Work

Bayesian optimized support vector regression with a Gaussian kernel for accurate prediction of the state of health of lithium-ion batteries used for electric vehicle applications

Year: 2025

Type: article

Abstract: The state of health SoH of lithium ion batteries plays a predominant role in ensuring the safe and reliable operation of electric vehicles. In this, a novel SoH estimation approach using support vector regression with a Gaussian kernel optimized using the Bayesian optimization technique (BO-SVR with a Gaussian kernel) was proposed. Unlike, traditional approaches that use the internal resistance, and battery capacity as input parameters, this study utilized the equivalent discharging voltage difference interval and equivalent charging voltage difference interval, as they capture the dynamic voltage characteristics associated with the battery degradation. The model was simulated using MATLAB 2023a. The mean absolute error, R 2 , root mean squared error, and mean squared error were considered as performance indicators. The simulation results indicated that the proposed BO-SVR with a Gaussian kernel model had superior performance to other kernel SVR and Gaussian Process Regression models, with a reduced RMSE of 0.0082, thus demonstrating its potential to predict the SoH more accurately.

Source: Global Energy Interconnection

DOI: 10.1016/j.gloei.2025.02.003

Authors: Selvaraj Vedhanayaki, Vairavasundaram Indragandhi

Institutions: Ganpat University, Vellore Institute of Technology University

Language: English

Topic: Battery Modeling and Management

Subfield: Automotive Engineering

Field: Engineering

Domain: Physical Sciences

SDGs: Affordable and clean energy, Industry, innovation and infrastructure

Open Access status: diamond

Funder: Royal Academy of Engineering