.png){kind=icon}
.png){kind=icon}
Return
Empowering Knowledge-Guided Intelligence for Accurate and Robust Few-Shot Lithium-Ion Battery State of Health Estimation
.png){kind=icon}
PRE
AI
R
K
S
D
J
J
S
H
W
C
DOI：10.1016/j.etran.2026.100582.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A knowledge-guided few-shot learning method is proposed for battery SOH estimation. • Pseudo labels are extracted combining both human expertise and machine learning. • A label propagation strategy is developed to enhance pseudo label availability. • A two-stage training scheme is developed for knowledge transfer. • An MAE of less than 1.3% is achieved, even with 2% of labelled training samples.
Keywords:
Few-shot learning
State of Health estimation
Knowledge-guided intelligence
Label propagation
Battery health monitoring
Journal
IF:
17
Papers:
547
Citations:
4.4K
Organization
Cited Papers