Skip to main navigation Skip to search Skip to main content

Battery Abnormal Diagnosis of Electric Loader Based on Data Mining

  • Qiang Zhang*
  • , Muxin He
  • , Ni Lin
  • *Corresponding author for this work
  • Ltd
  • Beijing Institute of Technology

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

Abstract

In order to address battery degradation and inconsistency in electric loaders, this study proposes a data-driven fault diagnosis method based on cloud platform data. The approach combines t-SNE for dimensionality reduction, K-means for clustering, and a 3σ-MSS multi-level screening strategy to identify abnormal battery cells. Short-term and long-term deviation analyses were conducted. Results show that the proportion of deviated voltage is below 6%, which can be mitigated via balancing charging. The method supports early fault warning and maintenance intervention.

Original languageEnglish
Title of host publication2025 IEEE 8th International Conference on Automation, Electronics and Electrical Engineering, AUTEEE 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages150-153
Number of pages4
ISBN (Electronic)9798331599072
DOIs
Publication statusPublished - 2025
Externally publishedYes
Event8th IEEE International Conference on Automation, Electronics and Electrical Engineering, AUTEEE 2025 - Shenyang, China
Duration: 28 Nov 202530 Nov 2025

Publication series

Name2025 IEEE 8th International Conference on Automation, Electronics and Electrical Engineering, AUTEEE 2025

Conference

Conference8th IEEE International Conference on Automation, Electronics and Electrical Engineering, AUTEEE 2025
Country/TerritoryChina
CityShenyang
Period28/11/2530/11/25

Keywords

  • anomaly detection
  • electric loader
  • fault diagnosis
  • power battery

Fingerprint

Dive into the research topics of 'Battery Abnormal Diagnosis of Electric Loader Based on Data Mining'. Together they form a unique fingerprint.

Cite this