Exergetic energy management of fuel cell electric vehicles using Tasmanian devil optimisation and recalling recurrent neural network algorithm

This paper presents an energy management strategy (EMS) for electric and fuel cell electric vehicles (EMFCEV) using a novel TDO-RERNN approach, integrating Tasmanian devil optimisation (TDO) and recalling recurrent neural network (RERNN) algorithms. The study focuses on identifying driving conditions, calculating fuel cell output, classifying battery state of charge (SoC), and optimising power distribution among energy storage systems (ESS). Implemented and evaluated on the MATLAB platform, the proposed method demonstrates improved performance over existing methodologies, offering significant enhancements in vehicle energy efficiency and fuel consumption reduction. The TDO-RERNN approach achieves a notable 99% accuracy in predicting and optimising energy usage and distribution in FCEVs.