.png){kind=icon}
.png){kind=icon}
Return
Multi-Grouping-Compatible Frame-Level Unsupervised Fault Diagnosis and Localization for Electric Vehicle Battery Packs in Realistic Conditions
.png){kind=icon}
PRE
AI
R
Y
J
F
H
Y
L
J
X
S
DOI：10.1016/j.etran.2026.100579.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Real-time battery fault diagnosis with faulty cell location in electric vehicles • Fault diagnosis method based on an Attention-Gated Recurrent Unit-Variational Autoencoder- StatFusion neural network enables frame-level diagnosis • Compatible across multiple pack designs without retraining for grouping changes • Fault analysis covers electrolyte leakage, connection anomalies, excessive aging, and internal short circuits. • Outperforms common methods on 500 vehicles with higher true positive rate
Keywords:
battery fault diagnosis
electric vehicle battery packs
attention-gated recurrent unit
variational autoencoder
real-time localization
Journal
IF:
17
Papers:
547
Citations:
4.4K
Organization
Cited Papers