Predicting Higher Heating Value of Sewage Sludges via Artificial Neural Network Based on Proximate and Ultimate Analyses

Abstract

The higher heating value (HHV) was an important factor for measuring the energy recovery price of sewage sludge, which was commonly determined by oxygen bomb calorimeter; however, there were problems of time consuming and high measurement cost. In this study, a back-propagation neural network (BPNN) model based on proximate and ultimate combination analysis was developed to predict the HHV of sewage sludge and the accuracy of the model was illustrated using statistical analysis. The results showed that the BPNN model had good accuracy, with a regression coefficient of 0.979 and 0.975 for the training and test groups, respectively. Several previously proposed linear models for predicting the HHV of sewage sludge were selected for comparison. The results showed that the BPNN model was the best among all models with the highest regression coefficient (0.975) and the lowest mean absolute deviation (0.385).