Nonlinear model predictive control of PEMFC anode hydrogen circulation system based on dynamic coupling analysis.

Anode hydrogen circulation system with purge valve has become one of the most effective ways for proton exchange membrane fuel cell (PEMFC) to alleviate performance degradation caused by nitrogen permeation and improve the hydrogen utilization rate. However, nonlinearity and multivariable coupling of the system make it difficult for PEMFC to achieve pressure and hydrogen supply synchronous tracking under complex operating conditions. To solve the problem, a novel nonlinear model predictive control scheme based on coupling analysis is proposed in this paper. Firstly, coupling characteristics of the hydrogen circulation system and the control pairing of the multivariable system are analyzed based on the relative gain array method. Then a novel control scheme based on an adaptive model predictive controller and a nonlinear model predictive controller is designed to keep the anode pressure stable and sufficient hydrogen supply, which use nonlinear observers to estimate the internal states online. Finally, the proposed controllers are implemented in control experiments of a 50 kW PEMFC anode hydrogen circulation system. The results demonstrate that the proposed control approach has great dynamic performance, anti-disturbance ability and can maintain a high hydrogen utilization rate under purge operation, current disturbance and uncertain intake pressure. • A PEMFC anode model with gas permeation and water transmission is proposed. • Multivariable coupling and control pairing of anode system are analyzed. • LPV-MPC and NMPC with state observers for anode pressure and flow regulation. • Dynamics of the recirculation system are improved under pressure fluctuations. • Trade-off between dynamic performance and efficiency of anode system is discussed. [ABSTRACT FROM AUTHOR]