Torrefaction of Bambara Groundnut Shell: experimental optimization and prediction of the energy conversion efficiency using statistical and machine learning approaches

Abstract

Optimising input variables is a necessity for efficient waste-to-energy conversion. Using a tubular furnace as the torrefaction medium, this study investigated the effects of four process parameters (viz., temperature, retention time, moisture contents and particle size) on the Higher Heating Value (HHV) and Energy Yield (EY) of Bambara Groundnut Shell (BBGS). An optimal HHV of 21.78 MJ/kg was obtained at a particle size of 1 mm, temperature of 260°C, retention time of 23 min and moisture content of 10%. The input–output relationships of the BBGS torrefaction process were modelled and validated using Response Surface Methodology (RSM) and Bayesian information criterion (BIC)-pruned stepwise regression (SR), resulting in a regression model that balances interpretability and strong prediction performance.

KEYWORDS:

• Bambara groundnut shells
• energy conversion efficiency
• higher heating value
• torrefaction
• waste-to-energy
• Stepwise regression

Disclosure statement

No potential conflict of interest was reported by the author(s).