摘要
为提高煤与瓦斯突出预测的准确性,基于判别分析理论,通过逐步判别法筛选出瓦斯放散初速度、瓦斯压力、软分层厚度3个煤与瓦斯突出敏感指标作为突出判别因子,将煤与瓦斯突出危险性分为4个等级作为Bayes判别分析的4个正态总体,建立了煤与瓦斯突出预测的Bayes-逐步判别分析模型。利用该判别模型对20个煤与瓦斯突出实例进行训练学习得出相应的判别函数,用回代估计的方法进行逐一验证,其误判率为0。将建立的判别模型应用于8个突出实例进行判别预测,其结果与实际情况完全吻合。
Bayes stepwise discriminant model is established to predict the coal and gas outbursts accurately and reliably based on the principle of discriminant analysis theory.Three sensitive indexes of coal and gas outburst such as initial speed of methane diffusion,gas pressure and soft layer thickness are regarded as the discriminant factors of the Bayes discriminant analysis model by stepwise discriminant analysis method.Meanwhile,the risk of coal and gas outburst is divided into four levels as the four normal states of Bayes discriminant analysis.The discriminant function are obtained through training of twenty sets of coal and gas outburst data by using the constructed discriminant model.Each of the twenty sets of samples is tested by using resubstitution method according to the Bayes discriminant method,and the misjudgment rate is equal to zero.The established discriminant model is applied to eight sets of coal and gas outburst examples for discriminant prediction,and the results are in perfect agreement with the actual situation.
作者
李长兴
关金锋
李回贵
辛程鹏
LI Changxing;GUAN Jinfeng;LI Huigui;XIN Chengpeng(College of Mining Engineering,Guizhou University of Engineering Science,Bijie 551700,China;School of Emergency Management and Safety Engineering,China University of Mining and Technology(Beijing),Beijing 100083,China)
出处
《中国矿业》
北大核心
2020年第2期117-123,共7页
China Mining Magazine
基金
贵州省重点学科矿业工程学科项目资助(编号：ZDXK[2016]13)
贵州省科技支撑计划项目资助(编号：黔科合支撑[2017]2820)、[2019]2876)
贵州省青年科技人才资助项目资助(编号：黔教合KY字[2017]296)
贵州省联合基金资助项目资助(编号：黔科合LH字[2015]7596)
