Artificial Intelligence Approach In Predicting Biomass-to-Biofuels Conversion Performances

Abstract

Energy and environmental biotechnologies have developed rapidly in recent decades, driven by societal challenges such as climate change, energy security and pollution control. The optimization of biotechnological processes is an essential issue for increasing the efficiency of bioconversion performances and ensuring product quality. The use of artificial intelligence and computer technology in conjunction with conventional simulation and modeling techniques may bring major benefits in defining optimal process parameters and reducing the overall process cost. This paper aims to provide insight on the potential applications of artificial intelligence in biomass-to-biofuels conversion processes, with a focus on predicting the production of biofuels by biochemical processes, given the high availability of wet bio-based resources. Appropriate artificial intelligence techniques are identified taking into account specific input and output variables for transesterification, alcoholic fermentation, anaerobic digestion and dark fermentation processes. A particular application for the computerized prediction of biomethane production generated by a mixture of selected organic waste was investigated by the authors using a two-factor central composite design methodology and STATISTICA 10 software. The predicted results were validated by performing a simultaneous lab-scale fermentation experiment that confirmed a high level of reliability in the use of computer technology for this specific application.