Fuel cell systems: Control as the key enabler

Part II: The potential of model predictive control

Abstract

This article uses an example to show the potential of model predictive control for fuel cell systems. It represents a modern and intuitive approach to solve the multivariable, coupled and constrained control problem and is thus a promising alternative to conventional control methods. For this purpose, the control problem from Part I of this two-part contribution is reviewed and an overview is given of the current control methods in industry and research. Based on this, the potential of a model predictive
controller is explained and an exemplary control design is presented for the application at hand. The effectiveness of this model predictive control design is then validated using two simulated scenarios. The results show that model predictive control is a suitable approach to systematically meet the requirements for future fuel cell systems in mobile applications.