支持向量机的给水管网水质综合评价研究

提出了基于支持向量机(SVM)的给水管网水质综合评价模型.在给水管网水质评价标准的基础上采用内插法获得学习样本,利用最小二乘支持向量机回归算法和高斯核函数对支持向量机进行训练,并用K-折交叉验证法优化模型参数,以优化参数的模型建立网络水质分级标准.将训练好的网络模型应用于水体实例并与其它几种评价模型加以比较,分析结果表明,SVM方法评价结果比较客观、合理,尤其在体现指标的极值作用方面具有独到的优势.

Research on Synthetic Evaluation of Water Quality in Water Supply Networks Based on the SVM

A model of comprehensive evaluation of water quality in water supply networks based on support vector machine(SVM) is proposed in this paper.Based on the standard of water quality assessment of water supply networks,study samples which are obtained by using interpolation method are trained with least squares support vector machine regression algorithm and Gauss kernel function and to get The optimization model parameter by using k-fold cross validation,then gotten assessment block.A case of the water quality is studied and compared with some other models.The result shows that SVM model is objective and reasonable,especially in the aspect of expressing the effect of extreme.