An Indeterminate Configuration for Energy Management in FCHEV System

The problem of energy management for fuel cell/ultra-capacitor hybrid electric vehicles (FCHEVs) in the face of uncertainty is looked at in this paper. Energy efficiency and the slow dynamics of the fuel cell system are important performance factors, as is the resilience of the solutions. Although the majority of previous research has focused on attaining the highest level of efficiency inside an environment that is predetermined, this study considers elements impacting the generation of energy, alteration, and consumption. Errors have been attributed to numerous factors, including variations in operational circumstances, estimating, and modelling. When applied to real-world scenarios, neglecting such uncertainties results in increased operational costs, decreased overall scheme competence, failed performance, and even impracticable results owing to limitations defilement. Research on a Neural Network (NN) based energy management system (EMS) is conducted to guarantee optimum but resilient performance when input parameters are unknown. The postulated NN-based approach incorporates ambiguity into the limitation ranges and cost function, shielding the scheme performance from issues related to optimality and viability. Considering the NN technique is renowned for its conservatism, we decrease computational costs and enable more nuanced decision making through the inclusion of conservatism level variables.