Parallel structure-based decentralized model predictive control of vehicle PEMFC anode circulation system

Abstract

Excessive pressure difference between the cathode and anode and insufficient hydrogen supply in proton exchange membrane fuel cell (PEMFC) can affect their stability, reliability and operating life. Therefore, in this paper, a decentralized model predictive controller (DMPC) based on the parallel structure of the ejector and hydrogen circulation pump is proposed to resist the disturbance of anode pressure and flow rate caused by current variation and purging. In addition, this structure can make up for the defect of the single ejector structure with too narrow operating range. Simulation results show that the proposed DMPC control strategy has better pressure stability and robustness than the traditional model predictive controller (MPC), and can achieve the ideal hydrogen excess ratio faster under low load conditions with step load current, variable reference and purge disturbance. In particular, even under loading current and purge disturbance condition of PEMFC hybrid vehicles, the DMPC with purge disturbance model has better pressure tracking performance and control stability, with an average absolute error of 56.69Pa and a root-mean-square error of 722.82Pa, which is 6.65 % lower than that of the MPC controller.