EV Charging Scheduling Using Clustering Algorithms in EV Fleet System

As electric vehicles (EVs) become more popular, managing the charging infrastructure for large EV fleets has become increasingly challenging. To address this, efficient scheduling of charging stations is crucial for maximizing charger usage, minimizing waiting times, and ensuring fair access for all...

Full description

Saved in:
Bibliographic Details
Published in:2025 International Conference in Advances in Power, Signal, and Information Technology (APSIT) pp. 1 - 6
Main Authors: Yadav, Maneesh, Jena, Satyaranjan, Panigrahi, Chinmoy Kumar, Naik, Amiya Kumar
Format: Conference Proceeding
Language:English
Published: IEEE 23.05.2025
Subjects:
Charging stations
Clustering algorithms
Customer Satisfaction
Elbow
Electric vehicle charging
EV Charging Scheduling
K-means Clustering
Maximum charger utilization
Optimization
Performance metrics
Resource management
Smart cities
Strain
Vehicle dynamics
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:As electric vehicles (EVs) become more popular, managing the charging infrastructure for large EV fleets has become increasingly challenging. To address this, efficient scheduling of charging stations is crucial for maximizing charger usage, minimizing waiting times, and ensuring fair access for all vehicles. This paper introduces a new approach to optimize EV charging scheduling using K-means clustering. The method groups EVs based on their proximity to available charging stations and their charging needs, creating clusters of vehicles with similar requirements. By using K-means, the approach allows for the dynamic allocation of EVs to stations, minimizing overall charging time and balancing the load across multiple chargers. Our simulations studies considers the four cases in EV fleet system having 75EVs, 150EVs, 225EVs and 300 EVs with a space availability of 20 charging slots each having 7.2 kW charger which demonstrates that 225 EVs outperforms rest of the cases with a almost 73 % better as compared to 75 EVs, similarly almost in every indices 13 % better to 150 EVs in fairness metrices. The approach is evaluated through key performance metrics, including station utilization, average waiting time, and fairness in distribution. By taking into account factors like battery state-of-charge and spatial distribution of vehicles, the model ensures efficient use of charging resources while reducing the strain on the grid. The results show that the proposed scheduling approach significantly outperforms traditional methods, improving charger utilization and reducing waiting times. This work highlights the potential of clustering algorithms to make EV charging scheduling more scalable, efficient, and fair, paving the way for better charging infrastructure in future smart cities and large fleet operations.
DOI:10.1109/APSIT63993.2025.11086104