Predicting Rock Burst in Underground Engineering Leveraging a Novel Metaheuristic-Based LightGBM Model

Rock bursts represent a formidable challenge in underground engineering,posing substantial risks to both infrastructure and human safety.These sudden and violent failures of rock masses are characterized by the rapid release of accumulated stress within the rock,leading to severe seismic events and structural damage.Therefore,the development of reliable prediction models for rock bursts is paramount to mitigating these hazards.This study aims to propose a tree-based model—a Light Gradient Boosting Machine(LightGBM)—to predict the intensity of rock bursts in underground engineering.322 actual rock burst cases are collected to constitute an exhaustive rock burst dataset,which serves to train the LightGBMmodel.Two population-basedmetaheuristic algorithms are used to optimize the hyperparameters of the LightGBM model.Finally,the sensitivity analysis is used to identify the predominant factors that may incur the occurrence of rock bursts.The results show that the population-based metaheuristic algorithms have a good ability to search out the optimal hyperparameters of the LightGBM model.The developed LightGBM model yields promising performance in predicting the intensity of rock bursts,with which accuracy on training and testing sets are 0.972 and 0.944,respectively.The sensitivity analysis discloses that the risk of occurring rock burst is significantly sensitive to three factors:uniaxial compressive strength(σc),stress concentration factor(SCF),and elastic strain energy index(Wet).Moreover,this study clarifies the particular impact of these three factors on the intensity of rock bursts through the partial dependence plot.

GRB 200612A:An Ultralong Gamma-Ray Burst Powered by Magnetar Spinning Down

GRB 200612A could be classified as an ultralong gamma-ray burst due to its prompt emission lasting up to~1020 s and the true timescale of the central engine activity t_(burst)≥4×10^(4) s.The late X-ray light curve with a decay index ofα=7.53 is steeper than the steepest possible decay from an external shock model.We propose that this X-ray afterglow can be driven by dipolar radiation from the magnetar spindown during its early stage,while the magnetar collapsed into the black hole before its spindown,resulting in a very steep decay of the late X-ray light curve.The optical data show that the light curve is still rising after 1.1 ks,suggesting a late onset.We show that GRB 200612A’s optical afterglow light curve is fitted with the forward shock model by Gaussian structured off-axis jet.This is a special case among GRBs,as it may be an ultralong gamma-ray burst powered by a magnetar in an off-axis observation scenario.

基于IMU-LiDAR紧耦合的煤矿防冲钻孔机器人定位导航方法

防冲钻孔机器人是冲击地压矿井卸压的关键设备,其在复杂卸压巷道的精确地图构建和的稳定导航是实现钻孔作业智能化的基础和前提。在分析激光雷达点云畸变成因和同步定位与地图构建(SLAM)算法缺陷的基础上,设计了基于惯性测量单元(IMU)连续时间轨迹的点云畸变矫正方法,建立了激光雷达和IMU的数据融合模型,提出了基于IMU-LiDAR紧耦合的防冲钻孔机器人定位建图方法。根据煤矿卸压巷道特点建立了密闭坡道模型,开展了建图效果仿真分析,结果表明,所提算法在定位精度、轨迹误差方面均优于现有常用算法。在此基础上,设计了基于改进人工势场法和快速扩展随机树的动态路径规划方法,建立了适用于防冲钻孔机器人的路径规划与导航融合方案,并设计了2种仿真运动场景,结果表明,所提路径规划方法在全局路径规划和动态路径规划的平均路径长度、平均运行时间、平均生成节点数等方面均具有较好的综合性能。为了进一步验证防冲钻孔机器人定位导航方法的实用性,在校内模拟巷道、地面实验基地和井下卸压巷道等场景下开展了多组对比实验,结果表明:将IMU数据与LiDAR数据紧耦合后,所提方法的定位建图精度明显提高,在特征退化场景中具有优越的定位建图性能,且规划路径的运算效率和路径代价方面均具有良好的表现,验证了所提定位导航方法在多种场景中的可行性和优越性。

Rock burst mechanism analysis in an advanced segment of gob-side entry under different dip angles of the seam and prevention technology

In order to investigate the frequent occurrences of rock burst in gob-side entry during the mining process of the mining zone No. 7, the mechanical model of main roof of fully-mechanized caving mining before breaking was established by the Winkler foundation beam theory, and the stress evolution law of surrounding rock with different dip angles of the seam during the mining process was analyzed by using FLAC3 D. The results show that: with the dip angle changing from 45° to 0°, the solid-coal side of gobside entry begins to form an L-shaped stress concentration zone at a dip angle of 30°, and the stress concentration degree goes to higher and higher levels. However, the stress concentration degree of the coalpillar side goes to lower and lower levels; the influence range and peak stress of the abutment at the lateral strata of adjacent gob increase with dip angle decreasing and reach a maximum value at a dip angle of 0°, but the tailgate is not affected; the abutment pressure superposition of two adjacent gobs leads to stress concentration further enhancing in both sides of gob-side entry. With the influence of strong mining disturbance, rock burst is easily induced by dynamic and static combined load in the advanced segment of gob-side entry. To achieve stability control similar to that in the roadway, the key control strategy is to reinforce surrounding rock and unload both sides. Accordingly, the large-diameter drilling and high-pressure water injection combined unloading and reinforced support cooperative control technology was proposed and applied in field test. The results of Electromagnetic Emission(EME) and field observation showed that unloading and surrounding rock control effect was obvious.

Comprehensive early warning of rock burst utilizing microseismic multi-parameter indices

Rock bursts have become one of the most severe risks in underground coal mining and its early warning is an important component in the safety management. Microseismic(MS) monitoring is considered potentially as a powerful tool for the early warning of rock burst. In this study, an MS multi-parameter index system was established and the critical values of each index were estimated based on the normalized multi-information warning model of coal-rock dynamic failure. This index system includes bursting strain energy(BSE) index, time-space-magnitude independent information(TSMII) indices and timespace-magnitude compound information(TSMCI) indices. On the basis of this multi-parameter index system, a comprehensive analysis was conducted via introducing the R-value scoring method to calculate the weights of each index. To calibrate the multi-parameter index system and the associated comprehensive analysis, the weights of each index were first confirmed using historical MS data occurred in LW402102 of Hujiahe Coal Mine(China) over a period of four months. This calibrated comprehensive analysis of MS multi-parameter index system was then applied to pre-warn the occurrence of a subsequent rock burst incident in LW 402103. The results demonstrate that this multi-parameter index system combined with the comprehensive analysis are capable of quantitatively pre-warning rock burst risk.

Application of high-pressure water jet technology and the theory of rock burst control in roadway

This paper puts forward using high-pressure water jet technology to control rock burst in roadway, and analyzes the theory of controlling rock burst in roadway by the weak structure zone model. The weak structure zone is formed by using high-pressure water jet to cut the coal wall in a continuous and rotational way. In order to study the influence law of weak structure zone in surrounding rock, this paper numerically analyzed the influence law of weak structure zone, and the disturbance law of coal wall and floor under dynamic and static combined load. The results show that when the distance between high-pressure water jet drillings is 3 m and the diameter of drilling is 300 mm, continuous stress superposition zone can be formed. The weak structure zone can transfer and reduce the concentrated static load in surrounding rock, and then form distressed zone. The longer the high-pressure water jet drilling is, the larger the distressed zone is. The stress change and displacement change of non-distressed zone in coal wall and floor are significantly greater than that of distressed zone under dynamic and static combined load. And it shows that the distressed zone can effectively control rock burst in roadway under dynamic and static combined load. High-pressure water jet technology was applied in the haulage gate of 250203 working face in Yanbei Coal Mine, and had gained good effect. The study conclusions provide theoretical foundation and a new guidance for controlling rock burst in roadway.

Quantitative calculation for the dissipated energy of fault rock burst based on gradient-dependent plasticity

The capacity of energy absorption by fault bands after rock burst wascalculated quantitatively according to shear stress-shear deformation curves considering theinteractions and interplaying among microstructures due to the heterogeneity of strain softeningrock materials. The post-peak stiffness of rock specimens subjected to direct shear was derivedstrictly based on gradient-dependent plasticity, which can not be obtained from the classicalelastoplastic theory. Analytical solutions for the dissipated energy of rock burst were proposedwhether the slope of the post-peak shear stress-shear deformation curve is positive or not. Theanalytical solutions show that shear stress level, confining pressure, shear strength, brittleness,strain rate and heterogeneity of rock materials have important influence on the dissipated energy.The larger value of the dissipated energy means that the capacity of energy dissipation in the formof shear bands is superior and a lower magnitude of rock burst is expected under the condition ofthe same work done by external shear force. The possibility of rock burst is reduced for a lowersoftening modulus or a larger thickness of shear bands.

Numerical simulation study on hard-thick roof inducing rock burst in coal mine

In order to reveal the dynamic process of hard-thick roof inducing rock burst, one of the most common and strongest dynamic disasters in coal mine, the numerical simulation is conducted to study the dynamic loading effect of roof vibration on roadway surrounding rocks as well as the impact on stability. The results show that, on one hand, hard-thick roof will result in high stress concentration on mining surrounding rocks; on the other hand, the breaking of hard-thick roof will lead to mining seismicity, causing dynamic loading effect on coal and rock mass. High stress concentration and dynamic loading combination reaches to the mechanical conditions for the occurrence of rock burst, which will induce rock burst. The mining induced seismic events occurring in the roof breaking act on the mining surrounding rocks in the form of stress wave. The stress wave then has a reflection on the free surface of roadway and the tensile stress will be generated around the free surface. Horizontal vibration of roadway surrounding particles will cause instant changes of horizontal stress of roadway surrounding rocks; the horizontal displacement is directly related to the horizontal stress but is not significantly correlated with the vertical stress; the increase of horizontal stress of roadway near surface surrounding rocks and the release of elastic deformation energy of deep surrounding coal and rock mass are immanent causes that lead to the impact instability of roadway surrounding rocks. The most significant measures for rock burst prevention are controlling of horizontal stress and vibration strength.

Frequency spectrum analysis on micro-seismic signal of rock bursts induced by dynamic disturbance

Blasting and breaking of hard roof are main inducing causes of rock bursts in coal mines with danger of rock burst,and it is important to find out the frequency spectrum distribution laws of these dynamic stress waves and rock burst waves for researching the mechanism of rock burst.In this paper,Fourier transform as a micro-seismic signal conversion method of amplitude-time character to amplitude-frequency character is used to analyze the frequency spectrum characters of micro-seismic signal of blasting,hard roof breaking and rock bursts induced by the dynamic disturbance in order to find out the difference and relativity of different signals.The results indicate that blasting and breaking of hard roof are high frequency signals,and the peak values of dominant frequency of the signals are single.However,the results indicate that the rock bursts induced by the dynamic disturbance are low frequency signals,and there are two obvious peak values in the amplitude-frequency curve witch shows that the signals of rock bursts are superposition of low frequency signals and high frequency signals.The research conclusions prove that dynamic disturbance is necessary condition for rock bursts,and the conclusions provide a new way to research the mechanism of rock bursts.

Prediction of rock-burst-threatened areas in an island coal face and its prevention:A case study

The island coal face arises in coal mines with the purpose of preventing gas explosion or maintaining the balance between mining and tunneling. However, its particular stress conditions in the surrounding rock may increase the difficulty of stress control in the coal face and in its mining roadways, especially when the coal seam, the roof, and the floor have rock-burst propensities. The high energy accumulated in the island coal face and in its roof and floor will intensify rock-burst propensity or even induce rock burst,which further result in great casualties and financial losses. Taking island coal face 2321 in Jinqiao coal mine as a case, we propose a method for the prediction of rock-burst-threatened areas in an island coal face with weak rock-burst propensity. Based on the analysis of the movement of the overlying roof and characteristics of stress distribution, this method combined numerical simulation with drilling bits to ensure the prediction accuracy. The effects of coal pillars with different widths on the mitigation of stress concentration in the coal face and on the prevention of rock burst are analyzed together with the mechanism behind. Finally, corresponding measures against the rock burst in the island coal face are proposed.

Gas-dynamic phenomena caused by rock mass tremors and rock bursts

Similar to coal, rock and gas ejections, rock mass tremors and rock bursts are among the most serious natural hazards accompanying the underground extraction of coal. Gas-dynamic phenomena caused by rock mass tremors and rock bursts observed as transient states of air parameters in mining headings,are usually generated as a result of a change in the geometry of headings and the release of considerable amounts of gases. Particular significance is attributed to transient states caused by disasters, which are often accompanied by rapid incidents, presenting threats to the life and health of the underground crew.In Polish mining there are known examples of transient states of air parameters recorded during gasdynamic phenomena, e.g. tremors and rock bursts. The paper presents the case studies of rapid seismic incidents to show how records in mine monitoring systems broaden the knowledge about the transient states of air parameters in mining headings generated because of them.

基于SSA-BP神经网络的岩爆烈度等级预测

随着深部开采战略在我国的发展,岩爆愈加成为我国资源开采时必须面对的地质灾害之一。为提高传统误差反向传播(Back Propagation,BP)神经网络模型进行岩爆预测的准确性与有效性,采用麻雀搜索算法(Sparrow Search Algorithm,SSA)优化传统BP神经网络,提出一种基于麻雀搜索算法优化BP神经网络的岩爆预测模型(SSA-BP模型)。在考虑岩爆产生的内外因基础上,选取相关岩爆预测指标,利用国内外100例已有工程岩爆数据建立SSA-BP模型,并与传统BP模型、粒子群算法(Particle Swarm Optimization,PSO)优化支持向量机(Support Vector Machines,SVM)模型对比。结果表明:SSA-BP预测模型的有效性和准确度皆高于传统BP模型和PSO-SVM模型,同时SSA-BP模型训练集的均方误差(Mean Square Error,MSE)为0.081,比传统BP模型(0.25)降低67.7%,可为类似工程的岩爆预测提供科学依据。

3-D modeling of rock burst in pillar No. 19 of Fetr6 chromite mine

Fetr6 is an underground mine in which chromite is extracted using stope and pillar mining method. Despite of all improving works such as roof supporting and replacing of ore pillars with concrete pillars, pillar No. 19 failed and other pillars failed progressively as a domino effect and 4000 m2 of mine collapsed within a few minutes, consequently. For detail investigation, two 3-D numerical models were developed by 3Dec. The first, a base model, was used for estimation of stress on pillars just before failure and the other for investigation of rock burst in pillar No. 19. The results show that discontinuity parameters such as friction angle and shear stiffness is critical parameters in this pillar failure. In addition, it indicates that W/H ratio equal 0.3, the lack of ore extraction strategy and inadequate roof support are the major reasons for this failure. In this paper, the procedure of study was described.

深部综采工作面多关键层-煤柱系统失稳诱冲机理研究

深部综采工作面多关键层-煤柱系统失稳型冲击地压是鄂尔多斯深部矿区常见的冲击地压类型,目前现场没有统一成熟的治理方案,冲击显现仍时有发生,以鄂尔多斯深部矿区门克庆矿3102工作面为工程背景,采用理论分析、数值模拟、现场实测等方法,对深部综采工作面多关键层-煤柱系统失稳诱冲机理进行研究,探讨了深部多关键层-煤柱系统的空间结构演化规律,研究了区段煤柱承载多关键层加载模式,分析了多关键层-宽煤柱系统的变形与应力转化函数关系,推导了多关键层-宽煤柱系统的失稳冲击工程判据。揭示了深部多关键层-煤柱系统失稳诱冲机理:当煤柱两侧采空区后,低位“ILZ”和面内“DLZ”破断关键岩层组加载至宽煤柱使其压缩变形,之后采场“DLZ”关键岩层组挠曲甚至破断,转移应力加载区段煤柱上,当其承载应力超过失稳阈值时即发生整体失稳冲击。数值模拟研究了不同缩面宽度条件下的采场区段煤柱应力大小与分布特征变化规律,随着区段承载煤柱宽度的增加,应力分布特征由平台型过渡至马鞍形,应力集中程度和整体冲击危险水平持续降低,现场实测验证了120 m区段煤柱的防冲安全性。

基于ISABO-SVM的冲击地压危险等级预测

为有效提高冲击地压危险等级预测的准确率,建立了基于改进的减法平均算法(ISABO)优化支持向量机(SVM)的冲击地压危险等级预测模型。通过提取声发射信号的上升时间和绝对能量等5个时域特征参数,结合煤岩体抗压强度,构建了冲击地压预测的特征参数数据库;根据特征参数数据库计算每个样本的综合危险指数,将冲击地压危险性划分为四个等级。以特征参数数据库和危险等级为模型输入和输出,通过ISABO优化SVM预测冲击地压危险等级过程中的核参数和惩罚因子。结果表明,与传统SVM模型以及减法平均算法(SABO)优化后的SVM模型相比,建立的ISABO-SVM模型预测准确率提高至98%。

Numerical analysis on the factors affecting post-peak characteristics of coal under uniaxial compression

The post-peak characteristics of coal serve as a direct reflection of its failure process and are essential parameters for evaluating brittleness and bursting liability.Understanding the significant factors that influence post-peak characteristics can offer valuable insights for the prevention of coal bursts.In this study,the Synthetic Rock Mass method is employed to establish a numerical model,and the factors affecting coal post-peak characteristics are analyzed from four perspectives:coal matrix mechanical parameters,structural weak surface properties,height-to-width ratio,and loading rate.The research identifies four significant influencing factors:deformation modulus,density of discrete fracture networks,height-to-width ratio,and loading rate.The response and sensitivity of post-peak characteristics to single-factor and multi-factor interactions are assessed.The result suggested that feasible prevention and control measures for coal bursts can be formulated through four approaches:weakening the mechanical properties of coal pillars,increasing the number of structural weak surfaces in coal pillars,reducing the width of coal pillars,and optimizing mining and excavation speed.The efficacy of measures aimed at weakening the mechanical properties of coal is successfully demonstrated through a case study on coal burst prevention using large-diameter borehole drilling.

Rock burst laws in deep mines based on combined model of membership function and dominance-based rough set

Rock bursts are spontaneous, violent fracture of rock that can occur in deep mines, and the likelihood of rock bursts occurring increases as depth of the mine increases. Rock bursts are also affected by the compressive strength, tensile strength, tangential strength, elastic energy index, etc. of rock, and the relationship between these factors and rock bursts in deep mines is difficult to analyze from quantitative point. Typical rock burst instances as a sample set were collected, and membership function was introduced to process the discrete values of these factors with the discrete factors as condition attributes and rock burst situations as decision attributes. Dominance-based rough set theory was used to generate preference rules of rock burst, and eventually rock burst laws analysis in deep mines with preference relation was taken. The results show that this model for rock burst laws analysis in deep mines is more reasonable and feasible, and the prediction results are more scientific.

基于ST-DBSCAN聚类算法的矿井冲击地压微震监测数据时空特性分析

矿井冲击地压是地下矿山中的一种常见地质灾害,对矿山的安全生产和员工的人身安全造成威胁。矿井常规的数据分析多采用数理统计方法进行危险预警分析,具有一定的片面性和局限性,对高维的微震监测数据时空特性分析明显不足,数据分析缺乏实时性,预警分析对特定软件的依赖性过大。为了提高冲击地压的预警能力,提出了一种基于ST-DBSCAN时空聚类算法的方法来分析矿井冲击地压微震监测数据的时空特性,相对于传统聚类算法,该方法表现出更高的鲁棒性。通过矿井真实微震数据验证分析,证明了ST-DBSCAN时空聚类算法能够满足冲击地压微震监测数据时、空、强分析要求,可以识别和分类工作面回采过程中的应力增高区,能够为矿井冲击地压的预警和管理提供决策支持。

Damage Model of Brittle Coal-Rock and Damage Energy Index of Rock Burst

Based on the mechanical experiment of brittle coal-rock and the damage mechanical theory, a damage model was established. Coal-Rock damage mechanical characteristic was researched. Furthermore, interior energy transformation mechanism of rock was analyzed from the point of view of damage mechanics and damage energy release rate of brittle coal rock was derived. By analyzing the energy transformation of rock burst, a new conception, damage energy index of rock burst, was put forward. The condition of rock burst was also established.

L-3-正丁苯酞对实验性自身免疫性脑脊髓炎的作用及其机制

目的 基于RhoA/ROCK信号通路探讨L-3-正丁苯酞(L-3-n-butylphthalide, NBP)对实验性自身免疫性脑脊髓炎(experimental autoimmune encephalomyelitis, EAE)小鼠的疗效及潜在机制。方法 32只雌性C57BL/6小鼠随机分为4组:对照组、EAE组、NBP低剂量组(L-NBP)和NBP高剂量组(H-NBP),每组8只。使用髓鞘少突胶质细胞糖蛋白(myelin oligodendrocyte glycoprotein 35-55, MOG35-55)作为抗原乳剂免疫诱导C57BL/6小鼠建立EAE模型。EAE模型制作成功后,L-NBP组和H-NBP组分别以3.25,6.5 mg/(kg·d)NBP腹腔注射,连续28 d。免疫当日起,每7 d记录各组小鼠的体质量,每天进行神经功能障碍评分;对脊髓组织进行HE染色和劳克坚牢蓝(Luxol fast blue, LFB)染色,观察病理改变;使用酶联免疫吸附试验法(ELISA)检测外周血白细胞介素-10(IL-10)、IL-1β、IL-6和IL-18含量;real time-PCR检测脊髓组织炎症相关因子肿瘤坏死因子(TNF)-α、TNF受体1(TNFR1)、IL-1β的表达;Western blot检测RhoA、ROCKⅠ及ROCKⅡ蛋白表达情况。结果 与对照组相比,EAE组小鼠体质量下降,神经功能障碍评分、组织病理学评分升高,IL-10含量降低,IL-18、IL-6和IL-1β含量增加,TNF-α、TNFR1及IL-1β mRNA表达升高,RhoA、ROCKⅠ及ROCKⅡ蛋白表达增高,差异均有统计学意义(P<0.05)。与EAE组相比,L-NBP和H-NBP组小鼠体质量增加,发病潜伏期延长,高峰期延迟,神经功能障碍评分、组织病理学评分降低,IL-10含量增高,IL-1β、IL-6和IL-18含量降低,TNF-α、TNFR1及IL-1β mRNA表达降低,RhoA、ROCKⅠ和ROCKⅡ蛋白表达降低,差异均有统计学意义(P<0.05)。与L-NBP组相比,H-NBP组小鼠体质量增加,发病潜伏期延长,高峰期延迟,神经功能障碍评分、组织病理学评分降低,IL-10含量增高,IL-1β、IL-6和IL-18含量降低,TNF-α、TNFR1和IL-1β mRNA表达降低,RhoA和ROCKⅡ蛋白表达降低,差异均有统计学意义(P<0.05)。结论 NBP能够减轻EAE小鼠外周及中枢炎症反应,进而改善EAE的神经障碍症状,其作用机制可能是通过抑制RhoA/ROCK信号通路来实现的。