Leveraging XGBoost machine learning with GWO-PSO-MVO optimization algorithms in an advanced biomass-assisted steam methane reforming for blue H2 production with CO2 capture: 4E analysis and life cycle assessment

The most common technology in the generation of hydrogen is steam methane reforming (SMR), which is, however, accompanied with a significant amount of CO2 emissions. This study presents an integrated multigeneration design, which is capable of producing blue hydrogen by combining biomass-assisted SMR and a chemical absorption unit that captures CO2 out of hydrogen-rich syngas, thus lowering the carbon footprint of hydrogen generation. The setup further incorporates cascade heat recovery through an organic Rankine cycle and an absorption chiller to convert the excess heat as emitted during combustion of biomass to electricity and chilled water. An intense evaluation, involving thermodynamic, economical, and environmental analysis and the life-cycle assessment (LCA), was carried out. As a result, the CO2 capture module captures 4901 kg/h of direct CO2 and decreases the total global warming potential (GWP) to 2.44 kg/kWh in comparison with conventional SMR. The economic analysis shows that the project has high feasibility with the net present value of 99.31 M$ and an investment payback period of 2.10 years. The Multi-Objective Multi-Verse Optimization (MOMVO) model maximized exergy efficiency and minimized exergo-economic cost and environmental impact at the same time, giving the optimal operating conditions with an exergy efficiency of 62.12%, a total exergy unit cost of 19.12 $/GJ, and a GWP of 1.019 kg/kWh. These findings underscore the potential of the suggested system in the efficient and sustainably blue hydrogen generation.