Model Guided Extremum Seeking and Active Disturbance Rejection Control for Efficiency Real-Time Optimization of PEMFC System

In order to achieve real-time optimization (RTO) of system efficiency, an efficiency online supervisory control based on model guided extremum seeking and active disturbance rejection control is proposed for proton exchange membrane fuel cell (PEMFC) system. The system uncertainty existed in the system efficiency optimization performs in both optimization process and control process. Uncertainty in the optimization process is mainly related to the operating conditions changing and the system components degradation. Model guided extremum seeking is proposed to solve the problem of maximum efficiency trajectory searching online. Levenberg–Marquardt algorithm-based semiempirical modeling is proposed as a feedforward guidance to compensate the load disturbance. The extremum seeking regarded as a supervisory function is proposed to sense the deviation of system efficiency map and also to make online corrections. The uncertainty in the control process is called dynamic uncertainty existed at every control cycle. To cope with this dynamic uncertainty, an active disturbance rejection control is proposed for real-time trajectory tracking. All performance verifications are conducted on a real PEMFC system. The experiments results show the proposed strategy can be aware of the maximum efficiency trajectory and also has the ability to achieve RTO of system efficiency.