Ensemble Learning Improves the Efficiency of Microseismic Signal Classification in Landslide Seismic Monitoring.

A deep-seated landslide could release numerous microseismic signals from creep-slip movement, which includes a rock-soil slip from the slope surface and a rock-soil shear rupture in the subsurface. Machine learning can effectively enhance the classification of microseismic signals in landslide seismic monitoring and interpret the mechanical processes of landslide motion. In this paper, eight sets of triaxial seismic sensors were deployed inside the deep-seated landslide, Jiuxianping, China, and a large number of microseismic signals related to the slope movement were obtained through 1-year-long continuous monitoring. All the data were passed through the seismic event identification mode, the ratio of the long-time average and short-time average. We selected 11 days of data, manually classified 4131 data into eight categories, and created a microseismic event database. Classical machine learning algorithms and ensemble learning algorithms were tested in this paper. In order to evaluate the seismic event classification performance of each algorithmic model, we evaluated the proposed algorithms through the dimensions of the accuracy, precision, and recall of each model. The validation results demonstrated that the best performing decision tree algorithm among the classical machine learning algorithms had an accuracy of 88.75%, while the ensemble algorithms, including random forest, Gradient Boosting Trees, Extreme Gradient Boosting, and Light Gradient Boosting Machine, had an accuracy range from 93.5% to 94.2% and also achieved better results in the combined evaluation of the precision, recall, and F1 score. The specific classification tests for each microseismic event category showed the same results. The results suggested that the ensemble learning algorithms show better results compared to the classical machine learning algorithms.

A Combined Method of Seismic Monitoring and Transient Electromagnetic Detection for the Evaluation of Hydraulic Fracturing Effect in Coal Burst Prevention.

In order to mitigate the risk of roof-dominated coal burst in underground coal mining, horizontal long borehole staged hydraulic fracturing technology has been prevailingly employed to facilitate the weakening treatment of the hard roof in advance. Such weakening effect, however, can hardly be evaluated, which leads to a lack of a basis in which to design the schemes and parameters of hydraulic fracturing. In this study, a combined underground-ground integrated microseismic monitoring and transient electromagnetic detection method was utilized to carry out simultaneous evaluations of the seismic responses to each staged fracturing and the apparent resistivity changes before and after all finished fracturing. On this basis, the comparable and applicable fracturing effects on coal burst prevention were evaluated and validated by the distribution of microseismic events and their energy magnitude during the mining process. Results show that the observed mining-induced seismic events are consistent with the evaluation results obtained from the combined seismic-electromagnetic detection method. However, there is a limited reduction effect on resistivity near the fractured section that induces far-field seismic events. Mining-induced seismic events are concentrated primarily within specific areas, while microseismic events in the fractured area exhibit high frequency but low energy overall. This study validates the rationality of combined seismic-electromagnetic detection results and provides valuable insights for optimizing fracturing construction schemes as well as comprehensively evaluating outcomes associated with underground directional long borehole staged hydraulic fracturing.

Risk Field of Rock Instability using Microseismic Monitoringdata in Deep Mining.

With the gradual depletion of surface resources, rock instability caused by deep high stressand mining disturbance seriously affects safe mining. To create effective risk management, a rockinstability risk field model using microseismic monitoring data is proposed in this study. Rockinstability risk was presented visually in 3D visualization. The in-situ microseismic monitoringdata was collected and analyzed to make calculation of peak ground velocity (PGV), peak groundacceleration (PGA), energy flux, energy and seismic moment. Indicator weights of PGV, PGA, energyflux are confirmed by using the analytic hierarchy process (AHP) to calculate risk severity. The Copulafunction is then used to solve the joint probability distribution function of energy and seismic moment.Then the spatial distribution characteristics of risk can be obtained by data fitting. Subsequently, thethree-dimensional (3D) risk field model was established. Meanwhile, the established risk field isverified by comparing monitoring data without disturbance and the blasting data with disturbance.It is suggested that the proposed risk field method could evaluate the regional risk of rock instabilityreasonably and accurately, which lays a theoretical foundation for the risk prediction and managementof rock instability in deep mining.

Experimental Study on Calibration of Amplitude-Frequency Measurement Deviation for Microseismic Sensors in Coal Mines.

Microseismic monitoring systems (MMS) have become increasingly crucial in detecting tremors in coal mining. Microseismic sensors (MS), integral components of MMS, profoundly influence positioning accuracy and energy calculations. Hence, calibrating these sensors holds immense importance. To bridge the research gap in MS calibration, this study conducted a systematic investigation. The main conclusions are as follows: based on calibration tests on 102 old MS using the CS18VLF vibration table, it became evident that certain long-used MS in coal mines exhibited significant deviations in frequency and amplitude measurements, indicating sensor failure. Three important calibration indexes, frequency deviation, amplitude deviation, and amplitude linearity are proposed to assess the performance of MS. By comparing the index of old and new MS, critical threshold values were established to evaluate sensor effectiveness. A well-functioning MS exhibits an absolute frequency deviation below 5%, an absolute amplitude deviation within 55%, and amplitude linearity surpassing 0.95. In normal operations, the frequency deviation of MS is significantly smaller than the amplitude deviation. Simplified waveform analysis has unveiled a linear connection between amplitude deviation and localization results. An analysis of the Gutenberg-Richter microseismic energy calculation formula found that the microseismic energy calculation is influenced by both the localization result and amplitude deviation, making it challenging to pinpoint the exact impact of amplitude deviation on microseismic energy. Reliable MS, as well as a robust MS, serve as the fundamental cornerstone for acquiring dependable microseismic data and are essential prerequisites for subsequent microseismic data mining. The insights and findings presented here provide valuable guidance for future MS calibration endeavors and ultimately can guarantee the dependability of microseismic data.

Multi-Disciplinary Monitoring Networks for Mesoscale Underground Experiments: Advances in the Bedretto Reservoir Project.

The Bedretto Underground Laboratory for Geosciences and Geoenergies (BULGG) allows the implementation of hectometer (>100 m) scale in situ experiments to study ambitious research questions. The first experiment on hectometer scale is the Bedretto Reservoir Project (BRP), which studies geothermal exploration. Compared with decameter scale experiments, the financial and organizational costs are significantly increased in hectometer scale experiments and the implementation of high-resolution monitoring comes with considerable risks. We discuss in detail risks for monitoring equipment in hectometer scale experiments and introduce the BRP monitoring network, a multi-component monitoring system combining sensors from seismology, applied geophysics, hydrology, and geomechanics. The multi-sensor network is installed inside long boreholes (up to 300 m length), drilled from the Bedretto tunnel. Boreholes are sealed with a purpose-made cementing system to reach (as far as possible) rock integrity within the experiment volume. The approach incorporates different sensor types, namely, piezoelectric accelerometers, in situ acoustic emission (AE) sensors, fiber-optic cables for distributed acoustic sensing (DAS), distributed strain sensing (DSS) and distributed temperature sensing (DTS), fiber Bragg grating (FBG) sensors, geophones, ultrasonic transmitters, and pore pressure sensors. The network was realized after intense technical development, including the development of the following key elements: rotatable centralizer with integrated cable clamp, multi-sensor in situ AE sensor chain, and cementable tube pore pressure sensor.

Optimization and Quality Assessment of Arrival Time Picking for Downhole Microseismic Events.

Arrival-time picking is a critical step in microseismic data processing, and thus the quality control of arrival results is necessary. Conventional picking methods may be inaccurate or inconsistent due to varied signal-to-noise ratios (SNR) and waveform patterns of the events recorded in different time sections. To address this issue, we propose a quality assessment method based on waveform similarity coefficients to evaluate arrival results and also a global optimization algorithm based on iterative cross-correlation to refine arrival times. The recordings after moveout correction are applied to calculate the intra-event and inter-event waveform coefficients for the quality assessment of arrival results. The residual time differences of intra-event and inter-event traces are calculated sequentially using an enhanced iterative cross-correlation method. In addition, the stacked waveform of each event after the intra-event residual time correction is introduced for global optimization to obtain the inter-event residual time discrepancies. We use both synthetic data and field data to validate the proposed method. The results indicate that the proposed method yields more robust and reliable results. The quality assessment of the optimized arrivals is greatly enhanced compared to the adjusted picks obtained from single event-based processing methods.

An Optimization Method for the Station Layout of a Microseismic Monitoring System in Underground Mine Engineering.

The layout of microseismic monitoring (MSM) station networks is very important to ensure the effectiveness of source location inversion; however, it is difficult to meet the complexity and mobility requirements of the technology in this new era. This paper proposes a network optimization method based on the geometric parameters of the proposed sensor-point database. First, according to the monitoring requirements and mine-working conditions, the overall proposed point database and model are built. Second, through the developed model, the proposed coverage area, envelope volume, effective coverage radius, and minimum energy level induction value are comprehensively calculated, and the evaluation reference index is constructed. Third, the effective maximum envelope volume is determined by taking the analyzed limit of monitoring induction energy level as the limit. Finally, the optimal design method is identified and applied to provide a sensor station layout network with the maximum energy efficiency. The method, defined as the S-V-E-R-V model, is verified by a comparison with the existing layout scheme and numerical simulation. The results show that the optimization method has strong practicability and efficiency, compared with the mine's layout following the current method. Simulation experiments show that the optimization effect of this method meets the mine's engineering requirements for the variability, intelligence, and high efficiency of the microseismic monitoring station network layout, and satisfies the needs of event identification and location dependent on the station network.

Machine Learning Based Identification of Microseismic Signals Using Characteristic Parameters.

Microseismic monitoring system is one of the effective means to monitor ground stress in deep mines. The accuracy and speed of microseismic signal identification directly affect the stability analysis in rock engineering. At present, manual identification, which heavily relies on manual experience, is widely used to classify microseismic events and blasts in the mines. To realize intelligent and accurate identification of microseismic events and blasts, a microseismic signal identification system based on machine learning was established in this work. The discrimination of microseismic events and blasts was established based on the machine learning framework. The microseismic monitoring data was used to optimize the parameters and validate the classification methods. Subsequently, ten machine learning algorithms were used as the preliminary algorithms of the learning layer, including the Decision Tree, Random Forest, Logistic Regression, SVM, KNN, GBDT, Naive Bayes, Bagging, AdaBoost, and MLP. Then, training set and test set, accounting for 50% of each data set, were prospectively examined, and the ACC, PPV, SEN, NPV, SPE, FAR and ROC curves were used as evaluation indexes. Finally, the performances of these machine learning algorithms in microseismic signal identification were evaluated with cross-validation methods. The results showed that the Logistic Regression classifier had the best performance in parameter identification, and the accuracy of cross-validation can reach more than 0.95. Random Forest, Decision Tree, and Naive Bayes also performed well in this data set. There were some differences in the accuracy of different classifiers in the training set, test set, and all data sets. To improve the accuracy of signal identification, the database of microseismic events and blasts should be expanded, to avoid the inaccurate data distribution caused by the small training set. Artificial intelligence identification methods, including Random Forest, Logistic Regression, Decision Tree, Naive Bayes, and AdaBoost algorithms, were applied to signal identification of the microseismic monitoring system in mines, and the identification results were consistent with the actual situation. In this way, the confusion caused by manual classification between microseismic events and blasts based on the characteristics of waveform signals is solved, and the required source parameters are easily obtained, which can ensure the accuracy and timeliness of microseismic events and blasts identification.

Deep Neural Networks for Detection and Location of Microseismic Events and Velocity Model Inversion from Microseismic Data Acquired by Distributed Acoustic Sensing Array.

Fiber-optic cables have recently gained popularity for use as Distributed Acoustic Sensing (DAS) arrays for borehole microseismic monitoring due to their physical robustness as well as high spatial and temporal resolutions. As a result, the sensors record large amounts of data, making it very difficult to process in real-/semi-real-time using the conventional processing routines. We present a novel approach, based on deep learning, for handling the large amounts of DAS data in real-/semi-real-time. The proposed neural network was trained on synthetic microseismic data contaminated with real-ambient noise from field data and was validated using field DAS microseismic data obtained from a hydraulic fracturing operation. The results indicate that the trained network is capable of detecting and locating microseismic events from DAS data and simultaneously update the velocity model to a high degree of precision. The mean absolute errors in the event locations and the velocity model parameters are 2.04, 0.72, 2.76, 4.19 and 0.97 percent for distance (x), depth (z), P-wave velocity, S-wave velocity and density, respectively. In addition to automation and computational efficiency, deep learning reduces human expert data handling during processing, thus preserving data integrity leading to more accurate and reproducible results.

Microseismic P-Wave Travel Time Computation and 3D Localization Based on a 3D High-Order Fast Marching Method.

The travel time computation of microseismic waves in different directions (particularly, the diagonal direction) in three-dimensional space has been found to be inaccurate, which seriously affects the localization accuracy of three-dimensional microseismic sources. In order to solve this problem, this research study developed a method of calculating the P-wave travel time based on a 3D high-order fast marching method (3D_H_FMM). This study focused on designing a high-order finite-difference operator in order to realize the accurate calculation of the P-wave travel time in three-dimensional space. The method was validated using homogeneous velocity models and inhomogeneous layered media velocity models of different scales. The results showed that the overall mean absolute error (MAE) of the two homogenous models using 3D_H_FMM had been reduced by 88.335%, and 90.593% compared with the traditional 3D_FMM. On that basis, the three-dimensional localization of microseismic sources was carried out using a particle swarm optimization algorithm. The developed 3D_H_FMM was used to calculate the travel time, then to conduct the localization of the microseismic source in inhomogeneous models. The mean error of the localization results of the different positions in the three-dimensional space was determined to be 1.901 m, and the localization accuracy was found to be superior to that of the traditional 3D_FMM method (mean absolute localization error: 3.447 m) with the small-scaled inhomogeneous model.

An Efficient Neural-Network-Based Microseismic Monitoring Platform for Hydraulic Fracture on an Edge Computing Architecture.

Microseismic monitoring is one of the most critical technologies for hydraulic fracturing in oil and gas production. To detect events in an accurate and efficient way, there are two major challenges. One challenge is how to achieve high accuracy due to a poor signal-to-noise ratio (SNR). The other one is concerned with real-time data transmission. Taking these challenges into consideration, an edge-computing-based platform, namely Edge-to-Center LearnReduce, is presented in this work. The platform consists of a data center with many edge components. At the data center, a neural network model combined with convolutional neural network (CNN) and long short-term memory (LSTM) is designed and this model is trained by using previously obtained data. Once the model is fully trained, it is sent to edge components for events detection and data reduction. At each edge component, a probabilistic inference is added to the neural network model to improve its accuracy. Finally, the reduced data is delivered to the data center. Based on experiment results, a high detection accuracy (over 96%) with less transmitted data (about 90%) was achieved by using the proposed approach on a microseismic monitoring system. These results show that the platform can simultaneously improve the accuracy and efficiency of microseismic monitoring.

Comparison of geomechanical deformation induced by megatonne-scale CO2 storage at Sleipner, Weyburn, and In Salah.

Geological storage of CO2 that has been captured at large, point source emitters represents a key potential method for reduction of anthropogenic greenhouse gas emissions. However, this technology will only be viable if it can be guaranteed that injected CO2 will remain trapped in the subsurface for thousands of years or more. A significant issue for storage security is the geomechanical response of the reservoir. Concerns have been raised that geomechanical deformation induced by CO2 injection will create or reactivate fracture networks in the sealing caprocks, providing a pathway for CO2 leakage. In this paper, we examine three large-scale sites where CO2 is injected at rates of ~1 megatonne/y or more: Sleipner, Weyburn, and In Salah. We compare and contrast the observed geomechanical behavior of each site, with particular focus on the risks to storage security posed by geomechanical deformation. At Sleipner, the large, high-permeability storage aquifer has experienced little pore pressure increase over 15 y of injection, implying little possibility of geomechanical deformation. At Weyburn, 45 y of oil production has depleted pore pressures before increases associated with CO2 injection. The long history of the field has led to complicated, sometimes nonintuitive geomechanical deformation. At In Salah, injection into the water leg of a gas reservoir has increased pore pressures, leading to uplift and substantial microseismic activity. The differences in the geomechanical responses of these sites emphasize the need for systematic geomechanical appraisal before injection in any potential storage site.

Rock pressure evaluation in coal face based on multi-factor decision-making theory.

The behavior of rock pressure is a natural and inevitable phenomenon during coal seam mining, resulting in numerous casualties and equipment damage annually. The ability to predict and assess the strength of rock pressure in the coal face beforehand has become crucial in preventing rock pressure accidents. This paper took the prediction of rock pressure strength in coal face as the research object, and based on the multi-factor decision-making theory, proposed a new method for the evaluation of rock pressure strength in coal face-"dual-dimension rock pressure strength evaluation method". Initially, the rock pressure strength index IA was obtained through the application of the law of sedimentary pressure control and microseismic monitoring data. The drilling data at the exploration scale served as references. Then, based on the rock pressure control mechanism, the rock pressure strength index IB was obtained by utilizing a type of Euclidean distance formula at the coal face scale. Finally, in order to mutually correct the two rock pressure strength indices, the rock pressure strength grade matrix was employed to acquire the rock pressure strength grade of the coal face. Applying this evaluation method to the coal face, the prediction outcomes aligned with the actual situation. Therefore, this method can provide a theoretical reference for the prediction of rock pressure strength and the prevention of rock pressure accidents in alternative areas.

Elastic wave prospecting of water-conducting fractured zones in coal mining.

In order to understand the development law of water-conducting fractures in overlying strata during the mining process of coal seam, an elastic wave exploration method based on key stratum theory is proposed to predict the height of water-conducting fracture zone. Taking Yushen mining area as the background, the development and evolution of fractures and the three-dimensional distribution characteristics of water-conducting fracture zone are studied by combining well-ground microseismic monitoring, high-density three-dimensional seismic exploration, borehole investigation, FLAC3D numerical simulation and similar physical simulation tests. The results indicate that the trial mining face's fracture-to-coal ratio ranges from 25.86 to 30.76, with the maximum fracture-to-coal ratio near the cutting eye at 30.76 and the minimum in the central portion of the trial mining face at 25.86. The primary characteristics of rock mass fracture distribution in the mined area are the development of fractures predominantly along high-angle and even vertical bedding planes. Within the fracture zone, fractures increase from top to bottom, with high-angle fractures developing in the lower section and high-angle and horizontal fractures developing simultaneously in the upper section. The water-conducting fracture zone undergoes a developmental process from inception to development, reaching its maximum height, and eventually stabilizing as coal seam mining progresses, overlying rock subsides, strata separation, and damage formation. The three-dimensional shape of the water-conducting fracture zone in the roof of the Yushen mining area exhibits a morphological pattern where the height of the fracture zone gradually decreases from the cutting eye towards the goaf. It also transitions from high to low along both sides and from the periphery towards the interior of the working face. In the trend and strike directions, it exhibits saddle-like characteristics. By comparing the monitoring results, the rationality of the elastic wave prospecting method for predicting the height of water-conducting fracture zones based on critical layer theory was verified. This research holds significant reference value for coal mining under similar geological conditions, especially in terms of water preservation during mining operations.

Microseismic Precursors of Coal Mine Water Inrush Characterized by Different Waveforms Manifest as Dry to Wet Fracturing.

Microseismic monitoring systems have been widely installed to monitor potential water hazards in limestone of the coal floor. The temporal and spatial distribution of rock fracture-induced microseismic events can be used as early warning indicators of potential water inrush from the coal floor. We established a microseismic monitoring system in the working face of Wangjialing coal mine. Besides traditional fluid-independent rock fracture-induced microseismic waveforms, fluid-dependent hybrid-frequency microseismic waveforms also play important roles in determining the microseismic precursors of water inrush. Hybrid-frequency microseismic waveforms have a sharp P wave and no obvious S wave phase. We infer that the first high-frequency signal is caused by the brittleness of the rock in the floor under the influence of the water pressure. The second low-frequency signal is caused by the water oscillations in the fractures. These hybrid-frequency waveforms represent the development of fracturing. In addition, the lifting height of the complete aquiclude above the confined water is very limited, and the water inrush from the floor is often closely related to these hidden faults. Therefore, the activation signal of hidden faults in the working face of coal mining can be monitored to effectively warn about the water inrush from the coal seam floor caused by faults. By analyzing different microseismic events, the monitoring and early warning of water disaster in the coal mine floor can be improved. This will help in taking measures in advance within the mine to ensure personnel safety and to reduce property losses.