Predictive HHV Model for Raw and Torrefied Sugarcane Residues

Abstract

Sugarcane residues (SCR), such as the bagasse and leaves, have the potential to be used as solid fuels either as a direct feed or when utilized after torrefaction. Like any fuel, higher heating value (HHV) is an important characteristic and factor to be considered for applications in combustion systems. Higher heating value is determined through calorimetric analysis, but would require complex experiments and skilled personnel to carry out such analysis. The technical challenges may, however, be avoided through the development of correlations for predicting the heating values of fuels. A predictive HHV model based on the proximate constituents of raw and torrefied SCR was developed in this study since existing models were found inadequate. Moisture is negatively correlated with HHV of fuels but is oftentimes excluded as a parameter in the models established. In principle, moisture does not contribute to the HHV, as it does not undergo combustion. However, it was observed that additional energy was required to release the moisture from the biomass matrix, thus contributing to the decrease in HHV. The general model was established through multivariate linear regression following the least squares method with data gathered from both actual experiments as well as those which have been reported in literature. The developed predictive HHV model has a coefficient of determination (r2) of at least 0.90 with a mean absolute error of <\u20096% and a mean bias error of <\u20091%. The predictive model was established in anticipation of potential utilization of SCR as a renewable source of energy.