BP人工神经网络模型在地下水水质评价中的应用  

Application of BP artificial neural network model in the evaluation of groundwater quality

作　　者：潘俊[1] 梁海涛[1] 岳丹丹[1] 赵磊[1] 

机构地区：[1]沈阳建筑大学市政与环境工程学院,辽宁沈阳110168

出　　处：《供水技术》2015年第6期6-11,共6页Water Technology

摘　　要：为了能够客观地对地下水水质进行综合评价,本文以西鞍山矿区为例,采用基于BP人工神经网络模型的评价方法对区内14个地下水水质监测点的水质进行了评价。考虑到地下水水质随季节性变化不大,以枯水期水质监测的主成分总硬度、溶解性总固体、硫酸盐、氯化物、铁和锰、硝酸盐、氟化物等指标作为评价因子,建立了地下水评价指标体系,并和模糊综合评价法的评价结果进行了比较,分类结果令人满意。评价结果表明,该模型设计合理、泛化能力强,对地下水水质评价具有较好的客观性、通用性和实用性,可为水质评价提供技术依据以及为有关部门治理水质提供理论依据和参考建议。In order to evaluate the groundwater quality objectively,taking west Anshan mining district for example,the water quality of 14 groundwater monitoring points in the district using the method based on BP artificial neural network model was evaluated. Considering the barely seasonal changes of groundwater quality,the evaluation index system of groundwater was established taking total hardness of principal component,dissolved total solids,sulfate,chloride,iron and manganese,nitrate,fluoride and other indicators monitored in dry season water quality as evaluation factors,and compared with the evaluation results of fuzzy comprehensive evaluation method,the classification results were satisfactory.The evaluation results showed that the model was reasonable and generalization ability was good,and had good objectivity,universal property and practicability to the evaluation of groundwater quality. It could provide technical basis for water quality evaluation,and provide theoretical basis and reference for the relevant departments to control water quality.

关 键 词：人工神经网络 水质评价 B-P算法 训练样本 矿区地下水 

分 类 号：TU991.11[建筑科学—市政工程]