Machine Learning Reduced Order Model for Cost and Emission Assessment of a Pyrolysis System

Biomass pyrolysis is a promising approach for producing economic and environmentally friendly fuels bioproducts. experiments show that feedstock properties have significant impact on product yields composition. Scientists are developing detailed chemical reaction mechanisms to capture the relationships between biomass composition yields. These can be computationally intensive. In this study, we investigate use of machine learning reduced order model (ROM) assessing costs emissions biorefinery. We developed Kriging-based ROM predict 314 samples based results kinetic mechanism. The integrated into process calculating mass energy in commercial-scale (2000 tonne/day) estimated biofuel 65 130 gallons per ton dry biomass. This minimum fuel-selling prices $2.62–$5.43 gallon ?13.62 145 kg CO2 MJ. achieved an average mean square error 1.8 × 10–20 absolute 0.53%. suggest ROMs facilitate rapid screening biorefinery systems.