Research On Water Quality Prediction Model Based On PLS And SVR

The water environment is an integral part of the global natural ecological environment.At present,the water environment protection and governance have been paid more and more attention by the government and society with China's increasingly serious pollution situation.Water quality prediction is an important part of water environment research,at the same time,it is one of the important subjects of modern environmental science theory research.Considering the complexity of the water environment system,the water quality prediction model based on Partial Least Squares Regression(PLS)and Support Vector Regression machine(SVR)is proposed,according to the characteristics of the water quality of the Miju River in the Erhai Lake valley.The main research results are as follows:In order to solve the problem of the prediction accuracy of the water quality predictioin model is poor because the components which is extracted with Principal Component Analysis(PCA)have low best explanatory to input variables,the partial least squares regression method is used to extract the components in this paper,so that the extracted components are the best explanatory to input variables and can maximumly explain the output variables.In this way,the model input variable is reduced and the prediction accuracy of the model is improved.Aiming at the problem that the BP neural network is poor in the generalization of the small sample data and the instability of the model output,this paper uses the Support Vector Regression machine to carry out the nonlinear regression of the water quality data of the Miju River,which improves the prediction accuracy of water quality prediction model.To solve the problem that the optimization efficiency of Genetic Algorithm(GA)is not high and it is easy to fall into the local minimum value,this paper proposes an improved Genetic Algorithm to dynamically adjust the probability of crossover and mutation.The initial parameters of the Support Vector Regression machine are optimized while improving the optimization ability.Based on the water quality data of Miju River in the Water Quality Monitoring Station of Erhai Lake in Dali,this paper applies the model proposed in this paper to the water quality prediction of Miju River and compares it with the current water quality prediction model.The simulation results show that the prediction results of water quality prediction model proposed in this paper are more accurate and reliable.