城市给水管网余氯的预测模型

选择余氯为研究对象,以南方某市给水管网水质监测的数据为基础,使用线性回归和非线性神经网络(ANN)方法建立模型,找到了一种利用在线监测数据和人工监测数据实时预测管网余氯的方法。通过建立给水管网水质模型,可以由监测系统动态回传的数据来实时的预测下一天人工点的水质。模拟的结果显示ANN模型比线性回归模型有更好的预测能力,预测的平均相对误差:ANN模型为14.9%,线性回归模型为25.8%。使用ANN模型可以实现实时预测。

Forecasting Model of Residual Chlorine in Urban Drinking Water Networks

The water quality indicator chlorine was taken as the studing object.Bases on the monitoring data from drinking water network of a southern city,the forecasting model of residual chlorine in urbam drinking water networks was set up with the application of the linear regression and non-linear artificial neural network (ANN).The residual chlorine concentrations in the water network could be forecasted by using on-line monitoring data and manual monitoring data.Simulation results showed that the ANN model gave better predictions than the regressive model.The average relative error of ANN was 14.9% and that of linear regression was 25.8%.The ANN model can forecast residual chlorine concentrations in real time.