摘要
本文以赵家寨矿井为研究对象,采集了奥灰岩水样7个,L1-4灰岩水样6个,L7-8灰岩水样25个,砂岩水样2个,第三系水样1个,第四系地下水样1个,老空水样1个。灰岩水样分析结果表明,不同含水层水化学类型有一定差异,如以L7-8灰岩水HCO3-SO4型水为主。以六大离子作为判别因子,经灰色关联度分析、判别分析以及BP神经网络判别模型3种判别方法,对突水水源判别对比,综合选择出适当的水源判别方法。研究表明,该矿井以六大离子作为判别因子时选择BP神经网络分析法进行预测的结果更加可靠。
In this paper,7 water samples of Ordovician limestone aquifer,6 water samples of L1 -4 limestone aquifer,25 water samples of L7 -8 limestone aquifer and 7 water samples of other aquifers were collected for the study from ZhaojiaZhai Coalmine.The results show that the water types of different aquifers were different in a certain degree,for example,the L7 -8 limestone water type is HCO3 -SO4.Choosing 6 ions as discriminant factors,grey relational grade analysis,discriminant analysis and BP neural network were used to distinguish the water sources.Then,the method of BP neural network was selected out to be the best suitable mathemati-cal discrimination method.
出处
《华北科技学院学报》
2014年第11期52-56,共5页
Journal of North China Institute of Science and Technology
基金
河北省自然科学基金(E2013508123)
中央高校基本科研业务费(AQ2013B01
3142014002)
关键词
充水水源
灰岩水
BP
神经网络
灰色关联度
判别分析
source of water -filled
BP neural network
grey relational grade analysis
discriminant analysis
