论文部分内容阅读
为准确判定煤矿矿井突水水源,基于多元混合模型理论,在水化学分析的基础上选取含水层和突水水样化学离子,建立矿井突水的多元混合模型。分析计算3个不同煤矿的矿井突水水样,并与BP神经网络方法和模糊综合评判法这2种传统突水水源分析方法相对比。结果表明:多元混合模型理论、BP神经网络方法和模糊综合评判方法均能够结合水化学数据对矿井突水水源进行有效判别,其中多元混合模型理论不仅简洁易懂,而且准确率高于传统的BP神经网络方法和模糊综合评判法。
In order to accurately determine water inrush from coal mines, a multivariate mixed model based on multivariate mixed model was proposed to select mulit-mixed models of water inrush in coal mines based on water chemistry analysis. The water inrush of mine in three different coal mines was analyzed and calculated, and compared with the two methods of BP neural network and fuzzy comprehensive evaluation. The results show that multivariate mixed model theory, BP neural network method and fuzzy comprehensive evaluation method can effectively discriminate water inrush water sources in combination with hydrochemical data. The multivariate mixed model theory is not only simple and easy to understand, but also has a higher accuracy than the traditional BP Neural Network Method and Fuzzy Comprehensive Evaluation Method.