Multi-source energy management in electric vehicles: Performance analysis using metaheuristic optimization strategies

Abstract

Hybrid energy storage systems (HESS) are considered as the optimal energy source for electric vehicles (EVs). Indeed, optimizing energy management efficiency and enhancing system lifespan are vital to improving the sustainability and reliability of EVs. However, the high charging and discharging battery current, and high temperature affected the battery lifetime of EVs. Therefore, this research introduces a Hunt Tracking Optimization-based Artificial Neural Network (HTO-ANN) control technique to enhance the battery lifespan in HESS of EVs, which enhances current distribution and overall energy efficiency. Furthermore, the ANN control strategy regulates the power from the Supercapacitor (SC), based on the EV speed to maintain the charge-sustaining and charge-depleting modes of the battery and SC. Indeed, the ANN stabilizes SC reference power and mitigates sudden transitions in battery current by adjusting the duty cycle of the DC-DC converter. Additionally, the HTO technique determines the optimal solution to fine-tune the hyperparameters of the ANN model, which enhanced convergence speed and prevent getting trapped in local minima. In the research work, three standard driving cycles including FTP75, J1015 and UDDS are validated to prove model efficacy, which confirms the battery life and reliability of EV energy storage systems. Finally, the developed HTO-ANN model achieved the battery SoC as 85.96%, SC SoC as 69.13%, and Vehicle speed as 13.45 km/hr for the FTP75 driving cycle.