支持向量机在矿井突水水源识别中的应用

Application of Water Source Identification in Mine Inflow Based on Support Vector Machines

为了正确地识别矿井突水水源,基于水化学成分对水源判别的重要性,选择K++Na+、Ca+2、Mg+2、Cl-、SO24-、HCO-3这6项指标作为特征向量,建立了矿井突水水源的支持向量机识别模型.此方法不仅结构简单,而且技术性能尤其泛化能力与BP神经网络相比有明显提高,能有效地识别矿井突水水源的类别,为防治水工作提供决策依据.

In order to identify the water source of mine inflow correctly, the K^＋ ＋Na^＋ Ca2^＋,Mg2^＋ Cl^-, SO4^2- ,HCO3^- are selected as the feature vectors based on the importance of 6 water hydrochemical element factors. Then, the determination model of water source in mine inflow is established based on Support Vector Machines in the paper. The results show that the SVM method is not only simple in structure, but also has markedly improved in technical performance and generalization ability, especially compared with the BP neural network. The method is able to identify the water source of mine inflow effectively based on support vector machines model, and provids basis for making decision on prevention and cure of mine inflow work.