| Autor: |
Yunhong Che, Xin Sui, Pallavi Bharadwaj, Daniel-Ioan Stroe, Remus Teodorescu |
| Rok vydání: |
2022 |
| Předmět: |
|
| Zdroj: |
Che, Y, Sui, X, Bharadwaj, P, Stroe, D I & Teodorescu, R 2022, Battery Lifetime Prediction and Degradation Reconstruction based on Probabilistic Convolutional Neural Network . in IEEE 13th International Symposium on Power Electronics for Distributed Generation Systems (PEDG) . IEEE, IEEE International Symposium on Power Electronics for Distributed Generation Systems (PEDG), 13th IEEE International Symposium on Power Electronics for Distributed Generation Systems, PEDG 2022, Kiel, Germany, 26/06/2022 . https://doi.org/10.1109/PEDG54999.2022.9923247 |
| DOI: |
10.1109/pedg54999.2022.9923247 |
| Popis: |
Capacity degradation of lithium-ion batteries influences their service abilities as energy storage systems. Lifetime prediction, historical degradation curve trajectory, and capacity estimation help prognose the long-term serviceability and timely health status of batteries, which guides early maintenance and intelligent management. This paper proposed a novel method to predict the lifetime, reconstruct the historical degradation curve and estimate the capacity via excavating the information hindered under partially charged capacity and temperature curves. Only partial raw data of temperature and charged capacity are needed for modeling without manual feature engineering. The convolutional neural network is used to build the lifetime prediction and capacity estimation models. In addition, the probabilistic regression is added to the establishment of the capacity estimation model, which could provide the probabilistic estimation results. Finally, transfer learning is adopted to update the model with a few available data of testing batteries. Results show that the predictions are accurate and reliable. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
