摘要
为了快速准确判别透水水源,为煤矿水害防治提供依据,系统分析了福建省龙永煤田透水水源35个标准水样的水化学特征,确定Ca2+、HCO3^-和SO4^2-为特征离子,建立BP神经网络判别模型,随机选取不同水源类型的9个水样作为检验样本,预判正确率88.89%,采集3个未知样本,验证其预判准确率为100%。
In order to distinguish the water-bursting sources quickly and accurately and to provide a basis for the prevention and control of coal mine water disaster,this paper systematically analyzes the hydrochemical characteristics of 35 standard water samples of the water-bursting sources in Longyong coal field in Fujian.Ca2+、HCO3^-and SO4^2 are selected as the characteristic ions for establishing discriminant model based on BP neural network.Nine water samples of different types of water sources are randomly selected as the test samples,the correct rate of predicted discrimination is 88.89%,and it’s 100%when selecting 3 unknown water samples.
作者
徐秀华
XU Xiuhua(Fujian Chuanzheng Communications College,Fuzhou,350007,China)
出处
《华北科技学院学报》
2019年第4期57-64,共8页
Journal of North China Institute of Science and Technology
关键词
透水水源
水化学特征
BP神经网络
判别模型
water bursting source
hydrochemical characteristics
BP neural network
discriminant model
