An artificial intelligence approach to predict a lower heating value of municipal solid waste

Abstract

The lower heating value is an important parameter to conduct the modeling of any fuel processing system. In this study, the relationship between the physical composition of municipal solid waste and its lower heating values was investigated using a Bayesian regularized artificial neural network (ANN) as an artificial intelligence approach. A new nonlinear regression model was also developed in this study. The artificial intelligence approach was compared using the developed and published correlations. The approach offers a high degree of correlation, and as a result, the ANN provides a useful tool for designing any thermolysis process for municipal solid waste.