On-Road Experimental Campaign for Machine Learning Based State of Health Estimation of High-Voltage Batteries in Electric Vehicles

Aug 5, 2026

Upload by Patrick

The study investigates machine learning approaches for estimating the state of health (SOH) of high-voltage electric-vehicle batteries using open-circuit voltage (OCV) data. Experiments rely on OCV measurements from 12 vehicles under different mileage conditions and seek correlations between OCV, stored energy, and battery SOH. Data were collected at Hyundai Motor Europe Technical Center GmbH using ETAS INCA and ETAS MDA. Six algorithms were compared, with random forest delivering the best SOH prediction accuracy. Compared with linear regression, it reduces mean absolute error by 96% for OCV and 97% for capacity (C) estimation, while clarifying differences in performance, complexity, and interpretability to support safer maintenance strategies.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages22

Reading Time55 min

AccessFree to download

Introduction

Experimental Data and Measurement Setup

Machine Learning Algorithms and Comparison

Results and Performance Evaluation

Implications for Maintenance Strategies

POLITECNICO DI TORINO Repository ISTITUZIONALE On-Road Experimental Campaign for Machine Learning Based State of Health Estimation of HighVoltage Batteries in Electric Vehicles Original On-Road Experimental Campaign for Machine Learning Based State of Health Estimation of High-Voltage Batteries in Electric Vehicles / Lelli, Edoardo; Musa, Alessia; Batista, Emilio; Misul, Daniela; Belingardi, Giovanni. -In: ENERGIES. -ISSN 1996-1073. -16:12(2023) . [10 .3390/en16124639] Availability: This version is available at: 11583/2979346 since: 2023-06-13T06:45:31Z Publisher: MDPI Published DOI:10.3390/en16124639 Terms of use: This article is made available under terms and conditions as specified in the corresponding bibliographic description in the repository Publisher copyright (Article begins on next page) 17 October 2023 energies Article On-Road Experimental Campaign for Machine Learning Based State of Health Estimation of High-Voltage Batteries in Electric Vehicles Edoardo Lelli 1, Alessia Musa 2, *, Emilio Batista 1, Daniela Anna Misul 2, * and Giovanni Belingardi 3 Citation: Lelli, E.; Musa, A.; Batista, E.; Misul, D.A.; Belingardi, G. On-Road Experimental Campaign for Machine Learning Based State of Health Estimation of High-Voltage Batteries in Electric Vehicles. Energies 2023, 16, 4639. [https://doi.org/](https://doi.org/)[ ](https://doi.org/)[10.3390/en16124639](10.3390/en16124639) Academic Editor: Satoru Okamoto Received: 14 April 2023 Revised: 24 May 2023 Accepted: 9 June 2023 Published: 11 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 Hyundai Motor Europe Technical Center GmbH, Hyundai-Platz, 65428 Ruesselsheim, Germany 2 Department of Energy (DENERG), Politecnico di Torino, 10129 Torino, Italy 3 Department of Mechanical and Aerospace Engineering (DIMEAS), Politecnico di Torino, 10129 Torino, Italy * Correspondence: alessia.musa@polito.it (A.M.); daniela.misul@polito.it (D.A.M.) Abstract: The present study investigates the use of machine learning algorithms to estimate the state of health (SOH) of high-voltage batteries in electric vehicles. The analysis is based on open-circuit voltage (OCV) measurements from 12 vehicles with different mileage conditions and focuses on establishing a correlation between the OCV values, the energy stored in the battery, and the battery SOH. The experimental campaign was conducted at the Hyundai Motor Europe Technical Center GmbH (Germany), and the data collection process took advantage of the ETAS Integrated Calibration and Application Tool (INCA) and the ETAS Measure Data Analyzer (MDA) software. Six machine learning algorithms are evaluated and compared, namely linear regression, k-nearest neighbors, support vector machine, random forest, classiﬁcation and regression tree, and neural network. Among the evaluated algorithms, random forest (RF) exhibits the best performance in predicting the state of health of high-voltage batteries, both for the OCV and the capacity (C) estimation. Speciﬁcally, if compared to the worst algorithm (i.e., linear regression), RF achieves a remarkable improvement with a reduction of 96% and 97% in the mean absolute error for the OCV and the C estimation, respectively. Furthermore, the comparison highlighted the main differences in the performance, complexity, interpretability, and speciﬁc features of the six algorithms. The ﬁndings of the present study will contribute to the development of efﬁcient maintenance strategies, thus reducing the risk of unexpected battery failures. Keywords: high-voltage batteries; state-of-health (SOH); machine learning (ML) algorithms 1. Introducti

FAQs

What data source is used to estimate battery state of health (SOH) in this study?

The study uses open-circuit voltage (OCV) measurements from 12 electric vehicles with different mileage conditions.

Which machine learning algorithm achieved the best SOH estimation performance?

Random forest (RF) showed the best performance for predicting battery SOH using both OCV and capacity (C) estimation.

How much improvement does random forest provide compared with linear regression?

Compared with linear regression, random forest reduces mean absolute error by 96% for OCV estimation and by 97% for capacity estimation.

Aug 5, 2026

Upload by Patrick