基于定量预测模型及预警系统的沙坪煤矿顶板突水能力分析  

Analysis of Shaping Coal Mine Roof Surge Capacity Based on Quantitative Prediction Model and Early Warning System

作　　者：张敏 ZHANG Min(Shanxi Dibao Energy Co.,Ltd.,Taiyuan,Shanxi 030045,China)

机构地区：[1]山西地宝能源有限公司,山西太原030045

出　　处：《自动化应用》2024年第16期212-214,219,共4页Automation Application

摘　　要：在煤矿开采过程中,面临顶板突水机理复杂和影响因素多变的问题,各因素相互联系的不确定性加大了顶板水害防治的难度。为提高煤矿顶板突水能力(WYC)预测的准确性和可靠性,分析了煤层顶板突水的主控因素,构建了基于SAA-CNN算法的WYC定量预测模型,并预测沙坪煤矿13煤开采顶板的WYC;分别使用BP、RF、SVM、CNN和SAA-CNN模型进行训练和误差对比,选择沙坪煤矿13102工作面,采用多种物探方法与钻孔相结合的方式进行超前探测,构建煤层顶板突水危险性动态监测预警系统。结果表明,WYC预测模型的准确率为SAA-CNN>RF>CNN>SVM>BP,SAA-CNN模型的平均绝对误差为0.87 m^(3)/h,平均相对误差为3.56%。In the process of coal mining,it faces the problem of complex mechanism of roof water breakout and variable influencing factors,and the uncertainty of interconnection of various factors increases the difficulty of roof water damage prevention and control.In order to improve the accuracy and reliability of the prediction of water-yielding capacity(WYC)of coal mine roof,the main controlling factors of coal seam roof water burst were analyzed,and a quantitative prediction model of WYC based on SAA-CNN algorithm was constructed,and the WYC of the roof of 13-coal mining in Shaping coal mine was predicted,and the training and error comparison were carried out by using the BP,RF,SVM,CNN,and SAA-CNN models,respectively,and Shaping coal mine 13102 working face was selected.Coal mine 13102 working face,using a variety of physical exploration methods and drilling holes combined with overdetection,to construct the dynamic monitoring and early warning system for the risk of water breakout on the roof of the coal seam.The results show that the accuracy of WYC prediction model is SAA-CNN>RF>CNN>SVM>BP,and the average absolute error of SAA-CNN model is 0.87 m^(3)/h,and the average relative error is 3.56%.

关 键 词：煤层顶板 突水能力 SAA-CNN算法 联合超前探测 

分 类 号：TD745[矿业工程—矿井通风与安全]