Fuel cell system humidity regulation and shutdown purge strategy using observer-based model predictive control to improve Time-to-Target and compressor energy performance

• A sliding mode observer is proposed to estimate the water content of fuel cell. • A model predictive control is used to fuel cell humidity regulating and purging. • Simulation and experiment results show the effectiveness of controller. The accumulation of water in fuel cell operation significantly influences its longevity and efficiency. This study focuses on humidity control and shutdown purge strategies in fuel cell systems, utilizing model predictive control (MPC) based on a fuel cell humidity sliding mode observer (SMO). The primary objective is to improve the time taken to reach set objectives and the energy efficiency of the air compressor in the fuel cell system. This will be achieved by adjusting the cathode air flow rate to manage humidity levels effectively, particularly during high-load operations and system shutdown. Experimental analysis conducted on an 80 kW Proton Exchange Membrane Fuel Cell system reveals that the observer’s estimations demonstrate remarkable accuracy, particularly for currents exceeding 100 A, showing minimal deviation from experimental results, with an overall relative error below 5 %. Simulations conducted under specifically designed conditions and the New European Driving Cycle demonstrate that the SMO-based MPC method enhances the efficiency of fuel cell humidity control and shutdown purging processes by reducing time and energy consumption by more than 40 % and 30 %, respectively, in comparison to fixed flow rate and interval flow rate methods. Moreover, the Hardware-in-Loop experimental results indicate that the developed SMO-based MPC method exhibits promising real-time operational capabilities, with the maximum relative error not exceeding 0.05 between simulation and experimental outcomes.