Multi-Objective Optimization of Sewage Treatment Based on BP Neural Network

With the over-exploitation of earth resources by human beings, the problem of environmental pollution is increasingly serious, and the pollution of surface water environment is an urgent problem to be solved at present. In this paper, six parameters of ammonia nitrogen (NH3-N) and total ammonia (TN) were selected as water quality evaluation indexes to establish the water quality grade evaluation system. With the water quality data of Jiulong Lake as training samples, a water quality identification model based on BP neural network was established. And by the golden section method to determine the optimal network structure, and by using MSE and R2 as indicators of accuracy test. The results show that the established water quality recognition based on BP neural network model has high precision, working for water quality assessment provides a good reference value.