A Proposed Equation for Predicting the Heating Value of Thai Bagasse

Several empirical equations, such as those by Dulong and Demirbas, have been widely used to estimate the higher heating value (HHV) of fuels from their elemental composition. However, when applied to Thai bagasse, these models exhibited substantial errors, with average deviations exceeding 10% from laboratory-tested values. To address this, 47 sets of elemental composition data for bagasse were compiled from Environmental Impact Assessment (EIA) reports of biomass power plants in Thailand. Two regression-based predictive models tailored to Thai bagasse were developed and evaluated. The performance of the developed models was evaluated against six established equations from international studies using four statistical indicators—Mean Absolute Error (MAE), Mean Bias Error (MBE), Average Bias Error (ABE), and Average Absolute Error (AAE). These indicators were selected to assess both the unsystematic bias and the overall predictive accuracy of the models. The proposed models demonstrated marked improvements, reducing MAE by 3.74% compared to the best existing equation. Model 1 (HHV = 17.137 – 0.161 × Mc) yielded MBE = –0.061 MJ/kg, MAE = 0.52 MJ/kg, ABE = +0.0171%, and AAE = 5.75%, while Model 2 (HHV = 1.496 + 0.157 C + 0.19 O) achieved MBE = +0.037 MJ/kg, MAE = 0.37 MJ/kg, ABE = +0.98%, and AAE = 4.12%. These results confirm that the developed models provide significantly improved HHV prediction accuracy for Thai bagasse, which provides better fuel management for bagasse-based power plants.