Mathematical models for predicting the higher heating value and ultimate analysis of biomass

Abstract

The higher heating value (HHV) and elemental composition are important properties of biomass and they are necessary to perform computer simulations of the biomass thermal conversion processes. Both properties require equipment to measure that can be expensive, while the proximate analysis required equipment that is typically found in a laboratory. This work presents a general correlation, based on proximate analysis of biomass, to calculate elemental composition and two new correlations, based on proximate and ultimate analyses of biomass and char used for prediction of the higher heating value. The correlations based on ultimate analysis and proximate analysis show an average absolute error (AAE) of 5.59% and 6.81%, respectively, an average bias error (ABE) of 0.17% and 1.52%, a mean bias error (MBE) of − 0.09 and − 0.06, and root mean square error (RMSE) of 1.52 and 1.98 concerning the value of HHV. To calculate the elemental composition from proximate analysis, the correlations were proposed of C, H and O. The AAE of these correlations were 3.86%, 7.49%, and 5.68%, the ABE were − 1.10%, − 0.22%, − 2.23%, the MBE were 0.58, 0.06, and 1.14 while the RMSE were 2.65, 0.6, 3.45 with respect to measured values of C, H, O, respectively. The AAE values of the correlation proposed in this study using ultimate analysis are lower than those found for the other studied correlations.