基于Hopfield神经网络的地下水环境质量评价  

Groundwater Quality Assessment Based on Hopfield Neural Network

作　　者：初海波[1] 卢文喜[1] 伊燕平[1] 龚磊[1] 

机构地区：[1]吉林大学环境与资源学院,吉林长春130026

出　　处：《水电能源科学》2011年第3期31-33,共3页Water Resources and Power

基　　金：国家自然科学基金资助项目(41072171)

摘　　要：针对传统水质评价方法存在人为因素影响等问题,结合Hopfield网络在模式识别应用中的特点建立了Hopfield网络模型,综合评价了金泉工业园区的地下水水质情况,并利用其水质评价结果,以ArcGIS为平台获得了地下水水质分布图,可反映水质的空间变化规律。评价结果表明,金泉工业园区大部分地区地下水水质呈现Ⅲ类,Ⅰ、Ⅱ类水集中于西北部,而Ⅴ类水主要分布于东南部,评价结果符合水质实际状况。To solve the problems such as the impact of human factors for water quality evaluation with traditional methods,combined with the advantages of Hopfield neural networks in pattern recognition,Hopfield network model is established to evaluate groundwater quality of Jinquan Industrial Park comprehensively.And with the use of water quality evaluation results,distribution graph of groundwater quality is obtained by using the platform of ArcGIS,which can reflect the spatial variation of water quality.Evaluation results show that the groundwater quality in most areas of Jinquan industrial park is Class Ⅲ of water quality criteria,and ClassⅠandⅡ water concentrate in the Northwest region while ClassⅤwater mainly distributes in the Southeast region;the evaluation results accord with the actual state of water quality.

关 键 词：HOPFIELD神经网络 MATLAB 地下水水质评价 金泉工业园区 

分 类 号：TV211.12[水利工程—水文学及水资源] X824[环境科学与工程—环境工程]