AI-Enabled Load Forecasting and Renewable Energy Optimization for Electric Vehicle Charging Stations
Innovative Computing Perspectives · 2026
DOI: 10.64229/947gge71 ↗Kurz gefasst
Fachartikel in der Zeitschrift Innovative Computing Perspectives, 2026. Verfasst von Nishchay Kshirsagar und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
As the number of electric vehicles continues to grow, managing peak demand, maintaining grid stability, and handling the variability of renewable energy are becoming increasingly difficult. Many existing studies look at forecasting or energy management in isolation, which leaves a gap when designing renewable__powered EV charging systems__especially for Indian cities with diverse climate and demand patterns. To address this, we present an integrated artificial intelligence (AI) framework that brings three components together: an Long Short-Term Memory (LSTM) model to forecast EV charging demand, a hybrid Random Forest-XG Boost model to predict renewable energy availability, and a Reinforcement Learning (RL) algorithm to manage energy flow in real time. The framework was tested using datasets from 2018 to 2024 for multiple Indian cities, sourced from NASA POWER, NREL, and Open EI. The results show clear improvements. The LSTM model reduces forecasting error by 26.8% compared to ARIMA, and the hybrid renewable energy predictor offers a 22.4% accuracy improvement. When combined with the RL-based optimization, renewable energy usage at charging stations increases from 62% to 80%, while grid dependence and operational costs both decrease__by 24.7% and 15%, respectively. Overall, this work offers a practical and scalable approach to building more efficient, reliable, and sustainable EV charging infrastructure suited for India’s smart cities.
Zitieren
APA
Kshirsagar, N., Somatkar, A., Gajalkar, S., Chahare, V., Gaikwad, M., & Majumder, H. (2026). AI-Enabled Load Forecasting and Renewable Energy Optimization for Electric Vehicle Charging Stations. Innovative Computing Perspectives. https://doi.org/10.64229/947gge71
Vancouver
Kshirsagar N, Somatkar A, Gajalkar S, Chahare V, Gaikwad M, Majumder H. AI-Enabled Load Forecasting and Renewable Energy Optimization for Electric Vehicle Charging Stations. Innovative Computing Perspectives. 2026. doi:10.64229/947gge71
BibTeX
@article{Kshirsagar2026,
author = {Kshirsagar, Nishchay and Somatkar, Avinash and Gajalkar, Shivam and Chahare, Vishal and Gaikwad, Mahendra and Majumder, Himadri},
title = {AI-Enabled Load Forecasting and Renewable Energy Optimization for Electric Vehicle Charging Stations},
journal = {Innovative Computing Perspectives},
year = {2026},
doi = {10.64229/947gge71}
}
EUVIMED