An artificial intelligence approach to predict gross heating value of lignocellulosic fuels


JOURNAL OF THE ENERGY INSTITUTE, vol.90, no.3, pp.397-407, 2017 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 90 Issue: 3
  • Publication Date: 2017
  • Doi Number: 10.1016/j.joei.2016.04.003
  • Page Numbers: pp.397-407
  • Keywords: Biomass, Gross heating value, Artificial neural network, Proximate analysis, PROXIMATE ANALYSIS, CALORIFIC VALUE, BIOMASS FUELS, COALS


The gross heating value (GHV) is one of the most significant properties of biomass fuels in designing and operating any fuel processing systems. This study deals with a new method to calculate the GHV from the proximate analysis of different kinds of lignocellulosic fuels by using Levenberg-Marquardt trained artificial neural network (ANN) as an artificial intelligence method. Furthermore, a new nonlinear regression model was developed for this study. The published correlations were employed with the various biomasses to obtain a comparison with the ANN model and developed nonlinear correlation in this study. The results indicate that the artificial intelligence approach offers a high degree of correlation and its robustness and capability to compute GHV of any lignocellulosic fuels from its proximate analysis. Therefore, the proposed artificial intelligence is highly promising tool to use in designing and operating of any thermolysis process for lignocellulosic fuels. (C) 2016 Energy Institute. Published by Elsevier Ltd. All rights reserved.