Fisher判别法在矿井突水水源识别中的应用——以钱营孜煤矿为例

Application of Fisher Discriminant Method in the Identification of Inrush Water Source of Coal Mine—A Case Study of Qianyingzi Coal Mine

摘要: 准确识别矿井突水水源对有效地预防矿井突水事故具有积极意义。本文以钱营孜矿为例，在系统分析矿井地质和水文地质条件的基础上，收集31个“四含水”、“砂岩裂隙水”、“灰岩水”共3类突水水源的常规水化学测试数据，利用Fisher判别原理，选取K⁺ + Na⁺，Ca²⁺，Mg²⁺，Cl^－，SO₄^2- ，HCO₃^-，pH值，总硬度，总碱度共9种判别指标(特征因子)，筛选出K⁺ + Na⁺和Mg²⁺为特征因子，建立Fisher水源判别模型，并用回代的方式对判别模型进行判别效果检验，结果表明回判总正确率为100%，判别效果好。该模型可用于矿井突水水源预测。

Abstract: To identify the source of mine water inrush accurately is of positive significance for effectively preventing mine water inrush accidents. Based on systematic analysis of the geological and hydrogeological conditions of the coal mine, the Qianyingzi mine was taken as an example, 31 samples from 3 water inrush sources, including the fourth aquifer, coal bearing sandstone aqui-fer, and the limestone aquifer were collected. Using the Fisher discrimination principle, the Fisher water source discrimination model was established by selecting 9 characteristic indica-tors, including K⁺ + Na⁺, Ca²⁺, Mg²⁺, Cl^－,SO₄^2- , HCO₃^- , pH, total hardness and total alkalinity, and K⁺ + Na⁺ and Mg²⁺ were selected as characteristic factors. The results showed that the total accuracy rate of back-propagation was 100%, indicating good discriminant effect. This model can be used for predicting the source of water inrush in mines.