Maximum power point tracking in fuel cells an AI controller based on metaheuristic optimisation

The increasing concern about global warming and the depletion of fossil fuel reserves has led to a growing interest in alternative energy sources, particularly fuel cells (FCs). These green energy sources convert chemical energy into electrical energy, offering advantages such as quick initiation, high power density, and efficient operation at low temperatures. However, the performance of FCs is influenced by changes in operating temperature, and optimal efficiency is achieved by operating them at their maximum power point (MPP). This study uses Proton Exchange Membrane Fuel Cells (PEMFCs) to charge electric vehicles (EVs), amplifying the voltage generated by the FC using the Interleaved Boost-Cuk (IBC) converter. The optimal tracking of the maximum power output is achieved using the Improved Mayfly optimized (IMO) Cascaded Adaptive Neuro Fuzzy Inference System (Cascaded ANFIS). The study uses MATLAB to simulate the task in various settings and analyze the relevant performances, demonstrating enhanced efficiency and power tracking outputs. The proposed converter efficiency has improved to 94% with a minimal part count of 2 switched configurations. configuration. The applied control logic, in my opinion, Cascaded ANFIS is capable of operating the BLDC with an operational efficiency of 98.92%, including better output voltage generations of 350 V.