Neural Network Predictive Control of a Tubular Solid Oxide Fuel Cell

The dynamic behavior and control of a tubular solid oxide fuel cell will be studied in this paper. The effect of fuel/air temperature and pressure will be investigated. Controlling the average stack temperature is the final objective of this study due to a high operating temperature of the system. In this case, temperature fluctuation induces thermal stress in the electrodes and electrolyte ceramics; therefore, the cell temperature distribution should be kept as constant as possible. A mathematical modeling based on first principles is developed. The fuel cell is divided into five subsystems and the factors such as mass/energy/momentum transfer, diffusion through porous media, electrochemical reactions, and polarization losses inside the subsystems are presented. Dynamic fuel-cell-tube temperature responses of the cell to step changes in conditions of the feed streams will be presented. A neural network model predictive controller (NNMPC) is then implemented to control the cell-tube temperature through manipulation of the temperature of the inlet air stream. The results show that the control system can successfully reject unmeasured step changes (disturbances) in the load resistance.