Research papers · 2022
Regularization-based Continual Learning for Fault Prediction in Lithium-Ion Batteries
Regularization-based Continual Learning for Fault Prediction in Lithium-Ion Batteries
Overview
The study compares regularization-based continual learning for battery fault prediction under changing tasks. Evaluation on real battery-wear data examines online elastic weight consolidation and other methods, highlighting sensitivity to task characteristics and ordering.
Author-supplied source information
Author-supplied information on arXiv.
6 pages, 5 figures, 4 tables. Accepted at CIRP ICME 2021. arXiv admin note: text overlap with arXiv:2101.00509
Version: arXiv v1 — manuscript revision