| 초록 |
Lithium-ion batteries have been used throughout the industry. For safety and performance, it is important to predict battery life accurately. In this study, we propose an efficient feature selection strategy for a data-driven battery life prediction model using active learning (AL) algorithm. The expected improvement in the prediction accuracy of the trained neural networks with a feature set was set as the objective function for AL. Gaussian process regression(GPR) is used to predict the approximate value of the function in the neighborhood. Among 4950 possible feature sets, we select the best feature set only after exploring 84 feature sets. This work is supported by the Korea Agency for Infrastructure Technology Advancement(KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (No. 22ATOG-C162087-02), and by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (NRF-2021R1C1C1004217). |
