Algae’s potential as a bio-mass source for bio-fuel production: MLR vs. ANN models analyses

Algae have gained significant attention as a sustainable bio-mass source for bio-fuel production due to their rapid growth rate and high lipid content. The increasing demand for renewable energy sources highlights the need for accurate predictive models to optimise bio-fuel production. This study hypothesises that artificial neural networks (ANNs) outperform multiple linear regression (MLR) in predicting the lower heating value (LHV) and gas yields of algae bio-mass. The methodology integrates elemental analysis with statistical and machine learning models to estimate LHV and gas yields, comparing the performance of ANN and MLR. Results indicate that ANN models provide higher predictive accuracy, capturing complex non-linear relationships, with an LHV prediction of 20.31 MJ/kg compared to 19.55 MJ/kg from MLR. These findings suggest that ANN models offer greater reliability in energy prediction, though at a higher computational cost. It is concluded that while ANN provides superior accuracy, MLR remains useful for simpler applications due to its efficiency and interpretability.