一种基于物理信息网络的电池快充动态安全边界识别方法
By using the physical information network architecture of TCN-FCNN-SG, the problem of non-destructive and online lithium plating identification in the vehicle battery management system is solved, and accurate identification of lithium plating boundaries and formulation of safe fast charging strategies are realized under complex working conditions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JILIN UNIVERSITY
- Filing Date
- 2026-05-09
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to achieve non-destructive, online identification of lithium plating behavior inside batteries under complex automotive conditions. Traditional methods suffer from strong interference and insufficient robustness, failing to meet the real-time early warning requirements of automotive battery management systems.
A Physical Information Network (PINN) based on a fusion architecture of Temporal Convolutional Network (TCN) and Fully Connected Neural Network (FCNN), combined with a soft gate component (SG), is used to construct a lithium-ion battery fast-charging lithium plating model, generating a dataset containing high-fidelity internal lithium plating features, thereby achieving non-destructive early identification of lithium plating.
Without adding new hardware sensors, it accurately captures the start and end boundaries of lithium plating, reduces reliance on implicit physical tags, provides dynamic safety boundary recognition, and supports the vehicle BMS in formulating adaptive safe fast charging strategies.
Smart Images
Figure CN122154500B_ABST