基于SVM-MFOA的煤矿采掘工作面瓦斯涌出量预测方法  

Prediction Method of Gas Emission in Coal Mining Face Based on SVM-MFOA

作　　者：苌延辉 Chang Yanhui(Shanxi Tiandi Wangpo Coal Industry Co.,Ltd.,Jincheng 048000,China)

机构地区：[1]山西天地王坡煤业有限公司,山西晋城048000

出　　处：《煤矿机械》2023年第7期183-186,共4页Coal Mine Machinery

基　　金：山西省科技攻关项目(20200512021-1);2021年山西天地王坡煤业有限公司项目(CG502106014)。

摘　　要：煤矿瓦斯涌出量受流量、风速、抽采负压、浓度、温度等多个因素的影响,预测过程精度较低。提出基于支持向量机与改进果蝇优化算法(SVM-MFOA)的煤矿采掘工作面瓦斯涌出量预测方法。以采掘煤层瓦斯量、瓦斯压力、位置深度、煤层厚度、每日开采工作量、采掘煤炭日产量作为训练集样本,将SVM与MFOA相融合;MFOA经过循环迭代计算后,完成瓦斯高值区的全局寻优,在SVM中引入最优参数,实现从低维特征空间向高维特征空间的映射,完成瓦斯涌出量预测。实验结果表明,该方法的预测精度高达97%,预测结果与真实值非常接近,可提高煤矿采掘工作安全性。The gas emission of coal mine is affected by many factors,such as flow,wind speed,negative pressure of extraction,concentration,temperature,etc.The prediction process has low accuracy.A prediction method of gas emission in coal mining face based on support vector machine and modified fruit fly optimization algorithm(SVM-MFOA)was proposed.The training set samples are the amount of gas in the coal seam mined,the gas pressure,the depth,the thickness of the coal seam,the daily mining workload,and the daily output of the coal mined,and SVM is combined with the MFOA.The MFOA completes the global optimization of the high gas value area after cyclic iterative calculation,introduces the optimal parameters into the SVM,realizes the mapping from the low dimensional feature space to the high dimensional feature space,and completes the prediction of gas emission.The experimental results show that the prediction accuracy of this method is as high as 97%,and the prediction results are very close to the real values,which can improve the safety of mining in coal mines.

关 键 词：采动影响 瓦斯涌出量预测 全局寻优 最优参数 高维特征空间 

分 类 号：TD712.5[矿业工程—矿井通风与安全]