Publication detail

Comparison of machine learning techniques for estimating battery health

SEDLAŘÍK, M. KAZDA, T. CAPKOVÁ, D. VYROUBAL, P.

Original Title

Comparison of machine learning techniques for estimating battery health

Type

article in a collection out of WoS and Scopus

Language

English

Original Abstract

The increasing number of lithium-ion (Li-ion) batteries across a wide range of industries emphasizes their reliability, which is primarily achieved through accurate estimation of State-of-Health (SOH). Numerous methods exist for determining SOH, and this study focuses on estimating battery life parameters using machine learning (ML), which appears to be a fast and promising approach for this application. Experimental measurements were performed using Constant Voltage Constant Current (CCCV) tests, from which discharge cycle parameters were extracted for SOH estimation. The suitability of these parameters was verified using Pearson correlation analysis, demonstrating their appropriateness for estimation as well as their redundancy, which adversely affects overfitting. Two types of ML methods are compared, each with its own advantages and disadvantages: Support Vector Regression (SVR) and Gaussian Process regression (GPR).

Keywords

Support Vector Regression, Gaussian Process Regression, Li-ion battery

Authors

SEDLAŘÍK, M.; KAZDA, T.; CAPKOVÁ, D.; VYROUBAL, P.

Released

21. 7. 2024

Publisher

Brno University of Technology Faculty of Electrical Engineering and Communication of the Brno University of Technology

Location

Brno

ISBN

978-80-214-6257-1

Book

Advanced Batteries Accumulators and Fuel Cells – 25th ABAF

Edition

25

Pages from

126

Pages to

128

Pages count

3

URL

https://www.aba-brno.cz/download/2024-ABAF-Proceeding.pdf

BibTex

@inproceedings{BUT189520,
  author="Marek {Sedlařík} and Tomáš {Kazda} and Dominika {Capková} and Petr {Vyroubal}",
  title="Comparison of machine learning techniques for estimating battery health",
  booktitle="Advanced Batteries Accumulators and Fuel Cells – 25th ABAF",
  year="2024",
  series="25",
  pages="126--128",
  publisher="Brno University of Technology Faculty of Electrical Engineering and Communication of the Brno University of Technology",
  address="Brno",
  isbn="978-80-214-6257-1",
  url="https://www.aba-brno.cz/download/2024-ABAF-Proceeding.pdf"
}