A novel hybrid model for lithium-ion batteries lifespan prediction with high accuracy and interpretability. (May 2023)

• Record Type: Journal Article
• Authors: Pang, Xiaoxian; Yang, Wei; Wang, Chengyun; Fan, Haosen; Wang, Le; Li, Junhao; Zhong, Shi; Zheng, Wenzhi; Zou, Hanbo; Chen, Shengzhou; Liu, Quanbing
• Is Part Of: Journal of energy storage. Volume 61(2023)
• Publication Date: 2023-05
• Subjects: Lithium-ion battery -- Machine learning -- Early prediction -- Remaining useful life -- Model interpretability; Energy storage -- Periodicals; Energy storage -- Research -- Periodicals; 621.3126
• DOI: 10.1016/j.est.2023.106728
• Languages: English
• ISSNs: 2352-152X

Abstract

Abstract: Machine learning can accurately predict the remaining useful life (RUL) of lithium-ion batteries because of its strong learning ability, efficient computing efficiency, and high accuracy. However, the prediction behavior and principle of many data-driven models as black box functions are unknown, and the potential of high accurate prediction requires further investigation. In view of these research gaps, this study proposes a novel hybrid model based on adaptive feature separable convolution (AFSC) and convolutional long short-term memory (ConvLSTM) network to improve the accuracy of RUL prediction and the interpretability of the model. The model extracts aging features from charging process data and can be applied to both early prediction and RUL prediction. Validation based on 124 commercial lithium iron phosphate battery aging data shows that the mean absolute error (MAE) of the early prediction results using the first 20 cycles is only 7 cycles, while the MAE of the RUL prediction is 0.12 cycles, both demonstrating excellent performance. In addition, the feature processing and prediction process of the model is analyzed through the visualization of upsampling and attention weights. Highlights: A new hybrid model for both battery life early prediction and RUL prediction is proposed. Used only the first 20 cycles data to achieve the early prediction with a MAE of 7 cycles. The MAE of the best RUL prediction is 0.12 cycles. Embedded two attention mechanisms reduces the error by 1.3–6 times. Combined the advantages of the two subnets, the error is reduced by 1.2–9. 7 and 1.3–7.3 times. Defined a contribution index of the original data points to the model and improved the interpretability.