作　　者： ; ; ; (李雅蕾）;

机构地区： 青岛理工大学环境与市政工程学院

出　　处： 《青岛理工大学学报》 2016年第4期75-81,共7页

摘　　要： 在分析国内外供水管网水质研究现状的基础上,对管网水的浊度特性及演变规律进行了深入研究,总结了饮用水浊度指标在管网中的变化规律,并据此筛选出对管网水浊度变化影响较大的三方面因素,即流速、水龄、总铁含量.在对主要因素进行连续监测的基础上,运用MATLAB分别构建BP和RBF神经网络模型对管网水浊度进行模拟分析,结果表明模拟值与实测值吻合良好,RBF神经网络的非线性逼近能力更强,预测精度更高. Based on the statistical analysis of the research status of water quality of water supply network at home and abroad, the paper carries out an in-depth study of the water turbidity properties and evolution rule with the general regularity of turbidity index in water supply network summarized. Three aspects such as flow rate, water age,and total iron concentrations greatly impacting on the turbidity in water supply network filtered out. Based on the continuous monitoring of the main factors, it makes a study of the turbidity in water supply network with the help of MATLAB to structure the BP and RBF neural network. The simulation result shows that the error between simulation values and the measured values is minor, and the nonlinear approximation ability of RBF neural network is stronger and the prediction accuracy is higher.

关 键 词： 供水管网 浊度模拟 神经网络 神经网络

领　　域： [建筑科学]