Prediction Of Yellow River Water Quality Index Based On ICEEMD-DEWOA-BP Combined Model

The Yellow River is of great significance to our economic development and the progress of human civilization.However,as far as the current situation is concerned,due to the imperfect water quality monitoring system in our country,there is still no more efficient and accurate water pollution early warning system.This paper proposes a new method to predict water quality index.The first 664of 676 data of Yellow River water quality index are selected as training set to predict the data value of water quality index in the last quarter of 2018?41 weeks-52 weeks,12 weeks in total?.On the basis of improved complete ensemble empirical mode decomposition?ICEEMD?,the Differential Evolution method is adopted to optimize the calculation process of the Whale Optimization Algorithm to avoid the Whale Optimization Algorithm falling into local optimization.Then the combined optimization method DEWOA is applied to the BP neural network.Finally,the decomposed data are integrated to obtain the final prediction data.Through the study of two monitoring stations in the upper and lower reaches of the Yellow River basin,it is found that ICEEMD-DEWOA-BP method has significantly improved the prediction effect on the four indexes of water quality indexes:pH,DO,COD_{M n},NH₃-N,compared with the traditional Autoregressive Integrated Moving Average model?ARIMA?,Support Vector Regression?SVR?,BP neural network and optimized WOA-SVR,WOA-BP,DEWOA-SVR,DEWOA-BP and other models.This provides a brand-new method for the monitoring and early warning of the water quality of rivers in our country,effectively improves the monitoring speed and efficiency of water pollution,and provides a strong guarantee for the monitoring and supervision policies of relevant departments.