87.81% when signal-to-noise ratios (SNRs) dropped below 20, 10, and 5 dB, respectively. This work contributes a promising approach to developing a more noise-resistant, low-cost seismic phase picking model, and offers potential significant gains for seismic monitoring techniques.

Electronic proceedings of IJCAI 2017

Noise reduction is a critical step in seismic data processing. A novel strong noise reduction network is proposed in this study. The network enhances the U-Net architecture with an improved inception module and coordinate attention (CA) mechanism, suppressing noise and enhancing signal clarity. These enhancements improve the network’s capability to distinguish between signal and noise in the time–frequency domain. We trained and tested our model on the STEAD dataset, which eliminated noise across various frequency bands, improved the signal-to-noise ratio (SNR) of seismic records, and reduced the waveform distortion significantly. Comparative analyses against U-Net, DeepDenoiser, and DnRDB models, using signals with SNRs ranging from −14 dB to 0 dB, demonstrated our model’s superior performance. At the same time, we demonstrated that the Inception Conv Block has a significant impact on the denoising ability of the network. Furthermore, validation using the “Di Ting” dataset and real noisy signals confirmed the model’s generalizability. These results show that the proposed model significantly outperforms the comparative methods in terms of the SNR, correlation coefficient (r), and root mean square error (RMSE), delivering higher-quality seismograms. The enhanced phase-picking accuracy underscores the potential of our approach to advance in geophysics applications.

A new first break picking for three-component (3C) vertical seismic profiling (VSP) data is proposed to improve the estimation accuracy of first arrivals, which adopts gesture detection calibration and polarization analysis based on the eigenvalue of the covariance matrix. This study aims at addressing the problem that calibration is required for VSP data using the azimuth and dip angle of geophones, due to the direction of geophones being random when applied in a borehole, which will further lead to the first break picking possibly being unreliable. Initially, a gesture-measuring module is integrated in the seismometer to rapidly obtain high-precision gesture data (including azimuth and dip angle information). Using re-rotating and re-projecting using earlier gesture data, the seismic dataset of each component will be calibrated to the direction that is consistent with the vibrator shot orientation. It will promote the reliability of the original data when making each component waveform calibrated to the same virtual reference component, and the corresponding first break will also be properly adjusted. After achieving 3C data calibration, an automatic first break picking algorithm based on the autoregressive-Akaike information criterion (AR-AIC) is adopted to evaluate the first break. Furthermore, in order to enhance the accuracy of the first break picking, the polarization attributes of 3C VSP recordings is applied to constrain the scanning segment of AR-AIC picker, which uses the maximum eigenvalue calculation of the covariance matrix. The contrast results between pre-calibration and post-calibration using field data show that it can further improve the quality of the 3C VSP waveform, which is favorable to subsequent picking. Compared to the obtained short-term average to long-term average (STA/LTA) and the AR-AIC algorithm, the proposed method, combined with polarization analysis, can significantly reduce the picking error. Applications of actual field experiments have also confirmed that the proposed method may be more suitable for the first break picking of 3C VSP. Test using synthesized 3C seismic data with low SNR indicates that the first break is picked with an error between 0.75 ms and 1.5 ms. Accordingly, the proposed method can reduce the picking error for 3C VSP data.