Study On Water Quality Forecasting And Evaluation Based On Combined Model

With the increasingly serious water pollution,water environment problem has become a major factor hindering the sustainable development of China's economy and society.how to curb water pollution and make more scientific,effective water resources governance is an extremely important issue.Reliable evaluation of water quality and accurate prediction of water pollution indicators are the key and difficult points in the management of water resources.Therefore,this paper establishes a water quality evaluation model and proposes a new water quality forecasting model based on the data from Lanzhou Xincheng bridge section in the Yellow Basin and Sichuan Panzhihua Longdong section in the Yangtze Basin,which provides a reference for water resources management and water pollution control.Based on “surface water quality standards”,this paper adopts the method of interpolate water quality index data in an equidistant and uniform manner to build research sample,and uses stratified sampling method to determine training sample.T-S fuzzy neural network is used to establish water quality evaluation model.The trained model is applied to the Yellow Basin and Yangtze Basin respectively and the results show its effectiveness and strong generalization ability.Considering the random,fluctuant and nonlinear characteristics of water quality index data,single forecasting model is difficult to achieve high accuracy.Therefore,this paper presents a new combined forecasting model with autoregressive integrated moving average model(ARIMA)and wavelet neural network(WNN),and implements bat algorithm(BA)to find the optimal weight of the individual models.Then,establishing combined forecasting model for each water quality index respectively,and compared with ARIMA,WNN,least squares support vector machine(LSSVM)and BP neural network.The results show that the combined model exhibits higher prediction accuracy.Finally,the predicted data of 2016 is put into the established water quality evaluation model,which performance further verifies its evaluation ability.