Smart Media Journal (스마트미디어저널)

THE KOREAN INSTITUTE OF SMART MEDIA (한국스마트미디어학회)
권호 표지
DOI QR코드

DOI QR Code

Improved SOH Prediction Model for Lithium-ion Battery Using Charging Characteristics and Attention-Based LSTM

충전 특성과 어텐션 기반 LSTM을 활용한 개선된 리튬이온 배터리 SOH 예측 모델

  • 류한일 (전남대학교 인공지능융합학과) ;
  • 이상훈 (전남대학교 정보보안협동과정) ;
  • 최덕재 (전남대학교 소프트웨어공학과) ;
  • 박혁로 (전남대학교 소프트웨어공학과)
  • Received : 2023.10.15
  • Accepted : 2023.11.20
  • Published : 2023.12.29
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, the need to prevent battery fires and accidents has emerged, as the use of lithium-ion batteries has increased. In order to prevent accidents, it is necessary to predict the state of health (SOH) and check the replacement timing of the battery with a lot of degradation. This paper proposes a model for predicting the degradation state of a battery by using four battery degradation indicators: maximum voltage arrival time, current change time, maximum temperature arrival time, and incremental capacity (IC) that can be obtained in the battery charging process, and LSTM using an attention mechanism. The performance of the proposed model was measured using the NASA battery data set, and the predictive performance was improved compared to that of the general LSTM model, especially in the SOH 90-70% section, which is close to the battery replacement cycle.

최근 리튬이온 배터리 사용이 늘어남에 따라 배터리 화재 및 사고 예방의 필요성이 대두되고 있다. 사고 예방을 위해서는 배터리 SOH(State of Health)를 예측하여 열화가 많이 진행된 배터리의 교체 시기를 확인하는 것이 필요하다. 본 논문에서는 배터리의 충전 과정에서 얻을 수 있는 최대 전압 도달 시간, 전류 변화 시간, 최대 온도 도달 시간, IC(Incremental Capacity) 등 4가지 배터리 열화 특성과 어텐션 메커니즘을 이용한 장단기 메모리(Long Short Term Memory, LSTM)를 사용하여 배터리의 열화 상태를 예측하는 모델을 제안한다. NASA에서 제공하는 배터리 데이터 세트를 사용해 제안하는 모델의 성능을 측정한 결과 일반적인 LSTM 모델을 사용하는 경우보다 예측성능의 개선을 확인할 수 있었고, 특히 배터리 교체 주기에 가까운 SOH 90-70% 구간에서 더 우수한 성능을 보였다.

Keywords

Acknowledgement

본 연구는 과학기술정보통신부 및 정보통신기획평가원의 지역지능화혁신인재양성사업의 연구결과로 수행되었음 (IITP-2023-RS-2022-00156287). 본 연구는 과학기술정보통신부 및 정보통신기획평가원의 인공지능융합혁신인재양성사업의 연구결과로 수행되었음(IITP-2023-RS-2023-00256629)

References

  1. H. Dai, G. Zhao, M. Lin, J. Wu and G. Zheng, "A Novel Estimation Method for the State of Health of Lithium-Ion Battery Using Prior Knowledge-Based Neural Network and Markov Chain," IEEE Transactions on Industrial Electronics, vol. 66, no. 10, pp. 7706-7716, Nov. 2019.
  2. Z. Guo, X. Qiu, G. Hou, B. Y. Liaw and C. Zhang, "State of health estimation for lithium ion batteries based on charging curves," Journal of Power Sources, vol. 249, pp. 457-462, Mar. 2014. https://doi.org/10.1016/j.jpowsour.2013.10.114
  3. Y. Li, Z. Wei, B. Xiong, and D. Vilathgamuwa, "Adaptive Ensemble-Based Electrochemical-Thermal Degradation State Estimation of Lithium-Ion Batteries," IEEE Transactions on Industrial Electronics, vol. 69, no. 7, pp. 6984-6996, Jul. 2022. https://doi.org/10.1109/TIE.2021.3095815
  4. X. Hu, D. Cao and B. Egardt, "Condition Monitoring in Advanced Battery Management Systems: Moving Horizon Estimation Using a Reduced Electrochemical Model," IEEE/ASME Transactions on Mechatronics, vol. 23, no. 1, pp. 167-178, Feb. 2017. https://doi.org/10.1109/TMECH.2017.2675920
  5. Z. Chen, M. Sun, X. Shu, R. Xiao and J. Shen, "Online State of Health Estimation for Lithium-Ion Batteries Based on Support Vector Machine," Applied Sciences, vol. 8, no. 6, pp. 925(1-13), Jun. 2018. https://doi.org/10.3390/app8060925
  6. D. Yang, X. Zhang, R. pan, Y. Wang and Z. Chen, "A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curve," Journal of Power Sources, vol. 384, pp. 387-395, Apr. 2018. https://doi.org/10.1016/j.jpowsour.2018.03.015
  7. J.H. Hong, D.H. Lee, E.R. Jeong, Y. Yi, "Towards the swift prediction of the remaining useful life of lithium-ion batteries with end-to-end deep learning," Applied Energy, vol. 278, pp. 115646(1-12), Nov. 2020.
  8. A. Eddahech, O. Briat, N. Bertrand, Y. Deletage and J. Vinassa, "Behavior and state-of-health monitoring of Li-ion batteries using impedance spectroscopy and recurrent neural networks," International Journal of Electrical Power & Energy Systems, vol. 42, no. 1, pp. 487-494, Nov. 2012. https://doi.org/10.1016/j.ijepes.2012.04.050
  9. W. Li, N. Sengupta, P. Dechent, D. Howey, A. Annaswamy and D. Sauer, "Online capacity estimation of lithium-ion batteries with deep long short-term memory networks," Journal of Power Sources, vol. 482, pp. 228863(1-11), Jan. 2021.
  10. L. Ren, L. Zhao, S. Zhao, H. Wang and L. Zhang, "Remaining Useful Life Prediction for Lithium-Ion Battery: A Deep Learning Approach," IEEE Access, vol. 6, pp. 50587-50598, Jul. 2018. https://doi.org/10.1109/ACCESS.2018.2858856
  11. L. Ren, J. Dong, X. Wang, Z. Meng, L. Zhao and M. Deen, "A Data-Driven Auto-CNN-LSTM Prediction Model for Lithium-Ion Battery Remaining Useful Life," IEEE Transactions on Industrial Informatics, vol. 17, no. 5, pp. 3478-3487, Jul. 2020.
  12. 박성윤, 이평연, 유기수, 김종훈, "증분용량 및 차동전압 분석법을 이용한 리튬이온 배터리 SOH 추정 방법 연구," 대한기계학회논문집 A권, 제 45권, 제 3호, 259-266쪽, 2021년 3월 https://doi.org/10.3795/KSME-A.2021.45.3.259
  13. M. Lin, J. Wu, J. Meng, W. Wang and J. Wu, "State of health estimation with attentional long short-term memory network for lithium-ion batteries," Energy, vol. 268, pp. 126706(1-12), Apr. 2023.
  14. H. Gabbar, A. Othman and M. Abdussami, "Review of Battery Management Systems (BMS) Development and Industrial Standards," Technologies, vol. 9, no. 2, pp. 28(1-23), Sep. 2021. https://doi.org/10.3390/technologies9020028
  15. G. Ji, Y. Ma and J.M. Lee, "Mitigating the initial capacity loss (ICL) problem in high-capacity lithium ion battery anode materials," Journal of Materials Chemistry, vol. 21, no. 27, pp. 9819-9824, 2011. https://doi.org/10.1039/c0jm03759a