Application of Machine Learning Methods in Arrival Time Picking of P Waves from Reservoir Earthquakes

Reservoir earthquake characteristics such as small magnitude and large quantity may result in low monitoring efficiency when using traditional methods. However, methods based on deep learning can discriminate the seismic phases of small earthquakes in a reservoir and ensure rapid processing of arrival time picking. The present study establishes a deep learning network model combining a convolutional neural network (CNN) and recurrent neural network (RNN). The neural network training uses the waveforms of 60 000 small earthquakes within a magnitude range of 0.8-1.2 recorded by 73 stations near the Dagangshan Reservoir in Sichuan Province as well as the data of the manually picked P-wave arrival time. The neural network automatically picks the P-wave arrival time, providing a strong constraint for small earthquake positioning. The model is shown to achieve an accuracy rate of 90.7% in picking P waves of microseisms in the reservoir area, with a recall rate reaching 92.6% and an error rate lower than 2%. The results indicate that the relevant network structure has high accuracy for picking the P-wave arrival times of small earthquakes, thus providing new technical measures for subsequent microseismic monitoring in the reservoir area.     

Measuring changes in fracture properties from temporal variations in anisotropic attenuation of microseismic waveforms

We investigate fracture‐induced attenuation anisotropy in a cluster of events from a microseismic dataset acquired during hydraulic fracture stimulation. The dataset contains 888 events of magnitude ?3.0 to 0.0. We use a log‐spectral‐amplitude‐ratio method to estimate change in over a half‐hour time period where fluid is being injected and an increase in fracturing from S‐wave splitting analysis has been previously inferred. A Pearson's correlation analysis is used to assess whether or not changes in attenuation with time are statistically significant. P‐waves show no systematic change in during this time. In contrast, S‐waves polarised perpendicular to the fractures show a clear and statistically significant increase with time, whereas S‐waves polarised parallel to the fractures show a weak negative trend. We also compare between the two S‐waves, finding an increase in with time. A poroelastic rock physics model of fracture‐induced attenuation anisotropy is used to interpret the results. This model suggests that the observed changes in t* are related to an increase in fracture density of up to 0.04. This is much higher than previous estimates of 0.025 ± 0.002 based on S‐wave velocity anisotropy, but there is considerably more scatter in the attenuation measurements. This could be due to the added sensitivity of attenuation measurement to non‐aligned fractures, fracture shape, and fluid properties. Nevertheless, this pilot study shows that attenuation measurements are sensitive to fracture properties such as fracture density and aspect ratio.     

Comparison of migration‐based location and detection methods for microseismic events

Microseismic monitoring in the oil and gas industry commonly uses migration‐based methods to locate very weak microseismic events. The objective of this study is to compare the most popular migration‐based methods on a synthetic dataset that simulates a strike‐slip source mechanism event with a low signal‐to‐noise ratio recorded by surface receivers (vertical components). The results show the significance of accounting for the known source mechanism in the event detection and location procedures. For detection and location without such a correction, the ability to detect weak events is reduced. We show both numerically and theoretically that neglecting the source mechanism by using only absolute values of the amplitudes reduces noise suppression during stacking and, consequently, limits the possibility to retrieve weak microseismic events. On the other hand, even a simple correction to the data polarization used with otherwise ineffective methods can significantly improve detections and locations. A simple stacking of the data with a polarization correction provided clear event detection and location, but even better results were obtained for those data combined with methods that are based on semblance and cross‐correlation.     

Microseismic Monitoring at a Limestone Mine

Improving microseismic monitoring efficiency and accuracy at a mine requires an integrated effort. This article discusses a case study which demonstrates that the monitoring efficiency and accuracy can be drastically improved through optimally using the existing array, efficient techniques for signal processing and noise separation, and the advanced location algorithm which not only offers a robust search scheme, but also features the techniques for efficient data utilization and further error detection and minimization. The study is a collaborated research between Penn State and NIOSH researchers for the better use of microseismic technique for mine safety, ground control and roof fall prediction.     

基于密集台阵研究2019辽宁抚顺M2.4矿震震源参数

分析矿震破裂机制及微震的时空分布能够为矿区灾害评估提供更多的有效信息.本研究基于密集台阵观测对2019年11月12日辽宁抚顺2.4级矿震开展震源参数研究,震源机制解显示地震破裂包含明显的非双力偶分量,表现为体积压缩的塌陷机制,且震源深度较浅,最佳拟合矩心深度为0.6 km.同时,对11月3日—25日记录的连续地震波形开展微震扫描,新检测出324个微震事件(-0.5~2.0级),定位结果显示在M 2.4矿震发生前M>1.0级事件显著增多,且在矿震位置存在近南北向的微震条带分布,微震序列随时间向深部迁移(约1.5 km),暗示存在断层活化迹象.结合震源破裂机制,我们认为此次事件与矿区塌陷破裂密切相关,同时伴随先存断裂的剪切滑动.本研究表明,基于密集台阵观测的地震矩张量反演和微震检测研究,对判定矿震类型及防范矿区灾害具有重要的研究意义.     

微震技术与应用研究综述

微震,是指微小的震动.在不同应用领域微震大小的尺度划分也各不相同.对于这种微小的地震信号在许多领域都有广泛的应用,例如金属矿山、油田、火山和地震预报等.随着近年来信息技术高速发展,微震监测技术被广泛应用到地下工程当中,如油田开发、矿山的安全生产、隧道的施工建设、水库大坝的选址等诸多方面,并逐渐工程化、系统化,形成实时监测微震的可视化系统.此外,利用自然界大量的微震现象,可以通过震源定位、速度和衰减成像等方法研究大断裂、火山活动、地震的发生发展趋势以及震源机制等.本文总结了各类微震技术与方法,并分析各种方法的优点和存在的问题.另外分类整理了微震在各个领域的应用,分析其不同应用领域下的微震时频率域特征和研究方法.最后得出结论微震技术无论是在科学研究还是工程生产中都起着非常重要的作用.     

The first power generation test of hot dry rock resources exploration and production demonstration project in the Gonghe Basin,Qinghai Province,China

Hot dry rock (HDR) is a kind of clean energy with significant potential. Since the 1970s, the United States, Japan, France, Australia, and other countries have attempted to conduct several HDR development research projects to extract thermal energy by breaking through key technologies. However, up to now, the development of HDR is still in the research, development, and demonstration stage. An HDR exploration borehole (with 236 °C at a depth of 3705 m) was drilled into Triassic granite in the Gonghe Basin in northwest China in 2017. Subsequently, China Geological Survey (CGS) launched the HDR resources exploration and production demonstration project in 2019. After three years of efforts, a sequence of significant technological breakthroughs have been made, including the genetic model of deep heat sources, directional drilling and well completion in high-temperature hard rock, large-scale reservoir stimulation, reservoir characterization, and productivity evaluation, reservoir connectivity and flow circulation, efficient thermoelectric conversion, monitoring, and geological risk assessment, etc. Then the whole-process technological system for HDR exploration and production has been preliminarily established accordingly. The first power generation test was completed in November 2021. The results of this project will provide scientific support for HDR development and utilization in the future.©2022 China Geology Editorial Office.     

Method of predicting tremors on the basis of seismicemission registered in exploitation workings

A method of tremor risk estimation on the basis of seismic emission registered in mining exploitation zones is presented in this paper. A considerable part of the emission registered in these conditions, is generated by mechanical devices used in the mine (miner transporters etc.). Therefore, it is not possible to evaluate the risk of mining rock-bursts properly on the basis of interpretation of emission using known theories and models describing rock mass fracturing. In a new approach presented in this paper, the tremor risk is determined on the basis of time variation of a parameter describing the attenuation of seismic vibrations of the rock medium. The attenuation is assessed from the analysis of seismic emission registered in the frequency range from 30 Hz to 1000 Hz. Information about this parameter is encapsulated in the seismic emission regardless of the means of its generation. The method of estimating the temporal variations of attenuation is based on the analysis of signal envelopes of the microseismic emission. Attenuation trends can be interpreted in a straightforward way. As the stress values keep increasing, the following processes take place in the rock mass: compaction (hardening), dilatancy (softening), and then possibly tremors. At the compaction stage attenuation values are lower, while they increase at the dilatancy stage. This mechanism has been proven in practice. Our results can be used for monitoring tremor risk.     

Imaging pre‐existing natural fractures using microseismic data

In hydraulic fracturing treatments, locating not only hydraulic fractures but also any pre‐existing natural fractures and faults in a subsurface reservoir is very important. Hydraulic fractures can be tracked by locating microseismic events, but to identify the locations of natural fractures, an additional technique is required. In this paper, we present a method to image pre‐existing fractures and faults near a borehole with virtual reverse vertical seismic profiling data or virtual single‐well profiling data (limited to seismic reflection data) created from microseismic monitoring using seismic interferometry. The virtual source data contain reflections from natural fractures and faults, and these features can be imaged by applying migration to the virtual source data. However, the imaging zone of fractures in the proposed method is strongly dependent on the geographic extent of the microseismic events and the location and direction of the fracture. To verify our method, we produced virtual reverse vertical seismic profiling and single‐well profiling data from synthetic microseismic data and compared them with data from real sources in the same relative position as the virtual sources. The results show that the reflection travel times from the fractures in the virtual source data agree well with travel times in the real‐source data. By applying pre‐stack depth migration to the virtual source data, images of the natural fractures were obtained with accurate locations. However, the migrated section of the single‐well profiling data with both real and virtual sources contained spurious fracture images on the opposite side of the borehole. In the case of virtual single‐well profiling data, we could produce correct migration images of fractures by adopting directional redatuming for which the occurrence region of microseismic events is divided into several subdivisions, and fractures located only on the opposite side of the borehole are imaged for each subdivision.     

基于时频稀疏性分析法的低信噪比微震事件识别与恢复

微震监测是直观评价压裂过程和压裂效果的有效手段.微震事件识别是微震监测的首要步骤.然而对于低信噪比微震监测数据,常规识别方法很难取得满意效果.基于微震事件在时频域中的稀疏性,本文提出利用Renyi熵值表示微震监测数据的时频稀疏程度,并以时频距离为约束条件,建立以低熵值的道数为判别阈值的目标函数.本文方法能在识别出微震事件的同时,恢复出较为清晰的微震事件.通过数值计算和对实际监测数据的测试,表明该方法对低信噪比的微震监测数据有较好的处理效果.