论文部分内容阅读
针对矿井地下水混合度较高的突水水源识别问题,运用多元统计分析原理和混合计算原理,建立焦作矿突水水源识别模型和混合模型.以实际数据作为训练样本,分别对它们进行分析与检验.研究结果表明:Logistic分析能有效建立混合度较低的突水水源识别模型,回估误判率较低;混合模型利用主成分分析分析结果建立四面体,损失较少的信息,可有效地确定地下水的混合比例,且利用示踪元素得到的预测值与实测值的总体误差相对较低.
Aiming at the identification of water inrush water source with high degree of groundwater mixing in mines, the identification model and the hybrid model of water inrush source in Jiaozuo Mine are established by using multivariate statistical analysis principle and hybrid computing principle. The actual data are used as training samples to analyze and test them respectively The results show that Logistic analysis can effectively establish a water inrush recognition model with low mixed degree, and the misjudgment rate is low. In the mixed model, the principal component analysis results can be used to establish tetrahedron with less loss information, which can effectively The mixing ratio of groundwater is determined, and the overall error between the predicted value and the measured value obtained by the tracer element is relatively low.