Flatness-based feedforward control of polymer electrolyte membrane fuel cell gas conditioning system

Manufacturers of automotive applications rely on high-performance testing environments for the development of polymer electrolyte membrane fuel cell (PEMFC) technologies. The main component of a PEMFC testbed is the gas conditioning system, which controls the inlet gas temperature, stack pressure, relative humidity and mass flow of the gas at the inlet of the PEMFC stack. This paper presents a control concept for a highly dynamic PEMFC gas conditioning system. The main control challenge lies in the decoupling of the governing thermodynamic quantities. Therefore, the nonlinear control concept of exact input-output linearisation with an extended PID control structure is applied. Constraints on the actuators are incorporated by formulating the resulting control law as an optimisation problem. The robustness with respect to parameter uncertainties for the model is shown by investigating the trajectory tracking error of the disturbed system. Simulation results show an overall good performance for the trajectory tracking and demonstrate the robustness of the model against parameter uncertainties.