Journal of IKEEE (전기전자학회논문지)

Institute of Korean Electrical and Electronics Engineers (한국전기전자학회)
권호 표지
DOI QR코드

DOI QR Code

Machine Learning Based State of Health Prediction Algorithm for Batteries Using Entropy Index

엔트로피 지수를 이용한 기계학습 기반의 배터리의 건강 상태 예측 알고리즘

  • Received : 2022.10.14
  • Accepted : 2022.11.11
  • Published : 2022.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

In order to efficeintly manage a battery, it is important to accurately estimate and manage the SOH(State of Health) and RUL(Remaining Useful Life) of the batteries. Even if the batteries are of the same type, the characteristics such as facility capacity and voltage are different, and when the battery for the training model and the battery for prediction through the model are different, there is a limit to measuring the accuracy. In this paper, We proposed the entropy index using voltage distribution and discharge time is generalized, and four batteries are defined as a training set and a test set alternately one by one to predict the health status of batteries through linear regression analysis of machine learning. The proposed method showed a high accuracy of more than 95% using the MAPE(Mean Absolute Percentage Error).

배터리를 효율적으로 관리하기 위해서는 배터리의 건강 상태와 잔여 수명을 정확하게 추정하고 관리하는 것이 중요하다. 배터리는 같은 종류여도 설비용량 및 전압 등의 특성이 다르며 학습용 모델을 위한 배터리와 모델을 통한 예측을 위한 배터리가 서로 다를 경우에는 정확도 측정에 한계가 있다. 본 논문에서는 전압의 분포와 방전 시간을 이용한 엔트로피 지수를 일반화하고 4개의 배터리를 각각 1개씩 교차적으로 훈련 집합과 테스트 집합으로 정의하여 기계학습의 선형회귀 분석을 통하여 배터리의 건강 상태를 예측하는 방법을 제안하였다. 제안된 방법은 평균 절대값 퍼센트 오차를 이용하여 95% 이상의 높은 정확도를 나타내었다.

Keywords

Acknowledgement

This research was the result of being supported by Ministry of Trade, Industry and Energy(MOTIE) in 2022 (No.20215910100030). This research was financially supported by the Ministry of Trade, Industry and Energy, Korea, under the "Regional Innovation Cluster Development Program(R&D, P0016222)" supervised by the Korea Institute for Advancement of Technology(KIAT).

References

  1. S. J. Jung, J. W. Hur, "Deep Learning Approaches to RUL Prediction of Lithum-ion Batteries," Journal of the Korean Society of Manufacturing Process Engineers, Vol.19, No.12, pp.21-27, 2020. DOI: 10.14775/ksmpe.2020.19.12.021
  2. S. R. Hong, M. Kang, H. G. Jeong, J. B. Baek, J. H. Kim, "State of Health Estimation for LithumIon Batteries Using Long-term Recurrent Convolutional Network," The transaction of the Korean Institute of Power Electronics, Vol.26, No.3, pp.183-191, 2021. DOI: 10.1109/IECON43393.2020.9254275
  3. P. Liu, Z. Sun, Z. Wang and J. Zhang, "Entropy-Based Voltage Fault Diagnosis of Battery Systems for Electric Vehicles," Energies 2018, Vol.11, No.1, p.136. DOI: 10.3390/en11010136
  4. T. Bak, S. Lee, "Accurate Estimation of Battery SOH and RUL Based on a Progressive LSTM with a Time Compensated Entropy Index," Proceedings of the Annual Conference of the PHM Society 2019, Vol.11, No.1, 2019. DOI: 10.36001/phmconf.2019.v11i1.833
  5. B. Saha, K. Goebel, "Battery data set," NASA AMES Progmostics Data Repository, 2007.
  6. S. W. Kim, K. Y. Oh, S. C. Lee, "Novel informed deep learning-based prognostics framework for on-board health monitoring of lithium-ion batteries," Applied Energy, Elsevier, Vol.315(C), 2022. DOI: 10.1016/j.apenergy.2022.119011