基于遗传BP神经网络的矿井突水水源识别  

Water Source Recognition of Mine Inflow Based on the GA-BP Neural Network

作　　者：王欣[1] 葛恒清[1] 张凯婷[1] 陈友群 WANG Xin;GE Heng-qing;ZHANG Kai-ting;CHEN You-qun(School of Physics and Electronic Electrical Engineering, Huaiyin Normal University, Huaian Jiangsu 223300, Chin)

机构地区：[1]淮阴师范学院物理与电子电气工程学院,江苏淮安223300

出　　处：《淮阴师范学院学报（自然科学版）》2017年第4期307-311,共5页Journal of Huaiyin Teachers College;Natural Science Edition

基　　金：江苏省科技厅产学研联合创新项目(BY2016062-01);淮安市产学研协同创新项目(HAC201605)

摘　　要：为了快速有效识别矿井突水水源,消除矿井水害,综合考虑各种水化学离子在水源识别中的重要性,选择Na^++K^+、Ca^(2+)、Mg^(2+)、Cl^-、SO_4^(2-)、HCO_3^-等6种水化学离子作为识别因子,提出了一种基于遗传算法优化BP神经网络的矿井突水水源识别方法.在同样的训练样本和待测样本下,将该方法的识别效果与BP神经网络、距离判别法、Bayes判别法等方法的识别效果进行比较.仿真结果表明该方法收敛速度更快,识别精度更高.In order to recognize the water source of mine inflow quickly and efficiently,and then eliminate the mine water disaster,considering the importance of the water chemistry ion in water source recognition,the Na^＋＋ K^＋,Ca^2＋,Mg^2＋,Cl^-,SO4^2-,HCO3^-were selected as evaluation index. A recognition method for water source of mine inflow of optimized BP neural network based on genetic algorithm（ GA） is presented. Under the same train samples and test samples,the recognition effect of this model is compared with the BP neural network,distance discrimination,Bayes discrimination. The computer simulations have showed that this method has faster convergence speed and higher recognition precision.

关 键 词：矿井突水 水源识别 遗传BP神经网络 

分 类 号：TD745[矿业工程—矿井通风与安全]