Hydrogen production as a green fuel in silica membrane reactor: Experimental analysis and artificial neural network modeling

In this work, artificial neural networks (ANNs) model has been developed for investigation of the silica membrane reactor (MR) performance during methanol steam reforming (MSR) reaction. Particularly, such parameters as the transmembrane pressure (from 0.5 to 1.5 bar), reaction temperature (from 513 to 573 K), gas hourly space velocity (GHSV) between 3300 and 10000 h-1 and Steam/MeOH molar ratio (from 1 to 3) have been taken to account from both experimental and modeling viewpoints in order to analyze their influences on the silica MR performance with respect to traditional reactor (TR) in terms of methanol conversion, CO selectivity, total hydrogen yield, hydrogen recovery, hydrogen and carbon monoxide compositions. The ANN model results have been validated by using portion of the experimental data. Moreover, regarding to optimization results of ANNs model, reaction temperature was selected as the most effective operating parameter in the silica membrane reactor and traditional reactor during MSR reaction.