Adaptive Load Balancing in Electric Vehicle Charging Networks

Choudhary P.K.; Arya K.V.; Yadav A.S.; Ciro Rodriguez R. · ICIIS 2025 - Next-Gen Engineering for Industry 5.0: Innovating Intelligent Systems for Human Centric Future: Proceedings of 2025 IEEE 19th International Conference on Industrial and Information Systems (2025) · DOI: 10.1109/iciis69028.2026.11450713

Resumen

The growing use of electric vehicles (EVs) brings challenges in managing charging stations, especially in presence of facilities with limited vehicles serving in single slot. Common strategies like the Shortest Distance Matching Strategy (SDMS) and Random Matching Strategy (RMS) are based on simple assumptions, such as all EVs being the same, charging stations having the same capacity, and travel times being constant. While the Load Balancing Matching Strategy (LBMS) solves some of these issues, it still does not adapt well to real-world situations where conditions vary a lot. In this work an improved method called Enhanced Load Balancing Matching Strategy (ELBMS) is proposed by including three main upgrades: consider EVs with different driving ranges, adjusts station capacities based on high-demand areas (hotspots), and calculates travel times using real-time traffic data. The performance of the proposed scheme d with popular existing such as SDMS, and RMS for varying numbers of EVs and station size. The results clearly demonstrate that the proposed ELBMS outperformed the existing methods. It reduces the average time users spend for charging, and spreads the load more evenly at service station (reaching valley-to-peak ratios as high as 85%), and works well even when infrastructure is limited. This research may provide a practical, scalable, and effective model for future EV charging networks that need to be both smart and sustainable.

Instituciones: Universidad Nacional Mayor de San Marcos