Estimation of Energy Consumption in Real-Time EV Sensor Data through Explainable AI and Machine Learning Algorithm

Sathishkumar S., Yogesh Rajkumar R.

2025

Abstract

Electric Vehicles (EVs) are a wonderful option for sustainability as they are changing the future of Transportation for the better by ensuring lesser dependency on fossil fuels and a reduced level of emissions. These enable EVs to collect huge volumes of real-time data on speed, acceleration, battery charge, and environment, all of which are critical for making energy efficient decisions. The real-time estimation of energy consumption using machine learning and explainable A.I. (XAI) to accurately interpret sensor data is the focus of this research. Mercury is the closest planet to sun. Like existing research, which mainly investigated energy consumption based on classical approaches or simple machine learning models, the current work utilizes state-of-the-art models, such as Random Forest and Neural Networks, using rich real-world data from Battery Electric Vehicles (BEVs) running in different driving scenarios. SHapley Additive explanations (SHAP) method is also used for model interpretability to understand how various parameters impact energy consumption, e.g., vehicle speed and battery current. This characterization not only facilitates improved accuracy in the prediction of energy consumption but also greatly aids the identification of determinants driving overall energy inefficiency during live operational conditions. This proposed approach builds on the previous work with increased accuracy and adaptability in prediction through XAI that aids in developing more refined strategies for energy management. In the long run, this study aids in optimizing EV capabilities, prolonging battery duration, and minimizing range anxiety, all of which are vital for increasing EV adoption and informing transportation electrification policy in the future.

Paper Citation

in Harvard Style

S. S. and R. Y. (2025). Estimation of Energy Consumption in Real-Time EV Sensor Data through Explainable AI and Machine Learning Algorithm. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 789-798. DOI: 10.5220/0013872900004919

in Bibtex Style

@conference{icrdicct`2525,
author={Sathishkumar S. and Yogesh R.},
title={Estimation of Energy Consumption in Real-Time EV Sensor Data through Explainable AI and Machine Learning Algorithm},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25},
year={2025},
pages={789-798},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013872900004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25
TI - Estimation of Energy Consumption in Real-Time EV Sensor Data through Explainable AI and Machine Learning Algorithm
SN - 978-989-758-777-1
AU - S. S.
AU - R. Y.
PY - 2025
SP - 789
EP - 798
DO - 10.5220/0013872900004919
PB - SciTePress