Open access
Aug 2026
Physics-Informed Transfer Learning for Cross-Condition State of Health Prediction of Lithium-Ion Batteries
This study proposes a physics-informed transfer learning (PITL) framework that combines a common health descriptor space, parameter function decomposition, and PINN–DeepHPM regularization, and partial freezing can lower adaptation cost, but its accuracy benefit is target dependent.
Jing-Han Ma, Junjun Wang, Changru Rong et al.
· Electronics · 0 citations