Data-Based Collaborative Control for Air Supply System of PEMFCs.

The air supply system in Proton Exchange Membrane Fuel Cells (PEMFCs) is a multi-input multi-output (MIMO) system with inherent disturbances, which presents complex nonlinearities and strong coupling. Traditional mechanistic modeling approaches are inadequate for this system. This paper aims to leverage a purely data-based modeling technique and proposes a coupled control architecture to effectively manage and optimize the air supply system. The MIMO model-free adaptive controller (MF AC) achieves its control obj ectives by simultaneously controlling the air compressor speed and throttle valve opening, ensuring that the air mass flow rate and cathode pressure of the air supply system track the desired values. By comparing the single-input single-output compact form dynamic linearization MFAC (SISO CFDL-MFAC) with the MIMO CFDL-MFAC, we propose that the latter exhibits better control performance for the air supply system of PEMFCs. This is because the MIMO CFDL-MFAC can handle multiple input and output variables simultaneously, taking into account the system's coupling effects, capturing the system's complex dynamic characteristics more effectively, and controlling it in a more precise manner.