Event-Triggered Robust Prescribed Performance Control for Proton Exchange Membrane Fuel Cell Temperature Considering Sensor Faults

This is the authors' abstract. We don't add key points for this paper.

Aiming at the practical limitations of proton exchange membrane fuel cell (PEMFC) systems, this article presents a model-independent robust stack temperature (ST) tracking control scheme with event-triggered communication and prescribed performance to improve the output efficiency and health state of the PEMFC. The main contributions lie in three aspects: 1) to exactly approximate the lumped disturbance (involving unmodeled dynamics, known/unknown disturbances, sensor faults, etc.), an event-triggered adaptive reduced-order extended state observer (ESO) is developed, which avoids large phase lag, the need for a detailed plant model, and unnecessary communication; 2) an event-triggered adaptive sliding mode controller integrating with a novel fixed-time dynamic prescribed performance function (FDPPF) is designed to qualitatively constrain the transient and steady-state behaviors of the ST tracking error, moreover, the control algorithm failure caused by the singularity that the tracking error exceeds the preset constraint can be strictly excluded; and 3) different event-triggering mechanisms are, respectively, constructed for intermittent ST measurement and control to save the communication resources in the sensor-to-controller and controller-to-actuator channels. The simulation results indicate that, under the disturbances of load current, system uncertainties, measurement noise, and sensor faults, the proposed control approach achieves superior ST regulation performance and satisfies the preset error constraint with less communication cost.