一种用于储能电站的单体电池的故障预警方法及系统
By using time-frequency domain feature extraction, relaxation analysis, and electrochemical mechanism training, a fault early warning network was developed, which solved the problems of accuracy and efficiency in early warning of single-cell battery faults in energy storage power stations, and achieved efficient and reliable fault identification and early warning.
Patent Information
- Application Number
- CN202610534132.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-22
- Publication Date
- 2026-07-17
- Estimated Expiration
- 2046-04-22
AI Technical Summary
Existing single-cell fault early warning technologies for energy storage power stations rely on a single operating parameter and fail to incorporate the battery's electrochemical response characteristics, resulting in serious issues of missed and false alarms. Furthermore, the models have weak generalization capabilities and cannot meet the requirements for long-term, high-safety-level operation.
By extracting time-frequency domain features, relaxing analysis, capacity differential benchmark verification and tensor coupling, multidimensional health features are constructed, and an early warning network is trained in conjunction with electrochemical mechanisms to improve the accuracy and efficiency of fault early warning.
It significantly improves the ability to identify fault signals, reduces missed and false alarms, enhances the generalization ability of the model, improves the efficiency and accuracy of fault early warning, and provides reliable protection for the safe and stable operation of energy storage power stations.
Smart Images
Figure CN122063458B_ABST
Abstract
Citation Information
Patent Citations
Energy storage battery fault diagnosis method and system based on data fusion algorithm
CN121049760A
Battery failure reason prediction method and computer storage medium
CN121721494A