摘要
由於鋰離子電池管理系統在工業應用中的廣泛使用，我們需要準確預測電池的剩餘使用壽命(Remaing useful life, RUL)來避免設備的損壞。電池的使用壽命隨著多次的充電與放電會導致電池的內阻增加和容量衰減，電池的續航力因而越來越差。我們需要可靠且精準的電池健康診斷來及時的對電池系統進行維護與更換。然而，電池的健康狀態與剩餘使用壽命要如何預測準確是我們最大瓶頸。此研究使用雙向長短期記憶搭配注意力機制的模型(Attention based bidirectional long short-term memory, Bi-LSTM-AT)進行線上的剩餘使用壽命預測，來實現更好的健康管理。此論文所提出的模型中將歸一化後的容量(Capacity)視為電池健康狀態(State of health, SOH)。本研究使用支持向量回歸模型(Support vector regression, SVR)獲得溫度的未來多步預測，且利用預測的未來溫度資料來更新線上的數據集，最後使用Bi-LSTM-AT模型預測電池的健康狀態。
在眾多鋰離子電池的資料集中，此論文使用Toyota 的鋰離子電池資料集來評估所提出的模型性能，再和一些現有的模型比較預測結果。我們相信有良好的溫度感知器以及訓練良好的SVR模型可以使此研究方法有更精準的預測結果。

Abstract
An accurate prediction of the remaining useful lifetime (RUL) in a battery management system is required due to its wide use in industrial applications. Side reactions make batteries suffer for continuous performance degradation throughout their service life in terms of capacity fades and internal resistance increases. A reliable and accurate battery health diagnostic is required for timely maintenance and replacement of battery systems. Nevertheless, the accuracy is the biggest bottleneck for battery state of health and RUL prediction. We propose a novel methodology for online RUL prediction using a bidirectional long short-term memory with an attention mechanism (Bi-LSTM-AT) model for better health management. The normalized capacity is used as a state of health (SOH) in the proposed model and a multi-step ahead forecast is obtained for the average temperature by using the support vector regression (SVR) model. The initial data set is updated online Based on the SVR model output, then the Bi-LSTM-AT model is used to predict the SOH of the battery. In cylindrical prismatic cell, thirteen batteries will be used in this study. Two batteries are used to evaluate the performance of the proposed model and some existing models and the additional eleven batteries’ RULs are predicted by adopting the purposed model.

References
[1] Wu J, Zhang C, Chen Z. An online method for lithium-ion battery remaining useful estimation using importance sampling and neural networks. Appl Energy 2016;173:134-40. https://doi.org/10.1016/j.apenergy.2016.04.057.
[2] Goebel K, Saha B, Saxena A, Celaya J, Christophersen J. Prognostics in battery health management. IEEE Instrum Meas Mag 2008;11(4):33-40. https://doi.org/10.1109/MIM.2008.4579269.
[3] Ali MU, Zafar A, Nengroo SH, Hussain S, Park GS, Kim HJ. Online remaining useful life prediction for lithium-ion batteries using partial discharge data features. Energies 2019;12(22):4366. https://doi.org/10.3390/en12224366.
[4] Chen L, Wang H, Chen J, An J, Ji B, Lyu Z, Cao W, Pan H. A novel remaining useful life prediction framework for lithium‐ion battery using grey model and particle filtering. Int J Energy Res 2020;44(9):7435-49. https://doi.org/10.1002/er.5464.
[5] Gou B, Xu Y, Feng X. State-of-health estimation and remaining-useful-life prediction for lithium-ion battery using a hybrid data-driven method. IEEE Trans Veh Technol 2020;69(10):10854-67. https://doi.org/10.1109/TVT.2020.3014932.
[6] Xu X, Yu C, Tang S, Sun X, Si X, Wu L. Remaining useful life prediction of lithium-ion batteries based on Wiener processes with considering the relaxation effect. Energies 2019;12(9):1685. https://doi.org/10.3390/en12091685.
[7] Wu Y, Li W, Wang Y, Zhang K. Remaining useful life prediction of lithium-ion batteries using neural network and bat-based particle filter. IEEE Access 2019;7:54843-54. https://doi.org/10.1109/ACCESS.2019.2913163.
[8] Jia J, Liang J, Shi Y, Wen J, Pang X, Zeng J. SOH and RUL prediction of lithium-ion batteries based on Gaussian process regression with indirect health indicators. Energies 2020;13(2):375. https://doi.org/10.3390/en13020375.
[9] Liu Y, Zhao G, Peng X. Deep learning prognostics for lithium-ion battery based on ensembled long short-term memory networks. IEEE Access 2019;7:155130-42. https://doi.org/10.1109/ACCESS.2019.2937798.
[10] Ungurean L, Micea MV, Cârstoiu G. Online state of health prediction method for lithium‐ion batteries, based on gated recurrent unit neural networks. Int J Energy Res 2020;44(8):6767-77. https://doi.org/10.1002/er.5413.
[11] Li Y, Li K, Liu X, Wang Y, Zhang L. Lithium-ion battery capacity estimation - A pruned convolutional neural network approach assisted with transfer learning. Appl Energy 2021;285:116410. https://doi.org/10.1016/j.apenergy.2020.116410.
[12] Li X, Zhang L, Wang Z, Dong P. Remaining useful life prediction for lithium-ion batteries based on a hybrid model combining the long short-term memory and Elman neural networks. J. Energy Storage 2019;21:510-8. https://doi.org/10.1016/j.est.2018.12.011.
[13] Peng W, Ye Z-S, Chen N. Joint online RUL prediction for multivariate deteriorating systems. IEEE Trans Ind Informatics 2019;15(5):2870-8. https://doi.org/10.1109/TII.2018.2869429.
[14] Guha A, Patra A. Online estimation of the electrochemical impedance spectrum and remaining useful life of lithium-ion batteries. IEEE Trans Instrum Meas 2018;67(8):1836-49. https://doi.org/10.1109/TIM.2018.2809138.
[15] Dong H, Jin X, Lou Y, Wang C. Lithium-ion battery state of health monitoring and remaining useful life prediction based on support vector regression-particle filter. J. Power Sources 2014;271:114-23. https://doi.org/10.1016/j.jpowsour.2014.07.176.
[16] Tao T, Zio E, Zhao W. A novel support vector regression method for online reliability prediction under multi-state varying operating conditions. Reliab Eng Syst Saf 2018;177:35-49. https://doi.org/10.1016/j.ress.2018.04.027.
[17] Zhao Q, Qin X, Zhao H, Feng W. A novel prediction method based on the support vector regression for the remaining useful life of lithium-ion batteries. Microelectron Reliab 2018;85:99-108. https://doi.org/10.1016/j.microrel.2018.04.007.
[18] Li X, Shu X, Yan W, Shen J, Xiao R, Yan W, Chen Z. An on-board remaining useful life estimation algorithm for lithium-ion batteries of electric vehicles. Energies 2017;10(5):691. https://doi.org/10.3390/en10050691.
[19] Shen D, Wu L, Kang G, Guan Y, Peng Z. A novel online method for predicting the remaining useful life of lithium-ion batteries considering random variable discharge current. Energy 2021;218:119490. https://doi.org/10.1016/j.energy.2020.119490.
[20] Ge M-F, Liu Y, Jiang X, Liu J. A review on state of health estimations and remaining useful life prognostics of lithium-ion batteries. Measurement 2021;174:109057. https://doi.org/10.1016/j.measurement.2021.109057
[21] Li Y, Zou C, Berecibar M, Nanini-Maury E, Chan JC, van den Bossche P, Van Mierlo J, Omar N. Random forest regression for online capacity estimation of lithium-ion batteries, Appl. Energy 2018;232:197–210. https://doi.org/10.1016/j.apenergy.2018.09.182
[22] Yang D, Zhang X, Pan R, Wang Y, Chen Z. A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curve. J. Power Sources 2018;384:387–95. https://doi.org/10.1016/j.jpowsour.2018.03.015.
[23] Severson KA, Attia PM, Jin N, Perkins N, Jiang B, Yang Z et al. Data-driven prediction of battery cycle life before capacity degradation. Nat Energy 2019;4(5):383-91. https://doi.org/ 10.1038/s41560-019-0356-8.
[24] Vapnik VN. (2000). The nature of statistical learning theory (2nd ed.). Springer. https://doi.org/10.1007/978-1-4757-3264-1.
[25] Storn R, Price K. Differential evolution - a simple and efficient heuristic for global optimization over continuous spaces. J. Glob Optim 1997;11:341-59. https://doi.org/10.1023/A:1008202821328.
[26] Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput 1997;9(8):1735-80. https://doi.org/10.1162/neco.1997.9.8.1735.
[27] Graves A, Schmidhuber J. Framewise phoneme classification with bidirectional LSTM and other neural network architectures. Neural Networks 2005;18(5-6):602-10. https://doi.org/10.1016/j.neunet.2005.06.042.
[28] Schuster M, Paliwal KK. Bidirectional recurrent neural networks. IEEE Trans Signal Process 1997;45(11):2673-81. https://doi.org/10.1109/78.650093.
[29] Xia T, Song Y, Zheng Y, Pan E, Xi L. An ensemble framework based on convolutional bi-directional LSTM with multiple time windows for remaining useful life estimation. Comput Ind 2020;115:103182. https://doi.org/10.1016/j.compind.2019.103182.
[30] Wang J, Wen G, Yang S, Liu Y. Remaining useful life estimation in prognostics using deep bidirectional LSTM neural network. In: Prognostics and system health management conference (PHM-Chongqing) 2018;p.1037-42. https://doi.org/10.1109/PHM-Chongqing.2018.00184.
[31] Bahdanau D, Cho K, Bengio Y. Neural machine translation by jointly learning to align and translate. 2015, arXiv preprint. https://arxiv.org/abs/1409.0473.
[32] Raffel C, Ellis DPW. Feed-forward networks with attention can solve some long-term memory problems. 2016, arXiv preprint. https://arxiv.org/abs/1512.08756.
[33] Liu G, Guo J. Bidirectional LSTM with attention mechanism and convolutional layer for text classification. Neurocomputing 2019;337:325-38. https://doi.org/10.1016/j.neucom.2019.01.078.
[34] Wang F-K, MamoT, Cheng X-B. Bi-directional long short-term memory recurrent neural network with attention for stack voltage degradation from proton exchange membrane fuel cells. J. Power Sources 2020;461:228170. https://doi.org/10.1016/j.jpowsour.2020.228170.
[35] Li W, Qi F, Tang M, Yu Z. Bidirectional LSTM with self-attention mechanism and multi-channel features for sentiment classification. Neurocomputing 2020;387:63-77. https://doi.org/10.1016/j.neucom.2020.01.006.