Past Infrastructures and Future Machine Intelligence (MI) for Biofuel Production: A Review and MI-Based Framework

Samuel Hansen,Amin Mirkouei

DOI: https://doi.org/10.1115/detc2018-86150

2018-08-26

Abstract:Recent interest in alternative energy sources, particularly biofuels from biomass, is becoming increasingly evident due to energy security and environmental sustainability concerns, such as depletion of conventional energy reserves and carbon footprint effects, respectively. Existing fuels (e.g., biodiesel and ethanol) are neither sustainable nor cost-competitive. There is a need to integrate the recent advanced manufacturing approaches and machine intelligence (MI) techniques (e.g., machine learning and artificial intelligence), targeted on the midstream segment (i.e., pre-/post-conversion processes) of biomass-to-biofuel supply chains (B2BSC). Thus, a comparative review of the existing MI approaches developed in prior studies is performed herein. This review article, additionally, proposes an MI-based framework to enhance productivity and profitability of existing biofuel production processes through intelligent monitoring and control, optimization, and data-driven decision support tools. It is further concluded that a modernized conversion process utilizing MI techniques is essential to seamlessly capture process-level intricacies and enhance techno-economic resilience and socio-ecological integrity of B2BSC.