Prediction of remaining useful life of battery cell using logistic regression based on strong tracking particle filter

Zhenbao Liu, Dasen Fan, Shuhui Bu, Chao Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

8 Scopus citations

Abstract

The RUL prediction of battery is an effective approach to improve the battery reliability and service life. This paper proposes a novel evaluation algorithm of battery states which is named logistic regression based on strong tracking particle filter for battery RUL prediction. The core idea of this algorithm is to approximate the non-linear and non-Gaussian process of state update of battery RUL prediction through logistic regression combining least square support vector machine. There are two main contributions: first, we combine logistic regression with least square support vector machine for RUL estimation; second, we introduce logistic regression with particle update by a strong tracking particle filter.

Original languageEnglish
Title of host publication2015 IEEE Conference on Prognostics and Health Management
Subtitle of host publicationEnhancing Safety, Efficiency, Availability, and Effectiveness of Systems Through PHAf Technology and Application, PHM 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781479918935
DOIs
StatePublished - 8 Sep 2015
EventIEEE Conference on Prognostics and Health Management, PHM 2015 - Austin, United States
Duration: 22 Jun 201525 Jun 2015

Publication series

Name2015 IEEE Conference on Prognostics and Health Management: Enhancing Safety, Efficiency, Availability, and Effectiveness of Systems Through PHAf Technology and Application, PHM 2015

Conference

ConferenceIEEE Conference on Prognostics and Health Management, PHM 2015
Country/TerritoryUnited States
CityAustin
Period22/06/1525/06/15

Keywords

  • Batteries
  • Least squares approximations
  • Logistics
  • Mathematical model
  • Prediction algorithms
  • Predictive models
  • Support vector machines

Fingerprint

Dive into the research topics of 'Prediction of remaining useful life of battery cell using logistic regression based on strong tracking particle filter'. Together they form a unique fingerprint.

Cite this