Lithium-ion Battery State of Health Estimation Using Empirical Mode Decomposition Sample Entropy and Support Vector Machine

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

Abstract

Accurate knowledge of the state of health (SOH) is important for battery management system to maintain safe operation of batteries and extend their lifetime. In order to improve the accuracy in estimating the SOH of a battery, a new method based on empirical mode decomposition sample entropy (EMDSE) and support vector machine (SVM) is proposed in this paper. Compared with the tradition SE-based method, EMD is used to filter the noise of the original signal. Then the EMDSE as a new feature is used for the SVM training, and the potential relationship between the SOH and the EMDSE is established. Finally, the effectiveness of the proposed method is verified by experimental results.
Original languageEnglish
Title of host publication2020 IEEE Applied Power Electronics Conference and Exposition (APEC)
Publication statusAccepted/In press - 2020

Fingerprint

Support vector machines
Entropy
Health
Decomposition
Lithium-ion batteries
Battery management systems

Keywords

  • Lithium-Ion Battery
  • State of Health
  • Sample Entropy
  • Empirical Mode Decomposition

Cite this

@inproceedings{1563ddf35671424886922dcac06b232e,
title = "Lithium-ion Battery State of Health Estimation Using Empirical Mode Decomposition Sample Entropy and Support Vector Machine",
abstract = "Accurate knowledge of the state of health (SOH) is important for battery management system to maintain safe operation of batteries and extend their lifetime. In order to improve the accuracy in estimating the SOH of a battery, a new method based on empirical mode decomposition sample entropy (EMDSE) and support vector machine (SVM) is proposed in this paper. Compared with the tradition SE-based method, EMD is used to filter the noise of the original signal. Then the EMDSE as a new feature is used for the SVM training, and the potential relationship between the SOH and the EMDSE is established. Finally, the effectiveness of the proposed method is verified by experimental results.",
keywords = "Lithium-Ion Battery, State of Health, Sample Entropy, Empirical Mode Decomposition",
author = "Xin Sui and Shan He and Daniel-Ioan Stroe and Remus Teodorescu",
year = "2020",
language = "English",
booktitle = "2020 IEEE Applied Power Electronics Conference and Exposition (APEC)",

}

Lithium-ion Battery State of Health Estimation Using Empirical Mode Decomposition Sample Entropy and Support Vector Machine. / Sui, Xin; He, Shan; Stroe, Daniel-Ioan; Teodorescu, Remus.

2020 IEEE Applied Power Electronics Conference and Exposition (APEC). 2020.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Lithium-ion Battery State of Health Estimation Using Empirical Mode Decomposition Sample Entropy and Support Vector Machine

AU - Sui, Xin

AU - He, Shan

AU - Stroe, Daniel-Ioan

AU - Teodorescu, Remus

PY - 2020

Y1 - 2020

N2 - Accurate knowledge of the state of health (SOH) is important for battery management system to maintain safe operation of batteries and extend their lifetime. In order to improve the accuracy in estimating the SOH of a battery, a new method based on empirical mode decomposition sample entropy (EMDSE) and support vector machine (SVM) is proposed in this paper. Compared with the tradition SE-based method, EMD is used to filter the noise of the original signal. Then the EMDSE as a new feature is used for the SVM training, and the potential relationship between the SOH and the EMDSE is established. Finally, the effectiveness of the proposed method is verified by experimental results.

AB - Accurate knowledge of the state of health (SOH) is important for battery management system to maintain safe operation of batteries and extend their lifetime. In order to improve the accuracy in estimating the SOH of a battery, a new method based on empirical mode decomposition sample entropy (EMDSE) and support vector machine (SVM) is proposed in this paper. Compared with the tradition SE-based method, EMD is used to filter the noise of the original signal. Then the EMDSE as a new feature is used for the SVM training, and the potential relationship between the SOH and the EMDSE is established. Finally, the effectiveness of the proposed method is verified by experimental results.

KW - Lithium-Ion Battery

KW - State of Health

KW - Sample Entropy

KW - Empirical Mode Decomposition

M3 - Article in proceeding

BT - 2020 IEEE Applied Power Electronics Conference and Exposition (APEC)

ER -