首页 | 本学科首页   官方微博 | 高级检索  
     检索      

RBF神经网络在平顶山市地表水评价中的应用
引用本文:李磊,孙卉,翟秋敏,郭志永.RBF神经网络在平顶山市地表水评价中的应用[J].安徽农业科学,2008,36(26).
作者姓名:李磊  孙卉  翟秋敏  郭志永
作者单位:河南大学资源与环境科学研究所,河南开封,475001
摘    要:为准确和客观地评价地表水环境质量状况,运用MATLAB软件中的神经网络工具箱,结合K均值聚类方法建立径向基函数网络,对平顶山市2004年市控5个地表水断面进行了环境质量评价。在评价前根据平顶山市的实际情况对训练样本范围进行更改,将训练和测试样本进行归一化处理,同时利用RAND函数对训练样本进行插值保证神经网络充分学习。结果发现,K均值聚类法能快速准确地确定网络中心,用建立的径向基函数网络进行地表水质量评价,其评价结果与单因子方法的评价结果一致,并且具有计算速度快、量化评价结果便于同类水质间互相比较的优点。

关 键 词:RBF神经网络  平顶山市  地表水  环境质量评价

Application of RBF Neural Network in the Surface Water Assessment in Pingdingshan City
LI Lei et al.Application of RBF Neural Network in the Surface Water Assessment in Pingdingshan City[J].Journal of Anhui Agricultural Sciences,2008,36(26).
Authors:LI Lei
Abstract:In order to assess the surface water environment quality accurately and objectively,neural network toolbox of MATLAB was used combining with K-means method.The surface water quality of five sections in Pingdingshan City in 2004 was assessed.Due to the actual situation,the scope of training samples were changed before assessment.The training and test samples were normalized.The RAND function was used to construct enough training samples in order to keep the network full learning.The result showed that using K-means method could determine the network center fast and accurately,and the result was the same with that by the single factor method.The RBF network could compute fast and quantify the result,which was advantageous for the comparison of same kind of water quality.
Keywords:RBF neural network  Pingdingshan City  Surface water  Environmental quality assessment
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号