This poses higher requirements for the refined operation of urban wastewater treat-ment plants.This study analyzes and compares different methods for predicting influent flow rates in wastewater plants,conducts correlation analysis on factors affecting influent conditions,and em-ploys a machine learning-based BP neural network model to predict the daily influent flow rates and chemical oxygen demand(COD)of wastewater plants.Utilizing operational data from two typical wastewater treatment plants located in the southern and northern regions of China,the results demonstrate that the model exhibits good prediction accuracy,with discrepancies from actual data kept within 5%.
关键词
污水处理厂/进水预测/短期预测/机器学习/神经网络
Key words
Sewage treatment plant/Water inflow forecast/Short term prediction/Machine learning/Neural network
维普
万方数据