Artificial Neural Network for Combined Steam-Carbon Dioxide Reforming of Methane

Abstract

The CH4 conversion, CO2 conversion, and H2/CO ratio were set as dependent variables, as the feed rate, flow rate and reaction temperature as independent variables in the complex reaction of methane. We used the Artificial Neural Network (ANN) technique to build a model of the process. The ANN technique was able to predict the reforming process with higher accuracy due to the training capability. The reaction temperature has the greatest effect on the CO2–CH4 reforming reaction. This is because the catalytic reaction temperature has a direct influence on the thermodynamic value and the reaction rate and the equilibrium state.