Artificial Neural Network-based harmonic mitigation in sustainable vehicle-to-home charging systems
Discover Applied Sciences · 31.05.2026 · 8 (8)
DOI: 10.1007/s42452-026-08932-3 ↗Kurz gefasst
Fachartikel in der Zeitschrift Discover Applied Sciences, online erschienen am 31.05.2026. Verfasst von Muhammad Zubair und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: University of Engineering and Technology, Lahore.
Abstract (Originalsprache)
Charging of Electric Vehicles (EVs) is a key pillar of sustainable mobility and intelligent energy management, particularly in Vehicle-to-Home (V2H) systems that enable bidirectional power sharing between EVs and the home. However, conventional EV battery charger design faces numerous challenges, including high inrush currents, low power factor, slow charging rates, DC-side voltage transients, and high total harmonic distortion (THD) on the AC side. These challenges result in less energy-efficient systems, reduced stability, and inefficient or impossible interaction with grid power. This study introduces an intelligent V2H charger architecture that employs a state-of-the-art Artificial Neural Network (ANN)-based controller, a dual-loop regulation design, and a novel Harmonic-Aware Squeeze–Excitation (HASE) layer to enhance performance. The system also uses an adjustable boost DC–DC converter for variable-voltage regulation and a Neutral Point–Switching Filter Compensator (NP-SFC) for harmonic mitigation and power factor correction. This work proposes an ANN controller implemented and validated in MATLAB/Simulink. The study also summarizes the comparative studies of the proposed controller with PID control strategies for V2H. The results obtained indicate that settling time has improved by up to 88%, voltage overshoot has decreased by 27.4%, and THD has reduced by 34.6%. In addition, inrush current decreased by 23.5%, and the power factor increased to 0.99. Overall, these results confirm that the proposed ANN + HASE controller exhibits faster dynamic response, improved power quality, and superior energy efficiency compared with PID and other baseline methods. The outcomes of the proposed research have high potential to advance efficient V2H-enabled energy systems, thereby contributing to the global Sustainable Development Goals (SDGs) and the Saudi Vision 2030 agenda by promoting clean energy, resilient infrastructure, and sustainable communities.
Förderung
- University of Engineering and Technology, Lahore
Zitieren
APA
Zubair, M., Manzoor, I., Shahid, M., Tariq, K. A., Jamal, M. Y., & Waleed, A. (2026). Artificial Neural Network-based harmonic mitigation in sustainable vehicle-to-home charging systems. Discover Applied Sciences, 8(8). https://doi.org/10.1007/s42452-026-08932-3
Vancouver
Zubair M, Manzoor I, Shahid M, Tariq KA, Jamal MY, Waleed A. Artificial Neural Network-based harmonic mitigation in sustainable vehicle-to-home charging systems. Discover Applied Sciences. 2026;8(8). doi:10.1007/s42452-026-08932-3
BibTeX
@article{Zubair2026,
author = {Zubair, Muhammad and Manzoor, Ijaz and Shahid, Muhammad and Tariq, Khawaja Adeel and Jamal, Muhammad Yasir and Waleed, Aashir},
title = {Artificial Neural Network-based harmonic mitigation in sustainable vehicle-to-home charging systems},
journal = {Discover Applied Sciences},
year = {2026},
volume = {8},
number = {8},
doi = {10.1007/s42452-026-08932-3}
}
EUVIMED