Improved Prediction of the Higher Heating Value of Biomass Using an Artificial Neural Network Model Based on the Selection of Input Parameters

Recently, biomass has become an increasingly widely used energy resource. The problem with the use of is its variable composition. most important property that determines content and thus performance fuels such as heating value (HHV). This paper focuses on selecting optimal number input variables using linear regression (LR) multivariate adaptive splines approach (MARS) to create artificial neural network model for predicting selected biomass. MARS data better than LR model. best modeling results were obtained a three neurons nine in hidden layer. was confirmed by high correlation coefficient 0.98. show (ANN) models are effective calorific woody field biomass, can be considered worthy simulation feedstocks their blends renewable fuel applications.