A correlation for optimal steam-to-fuel ratio in a biogas-fueled solid oxide fuel cell with internal steam reforming by using Artificial Neural Networks

Abstract

It is a challenge to find the optimal amount of water steam to be added to the fuel in a biogas-fueled solid oxide fuel cell (SOFC). An under-optimal steam fraction impedes steam reforming reactions, while an over-optimal steam fraction reduces SOFC performance due to a fuel shortage. Water production inside the cell, as a result of electrochemical reactions, and the carbon deposition issue rises the complexity of the problem. The main novelty of the current study is to propose a mathematical correlation for the optimal steam-to-biogas ratio as a function of operating temperature and biogas composition using a 3D simulation-trained Artificial Neural Network (ANN). This correlation can be simply used without having to study the complex phenomena inside the biogas-fueled SOFC. The results indicate the optimal steam-to-fuel ratio is highly dependent on the temperature and biogas methane content; that is, for a temperature range of 873K–1273K and a biogas CH4/CO2 range of 0.82–3, the optimal steam-to-fuel ratio varies within a range of 0.3–1.3. It is also observed that the optimal steam-to-fuel ratio decreases with temperature, but increases with the biogas methane content. The ANN-produced correlation shows a good agreement with the simulation results and can be reliably used by engineers.