この技術により、電池の開発コストを削減することができる。 Technique could reduce costs of battery development.
2022-05-05 アルゴンヌ国立研究所(ANL)
アルゴンヌ国立研究所で収集した6種類の電池化学を代表する300個の電池の実験データを使用して、さまざまな電池がどれくらいの時間サイクルし続けるかを判断することができるようになりました。
研究代表のポールソンは言いました。「この研究の価値は、異なる電池がどのように機能するかを特徴付ける信号を与えてくれたことです。我々ができることの一つは、既知の化学物質でアルゴリズムを訓練し、未知の化学物質で予測をさせることです 。この分野での更なる研究は、リチウムイオン電池の未来を導く可能性がある。」
＜関連情報＞
- https://www.anl.gov/article/researchers-now-able-to-predict-battery-lifetimes-with-machine-learning
- https://www.sciencedirect.com/science/article/abs/pii/S0378775322001495?via%3Dihub
機械学習による電池寿命の早期予測に向けた特徴量エンジニアリング Feature engineering for machine learning enabled early prediction of battery lifetime
Noah H.Paulson,Joseph Kubal,Logan Ward,Saurabh Saxena,Wenquan Lu,Susan J.Babinec
Journal of Power Sources Published:25 February 2022
DOI:https://doi.org/10.1016/j.jpowsour.2022.231127
Highlights
- Unique Li-ion dataset comprised 6 metal oxide cathode chemistries and 300 batteries.
- A single machine learning model accurately predicted cycle life across cathodes.
- Useful predictions required as few as one preliminary cycle.
- Broad feature sets are most accurate; categorically narrow sets are viable.
- A diverse training set improved predictive performance for new cathode chemistries.
Abstract
Accurate battery lifetime estimates enable accelerated design of novel battery materials and determination of optimal use protocols for longevity in deployments. Unfortunately, traditional battery testing may take years to reach thousands of cycles. Recent studies have shown that machine learning (ML) tools can predict lithium-ion battery lifetimes from 100 or fewer preliminary cycles, representing only a few weeks of cycling. Until now, conclusions about the efficacy and broad applicability of these predictions across a variety of cathode chemistries have been limited by available experimental information. In this work, we leverage a battery cycling dataset representing six cathode chemistries (NMC111, NMC532, NMC622, NMC811, HE5050, and 5Vspinel), multiple electrolyte/anode compositions, and 300 total carefully prepared pouch batteries to explore feature selection and battery chemistry’s role in ML battery lifetime predictions. A mean absolute error (MAE) of 78 cycles in prediction was seen for a chemistry-spanning test set from 100 preliminary cycles. Furthermore, an MAE of 103 cycles was seen when using only the first cycle. This study represents an in-depth investigation of strategies for feature selection for battery lifetime prediction, ML models’ generalization across multiple battery chemistries, and predictions beyond the training set in the chemical space.
