一种考虑电池健康状态的荷电状态与健康状态联合预测方法
By extracting the charging and discharging characteristics of lithium-ion batteries and combining correlation analysis and attention mechanisms, a learnable basis time series model and an adaptive filter are used for joint prediction. This solves the problem that the coupling relationship between the state of charge and the state of health in lithium-ion battery state estimation is not fully considered, and achieves high-precision and robust joint prediction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUAIYIN INSTITUTE OF TECHNOLOGY
- Filing Date
- 2026-06-22
- Publication Date
- 2026-07-17
AI Technical Summary
Existing lithium-ion battery state estimation methods fail to fully consider the coupling relationship between state of charge and state of health, resulting in insufficient prediction accuracy and robustness under complex operating conditions. Furthermore, deep learning models are time-consuming to train and overly dependent on sample data.
By extracting voltage, current, temperature, and time series features during battery charging and discharging, and combining correlation analysis and parallel block perception attention mechanism to screen health features, a learnable basis time series model and an adaptive unscented Kalman filter are used for joint prediction, and a second-order RC equivalent circuit model is used for dynamic correction. Chaotic mapping is used to optimize key parameters.
It improves the accuracy and robustness of the joint prediction of state of charge and health, reduces drift error under complex operating conditions, and enhances engineering applicability.
Smart Images
Figure CN122410331A_ABST