Adaptive neuro-fuzzy control for dual boost converter in fuel cell electric vehicles

Electric vehicles powered by fuel cells provide clean energy with zero emissions and are highly sustainable. Increased demand for sustainable transport has rendered hydrogen fuel cell electric vehicles (FCEVs) a feasible choice compared to conventional internal combustion engines. However, efficiency is significantly dependent on the strength of power electronic interfaces and smart control. This research combined an adaptive neuro fuzzy inference system (ANFIS) with two switched boost-converters to manage power from a fuel cell to a battery to deliver a constant voltage and to achieve optimal energy use. This improved the voltage regulation capability, particularly during load change. ANFIS combines the learning ability of neural networks with the decision structure of fuzzy logic. The hybrid approach is conducive to faster response, greater flexibility, and better energy efficiency over varying operating conditions. The ANFIS model recorded a value of root mean square error (RMSE) of 0.0024, confirming high accuracy in dynamic loading handling. Simulation outcomes support the effectiveness of the system and better performance when compared with traditional proportional integral (PI) controllers.