.png){kind=icon}
.png){kind=icon}
Return
A physics-informed adaptive forgetting mechanism with state-of-charge variation for low-temperature battery model identification
.png){kind=icon}
PRE
AI
Z
B
X
L
W
X
H
W
DOI：10.1007/s10800-026-02472-9.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Parameter identification for equivalent circuit models of lithium-ion batteries deteriorated at low temperatures. This decline was driven by intensified and rapidly shifting polarization dynamics, which were governed by the Arrhenius law. Conventional recursive algorithms, with their fixed forgetting factors, lacked the physicochemical basis to track these dynamics effectively. This paper introduced a physicochemically-informed adaptive mechanism that directly linked the algorithm's forgetting factor to the battery's internal state via the state-of-charge variation. The proposed low-temperature optimized identification method thereby embedded the battery's dynamics into the identification core. Benchmarking against second-order exponential fitting, a genetic algorithm, and standard forgetting factor recursive least squares on hybrid pulse power characterization and urban dynamometer driving schedule data from 0 to - 20 degrees C showed that the proposed method achieved the lowest mean absolute error and root mean square error under all conditions. Validation at extreme temperatures as low as - 30 degrees C confirmed its robustness, offering a reliable algorithmic solution for enhanced battery management system performance in wide-range low-temperature environments.
Keywords:
Low temperature
Lithium-ion battery
Forgetting factor
Parameter identification
State-of-charge variation
Journal
IF:
3
Papers:
963
Citations:
9.0K
Organization
Cited Papers