摘要
导水裂隙带高度预测是煤矿水害防治工作中的核心内容之一。本文在分析影响导水裂隙带高度因素基础上,结合灰色关联度理论,选取煤层开采方式、覆岩结构、采高、埋深及工作面斜长5个影响因素进行关联度分析,构建了基于粒子群优化算法(PSO)——支持向量机回归(SVR)的导高预测模型,并将此模型应用于招贤煤矿1307、1304、1305三个工作面导高的预测。结果显示:通过对PSO-SVR导高预测模型与传统经验公式、基于相似条件矿井实测导高曲线拟合方法的预测效果进行综合对比,PSO-SVR导高预测模型预测结果与实测值最为接近,精度较高且准确率更高。研究可以为招贤煤矿及黄陇煤田的矿井水害防治提供理论和技术支撑。
Prediction of the height of water flowing fractured zone is one of the core contents of coal mine water hazard prevention and control.In this paper,based on the analysis of factors affecting the height of water flowing fractured zone,and combined with the theory of Grey relevancy,five influencing factors including coal mining method,overburden structure,mining height,buried depth and face length were for analysis of relevancy degree.Afterwards,a prediction model of the height of water flowing fractured zone was established based on Particle Swarm Optimization(PSO)and Support Vector Regression(SVR).This model is applied to the prediction of the height of water flowing fractured zone of three working faces 1307,1304 and 1305 in Zhaoxian coal mine.The results show that the prediction results of PSO-SVR prediction model are closest to the measured values.The accuracy is higher by comprehensively comparing the prediction results of PSO-SVR prediction model with traditional empirical formula and measured curve fitting method.The research method of this paper can provide theoretical and technical support for mine water hazard prevention and control in Zhaoxian coal mine and Huanglong coal field.
作者
乔伟
韩昌民
李连刚
单景新
周宇
QIAO Wei;HAN Chang-min;LI Lian-gang;SHAN Jing-xin;ZHOU Yu(China School of Resources and Geosciences,China University of Mining and Technology,Xuzhou 221116,China;National Professional Center Laboratory of Basic Research on Mine Water Disaster Prevention and Control Technology,Xuzhou 221116,China;Shaanxi Jinyuan Zhaoxian Mining Co.,Ltd.,Wanbei Coal Power Group Co.,Ltd.,Baoji 721500,China)
出处
《煤炭科技》
2022年第4期77-84,共8页
Coal Science & Technology Magazine
基金
国家自然科学基金项目(41772302)。
关键词
煤矿顶板水害
导水裂隙带
支持向量机
粒子群算法
预测模型
roof water disaster
water flowing fractured zone
support vector regression
particle swarm optimization
prediction model
