Prediction of quality parameters of biomass pellets from proximate and ultimate analysis

The real-time prediction of crucial biomass pellet quality parameters such as higher heating value (HHV) and mechanical durability (MD) will allow for more efficient operation of energy production systems. Multiple linear regression (MLR) models were developed to predict HHV and MD from proximate and ultimate analysis of biomass pellets. A diverse range of biomasses from energy crops including pine, Miscanthus, reed canary grass, tall fescue and short rotation coppice willow were used to produce the pellets. HHV and MD of the pellets were predicted with coefficients of determination of 0.99 and 0.94, respectively, and standard errors of the estimate of 0.08 MJ kg−1 (Range: 16.39-18.92 MJ kg−1) and 0.49% (Range: 92.6-97.5%), respectively. This study demonstrates that MLR can be used to predict additional information of HHV and MD of biomass pellets from proximate and ultimate analysis. Important quality indices for diverse biomass pellets are also reported.