Experimental implementation of LPV-based internal model control for low pressure fuel cell in hydrogen-based renewable energy system

Low-pressure fuel cells are widely employed in hydrogen-based renewable energy power systems due to their high efficiency and low complexity. However, the hysteresis and nonlinearity of the fuel cell blower pose significant challenges for air supply control. This study proposes an internal model control strategy based on an identified linear parameter varying model, enabling accurate tracking control of the air supply with strong robustness. M-sequences are applied at three equilibrium points to identify the system dynamics and construct its linear parameter varying model, based on which a novel internal model controller is introduced. Experimental validation demonstrates that both the internal model controller designed using a single identified point and the PI controller exhibit overshoot or hysteresis in flow control when the system deviates from that point. Moreover, the internal model controller based on a model-switching strategy shows severe instability during transitions between models. In contrast, the internal model controller based on the linear parameter varying model achieves high-precision flow tracking under all tested conditions, with a mean absolute error of 0.16 g/s and a root mean square error of 0.28 g/s under small-step condition. Even under significant model mismatch, the proposed controller maintains excellent accuracy and stability.