论文部分内容阅读
为有效地预防矿井突水事故,及早识别突水水源是关键工作之一。根据矿井各含水层水化学成分的差异性,选取7种水化学成分指标作为突水水源识别的样本变量。在此基础上,采用主成分分析(PCA)与Fisher判别分析相结合的方法建立突水水源判别模型。以新庄孜煤矿不同水层的水化学特征资料中的33个为学习样本,12个为预测样本,对该模型进行检验和应用,并与传统Fisher判别分析模型的结果进行比较。研究结果表明:利用PCA与Fisher突水水源判别模型能够有效地消除样本变量指标间的相互影响,使突水水源判别结果更加准确。
In order to effectively prevent mine water inrush accidents, it is one of the key tasks to identify water inrush as early as possible. According to the difference of chemical composition of aquifers in the mine, seven kinds of chemical composition indexes were selected as the sample variables for water inrush recognition. Based on this, a discriminant model of inrush water was established by combining principal component analysis (PCA) with Fisher discriminant analysis. Taking 33 samples of hydrochemical characteristics of different water layers in Xinzhuangzi coal mine as samples and 12 samples as prediction samples, the model was tested and applied and compared with the traditional Fisher discriminant analysis model. The results show that the use of PCA and Fisher inrush water source discriminant model can effectively eliminate the interaction between the sample variables indicators, so that inrush water source discrimination results more accurate.