Microseismic Signals Research Articles

First-Arrival Picking for Microseismic Monitoring Based on Deep Learning

In microseismic monitoring, achieving an accurate and efficient first-arrival picking is crucial for improving the accuracy and efficiency of microseismic time-difference source location. In the era of big data, the traditional first-arrival picking method cannot meet the real-time processing requirements of microseismic monitoring process. Using the advanced idea of deep learning-based end-to-end classification and the prominent feature extraction advantages of a fully convolution neural network, this paper proposes a first-arrival picking method of effective signals for microseismic monitoring based on UNet++ network, which can significantly improve the accuracy and efficiency of first-arrival picking. In this paper, we first introduced the methodology of the UNet++-based picking method. And then, the performance of the proposed method is verified by the experiments with finite-difference forward modeling simulated signals and actual microseismic records under different signal-to-noise ratios, and finally, comparative experiments are performed using the U-Net-based first-arrival picking algorithm and the Short-Term Average to Long-Term Average (STA/LTA) algorithm. The results show that compared to the U-Net network, the proposed method can obviously improve the first-arrival picking accuracy of the low signal-to-noise ratio microseismic signals, achieving significantly higher accuracy and efficiency than the STA/LTA algorithm, which is famous for its high efficiency in traditional algorithms.

Determination of First Arrival Wave Type of Microseismic Signals and Approach to Wave Velocity Correction

Given the complex environment experienced in working mines, the vibration waves produced by processes such as rock fracture in deep formations usually show interference effects when monitored due to other signals, the so-called “clutter” in the signal, which are interfered with the clutter. At the same time, owing to the influence of system noise, the first arrival time and the arrival time difference values of the signals obtained cannot easily be determined accurately. The propagation model for the microseismic signals experienced and the discrimination method used to determine the first arrival wave type can be established using knowledge of the spatial geometry between the sensors used and the seismic source. Thus, the filtering of the actual from the abnormal wave signals is possible. Using the theory of signal cross-correlation in this work, a correction method for the arrival velocity of the first microseismic signal has been proposed and evaluated. By calculating the cross-correlation coefficient of the same source vibration signal and finding the position that corresponds to the maximum value of the cross-correlation coefficient, the arrival time difference between the signals seen in the two channels is obtained. Thus, the key conclusions can be drawn from the experiments carried out: when the signal-to-noise ratio of the original signal is low, the time difference can still be determined with high accuracy. Further, a wave velocity correction criterion has also been proposed, where the velocity correction of the S wave or the R wave can be realized by combining the spatial coordinate information on the blasting point and an algorithm representing the signal cross-correlation to arrival time difference is used.

A Combined Denoising Method for Microseismic Signals from Coal Seam Hydraulic Fracturing: Multithreshold Wavelet Packet Transform and Improved Hilbert-Huang Transform

Coal seam hydraulic fracturing (CSHF) has recently been applied to mitigate frequent regional rockburst risk in deep mines before mining practice, as an effective substitute for conventional labor-intensive and time-consuming rockburst prevention measures. Due to the complex nature of CSHF microseismic signals—e.g., nonstationary, transient, and low signal-to-noise ratio—conventional denoising methods tend to yield undesirable results that may preclude reliable evaluation of hydraulic fracturing performance using microseismic data. We propose an advanced denoising method MWPT-IHHT to achieve twice denoising in a fine and adaptive manner. This method combines a multithreshold wavelet packet transform (MWPT) and an improved Hilbert-Huang transform (IHHT), with each being improved compared to their conventional counterparts. A quantitative comparison using synthetic signals suggests the outperformance of the proposed method over the commonly used denoising methods in suppressing noises in terms of signal-to-noise ratio, signal similarity, and energy percentage. The desirable denoising results of two typical real CSHF signals in a CSHF test at Huafeng Coal Mine further demonstrate the applicability and effectiveness of the proposed MWPT-IHHT method.

Experimental and numerical study on the fracture characteristics of concrete under uniaxial compression

Since concrete infrastructure may lose its expected physical and functional properties over time, accurate inspection and assessment of such infrastructure systems is necessary to ensure safety and serviceability and prevent unsafe working conditions and the occurrence of hazards. In this paper, an experimental investigation was conducted on several concrete specimens under uniaxial compression to fully explore the fracture process and mechanical characteristics of concrete materials under compressive loading conditions. More specifically, a high-resolution industrial camera and a real-time geophysical acquisition system were employed to capture the fracture process and the microseismic (MS) and electromagnetic emission (EME) signals simultaneously. Based on digital image processing, a robust crack extraction algorithm was proposed for automatically identifying cracks on a concrete surface at the pixel level, and then two novel crack features were introduced to depict the dynamic crack propagation process quantitatively. Then, the correlation between MS, EME, stress drop and dynamic crack evolution characteristics was analyzed. The results indicate that: (1) the MS and EME signals of the concrete specimens have good consistency in the time domain, especially in the fully fractured stage; (2) the crack area, to a certain extent, can reflect the stress drop of a concrete specimen under the compression fracture process; (3) the crack propagation velocity parallel to the loading axis (along the vertical direction) is faster than that the perpendicular one to the loading axis (along the horizontal direction); and (4) the crack area, stress, MS signal, and EME signal reach their peak values nearly simultaneously. Finally, a novel mesh-free numerical model based on peridynamic (PD) theory was established to simulate the fracture process, and the displacement fields were presented to further reveal the failure mechanism of concrete under compressive loading.

A novel method for automatic identification of rock fracture signals in microseismic monitoring

Microseismic (MS) monitoring technology is able to offer 3-dimensional and real-time rupture characteristics of rocks by capturing and analyzing detectable elastic waves released from rock fracture, which has been widely applied in rock/mining engineering. The prerequisite of MS analysis is to identify MS signals with high efficiency and accuracy. However, the identification work is inevitably disturbed by vibration signals from the blasting activities involved in rock/mining engineering. Currently, the classification work of MS and blasting signals is mainly carried out by the on-site operators based on their own experience, which is time-consuming and subjectively sensitive. In this work, a new discriminant method is developed to automatically recognize these signals based on their time-frequency spectrum characteristics to offer more reliable data for MS monitoring analysis. Singular value decomposition is adopted to reduce data volume and obtain unique features. Random Forest (RF) is applied to achieve automatic recognition and get the most important features based on the variation of ‘out-of-bag’ error and Gini Index. These most important features can be used to optimize the RF classifier. The numerical experiments are conducted to verify our new method. The results prove that the method reaches 95.99%±0.1% accuracy in 100 repetitions for events collected from the field, and the optimized RF produces a more accurate result (97%) using fewer characteristics, which is higher than the methods reported in the previous literature. The proposed method diminishes the sensitivity of operators in the classification and promotes the accuracy and efficiency of the identification of MS signals data in the application of MS monitoring technology. Moreover, this strategy also offers a new strategy to process the acquired monitoring signals in other monitoring techniques.

Microseismic Signals in Heading Face of Tengdong Coal Mine and Their Application for Rock Burst Monitoring

Microseismic (MS) monitoring is an important and commonly used geophysical method in coal mines to predict rock burst which has great influence on safety production. MS monitoring technology and analysis method of the whole mine or working face have been matured, but its use in heading faces of coal mine is not mature due to small disturbances and narrow layout spaces. To carry out MS monitoring and early warning in the heading face, signal recognition must be adequately performed first, and monitoring objects and indicators must be obtained. Through field tests of MS systems at the 117 track gateway of Tengdong coal mine, interference signals of equipment operation and effective signals of coal vibration are accurately collected. After analysis, the waveform characteristics, spectrum, and propagation distance of the interference signals and coal vibration signal are different. Some effective signals with small energy (one-channel triggering) cannot be used as early warning indicators because they are concealed by interference signals. Through trial operation, it is found that large energy (three-channel and four-channel triggering) coal vibration events successfully predicted a rock burst. The MS system of 117 track gateway of Tengdong coal mine should be able to remove the interference signals in real time through the algorithm and take the number of large energy coal vibration signal rather than all coal vibration events as the predictor for rock burst risk monitoring.

A hybrid recognition model of microseismic signals for underground mining based on CNN and LSTM networks

Microseismic (MS) monitoring technology has been widely used to monitor ground pressure disasters. However, the underground mining environment is complex and contains many types of noise sources. Furthermore, the traditional recognition method entails a complex process with low recognition accuracy for MS signals, so it is difficult to serve for the safe production of mines. Therefore, this study established a hybrid model combining the singular spectrum analysis (SSA) method, convolutional neural networks (CNN), and long short-term memory networks (LSTM). First, the principal components of monitoring signals were extracted with the SSA method, and then spatial and temporal features of monitoring signals were separately extracted with the CNN and LSTM. Based on actual field data collected from Xiadian Gold Mine, the hybrid model was compared with the CNN, LSTM, and back-propagation networks (BP), as well as commonly used recognition methods including the support vector machine (SVM), decision tree (DT), K-nearest neighbor (KNN), and linear discriminant analysis (LDA). The results show that the proposed hybrid model can accurately extract data features of monitoring signals and further improve MS signals' recognition performance. Furthermore, the recognition accuracy of mechanical signals in monitoring signals is particularly increased using the hybrid model, which avoids confusion with MS signals.

The Use of Intelligent Sensing Algorithm for Internet of Things and Hash Spatial Information Location Technology

This exploration is aimed at quickly obtaining the spatial position information of microseismic focal points and increasing the accuracy of microseismic rapid positioning, to take timely corresponding measures. A microseismic focal point location system completely different from the traditional microseismic location method is proposed. The search engine technology is introduced into the system, which can locate the microseismic focal point quickly and accurately. First, the propagation characteristics of microseismic signals in coal and rock layers are analyzed, and the focal position information is obtained. However, the collected microseismic signal of the coal mine contains noise, so it is denoised at first. Then, a waveform database is established for the denoised waveform data and focal point position. The structure and mathematical model of the location‐sensitive hash (LSH) based on P stable distribution are introduced and improved, and the optimized algorithm multiprobe LSH is obtained. The microseismic location model is established according to the characteristics of microseismic data. The values of three parameters, hash table number, hash function family dimension, and interval size, are determined. The experimental data of the parameters of the search engine algorithm are analyzed. The results show that when the number of hash tables is 6, the dimension k of the hash function family is 14, and the interval size W is 8000, the retrieval time reaches a relatively small value, the recall rate reaches a large value, and the proportion of retrieved candidates is large; the parameters of the search engine algorithm of the measured coal mine microseismic data are analyzed. It is obtained that when the number of hash tables is 4, the dimension k of the hash function family is 6, and the interval size W is 500, the retrieval time reaches a relatively small value, the recall rate obtains a large value, and the proportion of retrieved candidates is large. The contents studied are of great significance to the evaluation of destructive mine earthquakes and impact risk.

Recognition of Rock Micro-Fracture Signal Based on Deep Convolution Neural Network Inception Algorithm

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.

Microseismic denoising assessment by polarization histograms

We have performed a comparison of microseismic data denoising methods based on their effect on the polarization attributes of 3C microseismic signals. The compared denoising methods include the classic band-pass filtering, and three recently proposed denoising techniques: restricted-domain hyperbolic Radon transform denoising, singular value decomposition (SVD)-based reduced-rank filtering, and empirical mode decomposition denoising. In order to draw the comparison, we denoised 3C synthetic data contaminated with noise extracted from field data records; calculated their rectilinearity, azimuth, and dip polarization attributes; and arranged them into histograms. The comparison has been drawn by measuring the distances between the polarization histograms of the clean and denoised data, assuming that one method outperforms another if the aforementioned distance is smaller. This strategy allows us to quantify the improvement in the calculated polarization attributes due to the different denoising processes. In addition, we also calculated the quality factor (QF) of the denoised signals, which adds value and robustness to the comparison. Our results indicated that the method based on SVD preserves the original polarization attributes better than the other techniques tested in this work. Moreover, it also retrieved the denoised signal with the highest QF. Finally, we tested the methods with field data and assessed their performance qualitatively on the basis of the insight gained from numerical tests with synthetic data.

Microseismic moment tensor based analysis of rock mass failure mechanism surrounding an underground powerhouse

The failure mechanism of surrounding rock is an important issue in the field of underground engineering. To quantitatively understand the failure process and failure mechanism of the surrounding rock mass in deep underground engineering, the underground powerhouse of the Houziyan hydropower station was selected as a case study. The study area was determined by analysing the spatiotemporal distribution of microseismic (MS) events and site investigations. After noise reduction, P-wave arrival time extraction and positioning of MS signals, the moment tensor inversion method was used to determine the failure mechanism and fracture direction of the surrounding rock mass. The results showed that (1) the failure processes and mechanism of the surrounding rock mass can be effectively analysed through MS monitoring with moment tensor analysis and (2) the failure type of the surrounding rock mass downstream of the main powerhouse was characterized by shear failure. The stress state in this region was a strike-slip stress state. (3) The failure type of the surrounding rock mass at the arch roof of the main transformer chamber was characterized by tensile failure. The tensile stress of the intermediate stress axis was obvious, and the tectonic stress presented a state of near uniaxial compression-biaxial tension.

Microseismic Monitoring, Positioning Principle, and Sensor Layout Strategy of Rock Mass Engineering

Microseismic monitoring technology can start from the most initial stage of rock deformation and track and monitor the progressive failure process of rock mass from the fracture of rock blocks to the instability of rock mass. Thus, the scientific nature of monitoring work is greatly promoted, and the accuracy and advance of the prediction of engineering and geological disasters are improved. In this paper, the fracture and instability of rock can be analyzed by analyzing the microseismic signals produced by rock failure; The location of microseismic source can be determined by multipoint synchronous data acquisition to determine the time when each sensor (at least 5) receives microseismic signals; Combined with practical engineering experience for underground engineering in growth, heading tunnel, put forward only sensor arrangement to take mobile, follow the semienclosed layout network. We hope to give some reference to the researchers who are concerned with microseismic monitoring technology.

Arrival Picking of Acoustic Emission Signals Using a Hybrid Algorithm Based on AIC and Histogram Distance

Automatic picking of the arrival of acoustic emission (AE) and microseismic signals is very important in geophysics and process monitoring. However, because an acoustic signal undergoes many reflections and refractions before being received by an AE sensor, its signal-to-noise ratio (SNR) varies and, in some cases, can be very low. A new hybrid method for arrival-time picking based on the Akaike information criterion (AIC) picker and Manhattan distance is presented. A criterion and a standard framework for determining an arrival point are formulated, and two key parameters are discussed, namely, the length of the time window and the size of the selected time series. Experimental results show that the proposed method can detect the arrival points of signals with various SNRs while overcoming some of the drawbacks of the conventional AIC method. It could provide a better alternative for determining the arrival points of AE signals in process monitoring and other fields.

Evaluation of Borehole Hydraulic Fracturing in Coal Seam Using the Microseismic Monitoring Method

Accurate evaluation of the influence range of borehole hydraulic fracturing (HF) in coal seam is crucial for optimizing the design scheme of HF. In this study, we adopted the microseismic (MS) monitoring technology to monitor and characterize the spatial shape of cracks caused by borehole HF in coal seam in an underground coal mine. And we also tested and analyzed the stress and moisture content changes of coal mass at different distances from the borehole after HF program. The number of MS waveforms and the energy of MS events show a good positive correlation with the water pressure curve. The response of the energy curve to the extension of the hydraulic cracks is ahead of the water pressure curve. Based on the short-time average/long-time average (STA/LTA), the interference signal recognition method (ISR) and the improved Akaike information criterion (AIC) method, we developed a comprehensive MS event detection and arrival time picking (CMDP) program that are suitable for the weak MS signals with low signal to noise ratio (SNR) induced by HF in coal seam. And then we were able to more accurately locate the MS events of the hydraulic cracks using the simplex source location method. We have conducted a comprehensive analysis of the relationship between the temporal and spatial distribution of MS events and hydraulic cracks propagation. The results show that there is an apparent correlation between the MS activities and HF operation. The expansion of hydraulic cracks generates MS events, and the larger the size of the cracks, the greater the energy of MS events. Based on the MS monitoring results, the HF produces a crack network of flat ellipsoid in the No. 6 coal seam, which indicates that there is obvious stimulated reservoir volume (SRV) fracturing effect during the borehole HF process. The influence radius of HF based on the moisture content is the smallest (about 20 m), followed by the stress monitoring (about 30 m), and the MS monitoring is the largest (about 40 m). The high-precision MS source location (location errors < 2.5 m) results combined with roadway roof watering phenomenon indicate that the influence radius of HF based on the moisture content and stress release may be underestimated. And the effective influence radius of the borehole HF is about 40 m for this HF program.

Deep Convolutional Neural Network for Microseismic Signal Detection and Classification

Reliable automatic microseismic waveform detection with high efficiency, precision, and adaptability is the basis of stability analysis of the surrounding rock mass. In this paper, a convolutional neural network (CNN)-based microseismic detection network (CNN-MDN) model was established and well trained to a high degree of accuracy using a dataset with 16,000 preprocessed waveforms. By comparison with other methods, 4000 waveforms were tested to evaluate the precision, recall, and F1-score. The results revealed that the CNN-MDN demonstrated the highest performance in microseismic detection. Moreover, the low sensitivity of the CNN-MDN to noise of different intensities was proved by testing on semi-synthetic data. The model also possesses good generalization ability and superior performance capability for microseismic detection under different geological structure backgrounds, and it can correctly detect the microseismic events with Mw ≥ 0.5. These preliminary results show that the CNN-MDN can be directly applied to unprocessed microseismic data and has great potential in real-time microseismic monitoring applications.

Automated Platform for Microseismic Signal Analysis: Denoising, Detection, and Classification in Slope Stability Studies

Microseismic monitoring has been increasingly used in the past two decades to illuminate (sub)surface processes, such as landslides, due to its ability to record small seismic waves generated by soil movement and/or brittle behavior of rock. Understanding the evolution of landslide processes is of paramount importance in predicting or even avoiding an imminent failure. Microseismic monitoring recordings are often continuous, noisy, and consist of signals emitted by various sources. Manually detecting and distinguishing the signals emitted by an unstable slope is challenging. Research on automated end-to-end denoising, detection, and classification of microseismic events, as an early warning system, is still in its infancy. To this effect, our work is focused on jointly evaluating and developing suitable approaches for signal denoising, accurate event detection, nonsite-specific feature construction, feature selection, and event classification. We propose an automated end-to-end system that can process big data sets of continuous seismic recordings fast and demonstrate applicability and robustness to a wide range of events (distant and local earthquakes, slidequakes, anthropogenic noise, etc.). Algorithmic contributions lie in novel signal processing and analysis methods with fewer tunable parameters than the state of the art, evaluated on two field data sets and benchmarked against the state of the art.

Automatic waveform-based source-location imaging using deep learning extracted microseismic signals

We have used an automatic unsupervised technique to extract waveform signals from continuous microseismic data. First, the time-frequency representation (scalogram) is obtained for the input microseismic trace. Second, the convolutional autoencoder (CAE) is used to extract the significant scalogram features related to the waveform signals and discard the rest. Third, the extracted features from the CAE encoder are considered as the input for the k-means clustering algorithm, in which the input samples are classified into waveform and nonwaveform components. The proposed algorithm is evaluated using several synthetic and field examples. We find that the proposed algorithm successfully extracts the waveform signals even in a noisy environment with a signal-to-noise-ratio as low as −10 dB. We compared the proposed algorithm to benchmark algorithms, for example, simple k-means and short-term and long-term average ratio methods, and find that the proposed algorithm performs best. We find that the detected waveform signals can enhance the resolution of microseismic imaging using a waveform-based reverse time migration method.

Identification of Microseismic Events in Rock Engineering by a Convolutional Neural Network Combined with an Attention Mechanism

Microseismic technology has widely been used in many rock engineering applications to shield workers from engineering hazards and monitor underground construction. To avoid the heavy workloads imposed by the manual recognition of many microseismic signals, this study proposes a new end-to-end training network architecture to automatically identify microseismic events. A dataset including not only easily identifiable microseismic signals but also barely distinguishable nontypical data has been collected from a practical rock engineering project for training and testing the network model. The applicability of various networks for this task is discussed to select the best method for microseismic recognition. We modify the residual skip connections to make them more suitable for the signal classification task. Then, the novel depthwise spatial and channel attention (DSCA) module is proposed. This module can autonomously learn how to weight information with different levels of importance, similar to human attention, which greatly improves the network performance without incurring additional computational costs. Theoretically, it can be a useful tool to replace traditional denoising algorithms and model the interdependencies between the different channels of a multichannel signal. Furthermore, the DSCA module and the modified residual connections are combined with a traditional convolutional network to obtain a novel network architecture named ResSCA and the results of comparative experiments are presented. Finally, single- and multichannel models are constructed based on ResSCA, which achieved improved accuracy rates. Their advantages and drawbacks are analyzed. This study presents a modified network architecture suitable for identifying and classifying complex signals to enable intelligent microseismic monitoring, which is valuable for various rock engineering applications.

Analysis of hydraulic fracture behavior and well pattern optimization in anisotropic coal reservoirs

Macrolithotypes control the pore-fracture distribution heterogeneity in coal, which impacts stimulation via hydrofracturing and coalbed methane (CBM) production in the reservoir. Here, the hydraulic fracture was evaluated using the microseismic signal behavior for each macrolithotype with microfracture imaging technology, and the impact of the macrolithotype on hydraulic fracture initiation and propagation was investigated systematically. The result showed that the propagation types of hydraulic fractures are controlled by the macrolithotype. Due to the well-developed natural fracture network, the fracture in the bright coal is more likely to form the “complex fracture network”, and the “simple” case often happens in the dull coal. The hydraulic fracture differences are likely to impact the permeability pathways and the well productivity appears to vary when developing different coal macrolithtypes. Thus, considering the difference of hydraulic fracture and permeability, the CBM productivity characteristics controlled by coal petrology were simulated by numerical simulation software, and the rationality of well pattern optimization factors for each coal macrolithotype was demonstrated. The results showed the square well pattern is more suitable for dull coal and semi-dull coal with undeveloped natural fractures, while diamond and rectangular well pattern is more suitable for semi-bright coal and bright coal with more developed natural fractures and more complex fracturing fracture network; the optimum wells spacing of bright coal and semi-bright coal is 300 m and 250 m, while that of semi-dull coal and dull coal is just 200 m.

Microseismic Signal Denoising and Separation Based on Fully Convolutional Encoder–Decoder Network

Denoising methods are a highly desired component of signal processing, and they can separate the signal of interest from noise to improve the subsequent signal analyses. In this paper, an advanced denoising method based on a fully convolutional encoder–decoder neural network is proposed. The method simultaneously learns the sparse features in the time–frequency domain, and the mask-related mapping function for signal separation. The results show that the proposed method has an impressive performance on denoising microseismic signals containing various types and intensities of noise. Furthermore, the method works well even when a similar frequency band is shared between the microseismic signals and the noises. The proposed method, compared to the existing methods, significantly improves the signal–noise ratio thanks to minor changes of the microseismic signal (less distortion in the waveform). Additionally, the proposed methods preserve the shape and amplitude characteristics so that it allows better recovery of the real waveform. This method is exceedingly useful for the automatic processing of the microseismic signal. Further, it has excellent potential to be extended to the study of exploration seismology and earthquakes.