基于改进的BP神经网络的地下水水质评价研究  

Research on the BP Neural Network Model of Groundwater Quality Evaluation

作　　者：甄祯[1] 何士华[1] 石崇喜[1] 高飞[1] 余兴奎[1] 

机构地区：[1]昆明理工大学电力工程学院,昆明650051

出　　处：《科学技术与工程》2010年第29期7128-7132,共5页Science Technology and Engineering

摘　　要：运用人工神经网络的理论和方法,建立地下水水质评价的BP神经网络模型。根据《地下水质量标准》(GB/T14848—1993),通过训练该模型对昆明盆地实测12个孔隙水井的地下水水质进行评价,并与层次分析法的评价结果进行了比较。对比结果表明,用BP神经网络模型评价地下水水质是可行的,该模型训练速度快、精度高,能如实地评价地下水质情况。By using artificial neural network theory and method of groundwater quality evaluation,the BP neural network model is established,according to the standard of groundwater quality（GB/T 14848—1993）.Through training the model for Kunming basin measured 12 pore water of groundwater quality evaluation,and the method of Analytic Hirrarchy Process evaluation results were compared.It is illustrated that this method of BP neural networks in groundwater quality evaluation is feasible,which has excellence as rapid speed and high precision,and it could evaluate the instance of groundwater quality.

关 键 词：BP神经网络 地下水水质评价 昆明盆地 

分 类 号：P641.12[天文地球—地质矿产勘探]