A flexible battery capacity estimation method using a portion of charging curve
- Resource Type
- Conference
- Authors
- Cao, Mengda; Zhang, Tao; Wang, Yu; Zhu, Wenkai; Liu, Yajie
- Source
- 2022 Global Reliability and Prognostics and Health Management (PHM-Yantai) Reliability and Prognostics and Health Management (PHM-Yantai),2022 Global. :1-5 Oct, 2022
- Subject
- Aerospace
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
General Topics for Engineers
Nuclear Engineering
Power, Energy and Industry Applications
Robotics and Control Systems
Transportation
Degradation
Estimation
Voltage
Market research
Robustness
Batteries
Bayes methods
Capacity estimation
sliding voltage window (SLW)
Bayesian Optimization (BO)
charging curve
- Language
Accurate capacity estimation is crucial for the battery management system (BMS) on the electrical vehicle. In this work, we utilize a deep learning structure to estimate the battery capacity with a portion sequence of the charging curve. A sliding voltage window of length 300mV with 10 mV a step is used to extract the corresponding discrete battery capacity, which is both selected as the input of the model. The model is further optimized by the Bayesian Optimization algorithm. The accuracy and robustness of the proposed model are verified on the Oxford battery dataset that it well captures the degradation trend of batteries with its RMSE less than 0.041Ah.