Skip to main navigation Skip to search Skip to main content

LSTM Model-Based Fault Detection for Electric Vehicle’s Battery Packs

  • Silesian University of Technology
  • NGK Ceramics Polska Sp. z o.o.

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

1 Citation (Scopus)

Abstract

An LSTM model-based fault detection method dedicated for a liquid leakage and liquid intrusion detection system was proposed and described. The method utilized two individual residual evaluation approaches. The first based on statistical analysis and the second based on model error modelling methodology. Both approaches were trained and tested using collected datasets of a prototyped laboratory stand, simulating liquid intrusion and liquid leakage faults of an electric vehicle’s battery pack with direct liquid cooling battery thermal management system. Obtained results were compared and discussed in details and have revealed notably higher robustness of one of the proposed approaches.

Original languageEnglish
Title of host publicationIntelligent and Safe Computer Systems in Control and Diagnostics
EditorsZdzislaw Kowalczuk
PublisherSpringer Science and Business Media Deutschland GmbH
Pages217-229
Number of pages13
ISBN (Print)9783031161582
DOIs
Publication statusPublished - 2023
Event15th International Conference on Diagnostics of Processes and Systems, DPS 2022 - Chmielno, Poland
Duration: 5 Sept 20227 Sept 2022

Publication series

NameLecture Notes in Networks and Systems
Volume545 LNNS
ISSN (Print)2367-3370
ISSN (Electronic)2367-3389

Conference

Conference15th International Conference on Diagnostics of Processes and Systems, DPS 2022
Country/TerritoryPoland
CityChmielno
Period5/09/227/09/22

Keywords

  • Battery pack
  • Electric vehicle
  • Liquid intrusion
  • Liquid leakage
  • Long short-term memory model
  • Model-based fault detection

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Signal Processing
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'LSTM Model-Based Fault Detection for Electric Vehicle’s Battery Packs'. Together they form a unique fingerprint.

Cite this