含瓦斯煤力学特性及其破坏风险等级预测模型  

Mechanical properties and prediction model of failure risk level of gas-bearing coal

作　　者：杨科[1,2,3] 许日杰 张杰[1,2,3] 池小楼 刘文杰 郑诗章[1,2,3] 刘旺军 YANG Ke;XU Rijie;ZHANG Jie;CHI Xiaolou;LIU Wenjie;ZHENG Shizhangg;LIU Wangjun(State Key Laboratory of Mining Response and Disaster Prevention and Control in Deep Coal Mines,Anhui University of Science&Technology,Huainan,Anhui 232001,China;Key Laboratory of Safe and Effective Coal Ministry of Education,Anhui University of Science&Technology,Huainan,Anhui 232001,China;School of Mining Engineering,Anhui University of Science&Technology,Huainan,Anhui 232001,China;Luling Coal Mine,Huaibei Mining Co Ltd,Suzhou,Anhui 234000,China;Department of Environment,Land and Infrastructure Engineering,Politecnico di Torino,Torino 10129,Italy)

机构地区：[1]安徽理工大学深部煤矿采动响应与灾害防控国家重点实验室,安徽淮南232001 [2]安徽理工大学煤矿安全高效开采省部共建教育部重点实验室,安徽淮南232001 [3]安徽理工大学矿业工程学院,安徽淮南232001 [4]淮北矿业股份有限公司芦岭煤矿,安徽宿州234000 [5]都灵理工大学环境、土地和基础设施工程学院,意大利都灵10129

出　　处：《采矿与安全工程学报》2025年第2期404-417,共14页Journal of Mining & Safety Engineering

基　　金：国家自然科学基金区域创新发展联合基金重点项目(U21A20110);国家重点研发计划项目(2023YFC2907502);安徽省自然科学基金项目(2308085QE149)。

摘　　要：为研究深部煤岩瓦斯动力灾害致灾机理及风险判识方法,以两淮矿区芦岭矿煤样为研究对象,开展了不同初始瓦斯压力下煤样单轴压缩试验,分析了瓦斯压力变化对煤体力学特性、破坏特征及声发射事件的影响规律,建立了基于卷积随机向量函数连接网络(CRVFL)的含瓦斯煤破坏风险等级预测模型。研究结果表明:含瓦斯煤应力-应变曲线峰值应力前存在明显塑性应力降现象,局部应变能得到提前释放,随着瓦斯压力升高,煤样峰值应力与弹性模量均呈指数型函数减小;煤样破坏形态与碎屑分布受瓦斯压力变化影响明显,随着瓦斯压力升高,破坏模式由剪切破坏逐渐转变为张拉破坏,分形维数呈对数型函数增长;含瓦斯煤渐进破坏过程中会产生明显的声发射能量信号激增现象,最大声发射能量幅值随瓦斯压力升高先急剧减小后稳定下降;CRVFL模型能够对含瓦斯煤破坏风险等级进行有效预测,准确率均大于99%,体现了模型的稳定性与数据泛化能力。研究成果为深部高瓦斯煤层煤岩瓦斯动力灾害风险判识提供了理论基础。To investigate the mechanism and risk identification of gas dynamic disasters in deep coalrock mass,uniaxial compression tests were conducted on coal samples from the Luling Coal Mine in the Huaibei and Huainan mining areas under different initial gas pressures,and the influences of gas pressure changes on mechanical properties,failure characteristics,and acoustic emission(AE)events of the coal samples were analyzed.Based on the convolutional random vector functional link(CRVFL)network,a prediction model for the failure risk level of gas-bearing coal was established.The results indicate that:The stress-strain curve of gas-bearing coal exhibits a distinct plastic stress drop before the peak stress,where local strain energy is released prematurely.With the rise of gas pressure,both the peak stress and elastic modulus of the coal samples decrease exponentially.The failure mode and debris distribution of the coal samples are remarkably affected by changes in gas pressure.With the rise of gas pressure,the failure mode gradually shifts from shear failure to tensile failure,and the fractal dimension increases logarithmically.During progressive failure of gas-bearing coal,an obvious surge of AE energy is observed.The maximum amplitude of AE energy initially drops sharply and then stabilizes as gas pressure rises.The CRVFL model can effectively predict the failure risk level of gas-bearing coal with an accuracy rate of greater than 99%,which demonstrates the stability and data generalization ability of the model.The research results provide a theoretical foundation for risk identification of gas dynamic disasters in deep coal seams with high gas contents.

关 键 词：岩石力学 含瓦斯煤 声发射 机器学习 破坏预测 

分 类 号：TD315[矿业工程—矿井建设] TD712