煤及瓦斯突出预测的距离判别分析法及应用_王超.pdfVIP

26 4 Vol.26 No.4 　2009 12 Journal of Mining Safety Engineering Dec.2009 　 :1673-3363(2009)04-0470-05 王　超, 宋大钊, 杜学胜, 张振国, 朱　栋, 杨道伟 ( , , , 　221008) :将距离判别分析法应用在煤与瓦斯突出预测中,建立煤与瓦斯突出预测的距离判别分析 模型.选取瓦斯放散初速 、煤的坚固性系数、瓦斯压力、煤体破坏类型和开采深 等5 项影响突 出的主要指标作为距离判别分析模型的判别因子,以我国典型突出矿井的煤与瓦斯突出实例作 为学习样本进行训练,建立相应判别函数, 经过训练后的模型回判估计的误差率为零.利用该模 型对云南恩洪煤矿突出实例进行预测,结果与实际情况完全符合.研究结果表明, 距离判别分析 模型识别性能良好, 预测精 高,在研究煤与瓦斯突出预测中具有良好的实用性和有效性. :距离判别分析;煤与瓦斯突出;预测 :TD 713.2　　:A Prediction of Coal and Gas Outburst Based on Distance Discriminant Analysis Method and Its Application WANG Chao, SONG Da-zhao, DU Xue-sheng, ZHANG Zhen-guo, ZH U Dong, YANG Dao-wei (School of Safety Engineering, State ey Laboratory of Coal Resources and Mine Safety, National Engineering Research Center for Coal Gas Control, China University of Mining Technology, Xuzhou, Jiangsu 221008, China) Abstract:Based on the principle of Mahalanobis distance discriminant analysis (DDA ), a dis- tance discriminant analysis model to predict the possibility and type of coal and gas outburst was established.Five effect indexes including initial speed of methane diffusion, consistent co- efficient of coal, gas pressure, destructive style of coal, and mining depth were regarded as dis- criminant factors of the model.Discriminant functions were obtained through training a large set of studying samples from the instance of typical coal and gas outburst mines;and the ratio of mistake-distinguish is considerably low zero after the DDA model was trained.Coal and gas burst instances of Yunnan Enhong mine were used to verify this model, showing that the DDA model has an excellent performance and a high prediction accuracy . Key words:distance discriminant analysis;coal and gas outbur