Performance Evaluation of Multi-Objective Optimization Techniques for a Hybrid EV Charging Station
Energies · 14.11.2026 · 19 (22) · 5497
DOI: 10.3390/en19225497 ↗Kurz gefasst
Fachartikel in der Zeitschrift Energies, online erschienen am 14.11.2026. Verfasst von Sana Sultan und 3 weiteren Autor:innen. Die Arbeit nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Electric Vehicles (EVs) have become a global trend due to their environmental benefits, which reduce the dependency of the transport sector on fossil fuels. The rise of EVs has led to the adoption of an effective infrastructure for charging them, to provide convenience to users and reduce their range anxiety. Hybrid EV charging stations are proposed that integrate renewable energy sources such as solar and wind to reduce the reliance on the utility grid. However, the intermittent nature of renewable energy sources introduces instability in managing the charging stations. Moreover, the stochastic demand of EV charging poses challenges for the energy management of hybrid EVCSs. This study presents a performance evaluation of multi-objective optimization techniques for the energy management of hybrid EVCSs. It incorporates a Reinforcement Learning demand response mechanism to manage the load of EV charging. A multi-objective optimization problem is formulated to minimize operating cost and Loss of Power Supply Probability (LPSP), while evaluating grid dependency and renewable-energy utilization. Four different optimization techniques—Multi-Objective Particle Swarm Optimization (MOPSO), Non-dominated Sorting Genetic Algorithm II (NSGA-II), Non-dominated Sorting Genetic Algorithm III (NSGA-III), and Multi-Objective Evolutionary Algorithm based on Decomposition (MOEA/D) are implemented under identical operating conditions, objective functions, and constraints. Their performance is assessed using Pareto-front solutions, convergence, diversity, computational efficiency, and robustness over repeated runs. The results demonstrate distinct algorithmic behaviours and trade-offs among cost and reliability objectives. NSGA-II achieves the lowest cost and better robustness, while MOEA/D has the lowest LPSP and strong computational efficiency. NSGA-III shows better convergence, and MOPSO provides favourable diversity.
Zitieren
APA
Sultan, S., Faraz, S. M., Azeem, F., & Narejo, G. B. (2026). Performance Evaluation of Multi-Objective Optimization Techniques for a Hybrid EV Charging Station. Energies, 19(22), 5497. https://doi.org/10.3390/en19225497
Vancouver
Sultan S, Faraz SM, Azeem F, Narejo GB. Performance Evaluation of Multi-Objective Optimization Techniques for a Hybrid EV Charging Station. Energies. 2026;19(22):5497. doi:10.3390/en19225497
BibTeX
@article{Sultan2026,
author = {Sultan, Sana and Faraz, Sadia Muniza and Azeem, Fawad and Narejo, Ghous Bakhsh},
title = {Performance Evaluation of Multi-Objective Optimization Techniques for a Hybrid EV Charging Station},
journal = {Energies},
year = {2026},
volume = {19},
number = {22},
pages = {5497},
doi = {10.3390/en19225497}
}
EUVIMED