Optimal Control Design for Proton Exchange Membrane Fuel Cell via Genetic Algorithm

Water management in PEM fuel cell has become an important issue since it can improve cell performance and lifetime. This paper presents a dynamic model of Proton Exchange Membrane Fuel Cell (PEMFC), and an optimal PID controller based on Genetic Algorithm has been developed. This process can control the cell voltage with changes like current disturbance and uncertainty in model. The proposed controller is compared with CHR-PID (Chien-Hrones-Roswick approach to design a classic PID controller) and predictive control. The comparison shows that an optimal controller has good performance. Models and optimal PID controller are implemented in MATLAB and SIMULINK environment. Simulation results show that, this approach can adjust the anode and cathode water mole fractions. When changes applied to model, the cell voltage oscillates and after little time converges to setpoint and it would be constant. Thus, the cell water content maintains within a satisfactory interval irrespective of changes in the cell current and parameter in model. This process is caused to extend the lifetime of PEM fuel cell stack.