A SOH online estimation method and device, electronic equipment and storage medium
By combining real-time data acquisition and preprocessing with multi-mode stationary point identification and OCV-SOC relationship calibration, and by combining cross-day state continuation and physical rule calculation of capacity attenuation, the SOH value is predicted using a random forest model. This solves the problems of low data utilization and poor reliability of traditional methods under frequency modulation conditions, and achieves high-precision SOH estimation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 深圳织算科技有限公司
- Filing Date
- 2026-04-07
- Publication Date
- 2026-07-21
AI Technical Summary
Traditional SOH estimation methods suffer from low data utilization and high estimation failure rate under complex frequency modulation conditions, making them unsuitable for high-frequency dynamic conditions. Furthermore, data-driven methods lacking physical constraints have poor interpretability.
By preprocessing real-time battery data, using a multi-mode resting point identification mechanism and dynamic calibration of the OCV-SOC relationship curve, and combining a cross-day state continuation mechanism to match and screen resting point combinations, capacity decay is calculated based on physical rules, and SOH values are predicted by training a random forest model. Finally, SOH values are obtained through adaptive weighted fusion or decay weight filtering.
It significantly improves the accuracy and reliability of SOH estimation, adapts to complex working conditions, and enhances data utilization and engineering applicability.
Smart Images

Figure CN121978573B_ABST