Accurate regulation of air flow rate and cathode pressure is crucial for enhancing the efficiency and durability of proton exchange membrane fuel cell (PEMFC) systems. However, the PEMFC air supply system exhibits strong nonlinearity, multivariable coupling, and sensitivity to parameter variations and external disturbances, which make precise mathematical modeling extremely challenging and consequently limit the effectiveness of conventional model-based control approaches. To address these issues, a data-driven sliding mode control strategy that requires neither prior model information nor structural knowledge is developed. First, a dynamic linearized data model capable of describing both separable and inseparable disturbances is constructed, and its dynamic equivalence to the original nonlinear system is rigorously established. A wavelet neural network is then employed to estimate the unknown parameters online, thereby improving estimation accuracy. Based on these data models, a data-driven sliding mode controller is designed for the coordinated regulation of air flow rate and cathode pressure. Furthermore, a novel hyperbolic reaching law is introduced to adaptively adjust the convergence rate and effectively alleviate chattering. Theoretical analysis proves that the proposed control scheme guarantees convergence to a quasi-sliding mode. Finally, comparative simulations with benchmark controllers demonstrate the effectiveness and superiority of the proposed method.
Bao et al. (2026) studied this question.