AI Prediction of H₂ Production in a Biomass Supercritical Water Gasification Fluidized Bed under Pulsating Inlets

Supercritical water gasification (SCWG) is a promising approach with the advantages of high gasification rate, hydrogen-rich gas production, and environmental friendliness. In this work, a computational fluid dynamic-discrete element method (CFD-DEM) model is proposed to investigate and optimize the process of supercritical water (SCW) biomass gasification. The effect of pulsating SCW inlets on gasification and artificial neural network prediction of H₂ yields were conducted. The sinusoidal SCW inlet affects the bubble behavior in the bed, but has little effect on the heat transfer. When the sinusoidal period is 1.0 s, the H₂ yields are higher. Based on the BP neural network model, the H₂ yields were predicted and the optimized operating conditions are specified. This study will provide a theoretical foundation and guidance for industrial applications.