In order to improve the automation level of predicting the effluent quality of papermaking wastewater and enhance the wastewater treatment capacity,this study adopts an augmented matrix to perform dimensionality enhancement on the raw data of papermaking wastewater,and then determines the optimal number of features through dynamic chronic feature analysis.On this basis,regression prediction analysis is performed on the sample data with the current number of features using a random forest model.To verify the effectiveness of the dynamic chronic feature random forest composite model,the prediction error of this model was statistically analyzed and compared with other models.It was found that the average absolute percentage error(MAPE),mean square root error(RMSE),and coefficient of determination R2 of the prediction results of the dynamic chronic feature random forest composite model were 0.02,2.88,and 0.81,respectively,which were significantly better than the traditional chronic feature support vector regression composite model.This showed an ideal level of accuracy in predicting the effluent quality of papermaking wastewater and had certain application value.
维普
万方数据