| Peer-Reviewed

State-of-charge Estimation of Lithium-ion Batteries Based on PSO-BP Neural Network

Received: 25 October 2021    Accepted: 15 November 2021    Published: 17 November 2021
Views:       Downloads:
Abstract

The state of charge (SOC) of lithium-ion battery is a variable and cannot be measured directly by sensors. Therefore, accurate estimation of battery state of charge is the key to ensure the safe and reliable operation of battery management system (BMS) and reduce the required battery cost. In the research of estimating the state of charge of lithium-ion batteries, the initial setting of the weights and thresholds of the BP neural network easy to falls into the local minimum problem, which makes the SOC estimation insufficiently accurate. Therefore, a method of SOC estimation of lithium-ion battery based on particle swarm optimization (PSO) and BP neural network is proposed in this paper. Taking lithium manganese battery (LiMn2O4) as the object, use the multi-physics simulation platform COMSOL to conduct charging and discharging experiments on it, and collect the relevant performance parameters of the battery. Under the condition of constant temperature and constant current, the SOC of the battery is inferred according to the voltage and discharge rate. Building a PSO-BP neural network model with voltage and discharge rate as input and battery SOC as output. The performance of SOC estimation is evaluated from the aspects of overall correlation, training time and robustness. It is compared with the estimation method based on BP neural network. The simulation results show that the absolute error of the estimation method based on PSO-BP neural network is 2.68%, which is 3.18% higher than that of BP neural network, and the accuracy is higher. The proposed method has more advantages.

Published in International Journal of Energy and Power Engineering (Volume 10, Issue 6)
DOI 10.11648/j.ijepe.20211006.13
Page(s) 115-120
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2024. Published by Science Publishing Group

Keywords

Lithium-ion Batteries, State of Charge Estimation, Particle Swarm Optimization, BP Neural Network

References
[1] Saite Wu. Research on current status of domestic energy utilization and prospects of heat pump technology [J]. Resource Conservation And Environmental Protection, 2019, 09: 35-36.
[2] Yihuan Li, Kang Li, Jian Yu. Estimation Approaches for States of Charge and Health of Lithium-ion Battery [J]. Power Generation Technology, 2021, 42 (05): 537-546.
[3] Yifeng Guo, Zeshuang Zhao, Limin Huang. SoC Estimation of Lithium Battery Based on Improved BP Neural Network [J]. Energy Procedia, 2017, 105: 4153-4158.
[4] Qi Yang, Huan-wu Sun, Feng-bo zhang. Research on SOC Improved Unscented Kalman Filter Estimation Algorithm for Lithium Batteries [J]. Machinery Design & Manufacture, 2021 (10): 220-224.
[5] Anton JA, Nieto PG, Viejo CB, Vilan JV. Support vector machines used to estimate the battery state of charge. IEEE Trans Power Electron 2013; 28 (12): 19-26.
[6] Sahinoglu GO, Pajovic M, Sahinoglu Z, Wang Y, Orlik PV, Wada T. Batterystate-of-charge estimation based on regular/recurrent Gaussian process regression. IEEE Trans Ind Electron 2018; 65 (5): 11-21.
[7] Wei He, Nicholas Williard, Chaochao Chen, Michael Pecht. State of charge estimation for Li-ion batteries using neural network modeling and unscented Kalman filter-based error cancellation [J]. International Journal of Electrical Power and Energy Systems, 2014, 62: 783-791.
[8] Hicham Chaoui, Chinemerem C, Ibe-Ekeocha, Hamid Gualous. Aging prediction and state of charge estimation of a LiFePO 4 battery using input time-delayed neural networks [J]. Electric Power Systems Research, 2017, 146: 189-197.
[9] Yujie Wang, Chenbin Zhang, Zonghai Chen. A method for state-of-charge estimation of Li-ion batteries based on multi-model switching strategy [J]. Applied Energy, 2015, 137: 427-434.
[10] Shurui Ren, Qi Wang. Estimation of Power Battery SOC Based on BAS-BP Neural Network [J]. Automation & Instrumentation, 2021, 36 (08): 87-91.
[11] Angelo Bonfitto. A Method for the Combined Estimation of Battery State of Charge and State of Health Based on Artificial Neural Networks [J]. Energies, 2020, 13 (10).
[12] Shafqat Wafa, Malik Sehrish, Lee KyuTae, Kim DoHyeun. PSO Based Optimized Ensemble Learning and Feature Selection Approach for Efficient Energy Forecast [J]. Electronics, 2021, 10 (18): 2188-2188.
[13] Kim Tae-Young, Cho Sung-Bae. Optimizing CNN-LSTM neural networks with PSO for anomalous query access control [J]. Neurocomputing, 2021, 456: 666-677.
[14] Hein Thiemo, Ziegler Andreas, Oeser David, Ackva Ansgar. A capacity-based equalization method for aged lithium-ion batteries in electric vehicles [J]. Electric Power Systems Research, 2021, 191.
[15] LiuWang Kang, Xuan Zhao, Jian Ma. A new neural network model for the state-of-charge estimation in the battery degradation process [J]. Applied Energy, 2014, 121: 20-27.
[16] Weiyang Dai, Zhong Tang, Chenhao Shi, Zeshuang Li. SOC estimation of power battery based on QGA-QBP [J]. Chinese Journal of Power Sources, 2021, 45 (07): 893-896.
Cite This Article
  • APA Style

    Biao Yang, Yinshuang Wang, Hao Gao. (2021). State-of-charge Estimation of Lithium-ion Batteries Based on PSO-BP Neural Network. International Journal of Energy and Power Engineering, 10(6), 115-120. https://doi.org/10.11648/j.ijepe.20211006.13

    Copy | Download

    ACS Style

    Biao Yang; Yinshuang Wang; Hao Gao. State-of-charge Estimation of Lithium-ion Batteries Based on PSO-BP Neural Network. Int. J. Energy Power Eng. 2021, 10(6), 115-120. doi: 10.11648/j.ijepe.20211006.13

    Copy | Download

    AMA Style

    Biao Yang, Yinshuang Wang, Hao Gao. State-of-charge Estimation of Lithium-ion Batteries Based on PSO-BP Neural Network. Int J Energy Power Eng. 2021;10(6):115-120. doi: 10.11648/j.ijepe.20211006.13

    Copy | Download

  • @article{10.11648/j.ijepe.20211006.13,
      author = {Biao Yang and Yinshuang Wang and Hao Gao},
      title = {State-of-charge Estimation of Lithium-ion Batteries Based on PSO-BP Neural Network},
      journal = {International Journal of Energy and Power Engineering},
      volume = {10},
      number = {6},
      pages = {115-120},
      doi = {10.11648/j.ijepe.20211006.13},
      url = {https://doi.org/10.11648/j.ijepe.20211006.13},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijepe.20211006.13},
      abstract = {The state of charge (SOC) of lithium-ion battery is a variable and cannot be measured directly by sensors. Therefore, accurate estimation of battery state of charge is the key to ensure the safe and reliable operation of battery management system (BMS) and reduce the required battery cost. In the research of estimating the state of charge of lithium-ion batteries, the initial setting of the weights and thresholds of the BP neural network easy to falls into the local minimum problem, which makes the SOC estimation insufficiently accurate. Therefore, a method of SOC estimation of lithium-ion battery based on particle swarm optimization (PSO) and BP neural network is proposed in this paper. Taking lithium manganese battery (LiMn2O4) as the object, use the multi-physics simulation platform COMSOL to conduct charging and discharging experiments on it, and collect the relevant performance parameters of the battery. Under the condition of constant temperature and constant current, the SOC of the battery is inferred according to the voltage and discharge rate. Building a PSO-BP neural network model with voltage and discharge rate as input and battery SOC as output. The performance of SOC estimation is evaluated from the aspects of overall correlation, training time and robustness. It is compared with the estimation method based on BP neural network. The simulation results show that the absolute error of the estimation method based on PSO-BP neural network is 2.68%, which is 3.18% higher than that of BP neural network, and the accuracy is higher. The proposed method has more advantages.},
     year = {2021}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - State-of-charge Estimation of Lithium-ion Batteries Based on PSO-BP Neural Network
    AU  - Biao Yang
    AU  - Yinshuang Wang
    AU  - Hao Gao
    Y1  - 2021/11/17
    PY  - 2021
    N1  - https://doi.org/10.11648/j.ijepe.20211006.13
    DO  - 10.11648/j.ijepe.20211006.13
    T2  - International Journal of Energy and Power Engineering
    JF  - International Journal of Energy and Power Engineering
    JO  - International Journal of Energy and Power Engineering
    SP  - 115
    EP  - 120
    PB  - Science Publishing Group
    SN  - 2326-960X
    UR  - https://doi.org/10.11648/j.ijepe.20211006.13
    AB  - The state of charge (SOC) of lithium-ion battery is a variable and cannot be measured directly by sensors. Therefore, accurate estimation of battery state of charge is the key to ensure the safe and reliable operation of battery management system (BMS) and reduce the required battery cost. In the research of estimating the state of charge of lithium-ion batteries, the initial setting of the weights and thresholds of the BP neural network easy to falls into the local minimum problem, which makes the SOC estimation insufficiently accurate. Therefore, a method of SOC estimation of lithium-ion battery based on particle swarm optimization (PSO) and BP neural network is proposed in this paper. Taking lithium manganese battery (LiMn2O4) as the object, use the multi-physics simulation platform COMSOL to conduct charging and discharging experiments on it, and collect the relevant performance parameters of the battery. Under the condition of constant temperature and constant current, the SOC of the battery is inferred according to the voltage and discharge rate. Building a PSO-BP neural network model with voltage and discharge rate as input and battery SOC as output. The performance of SOC estimation is evaluated from the aspects of overall correlation, training time and robustness. It is compared with the estimation method based on BP neural network. The simulation results show that the absolute error of the estimation method based on PSO-BP neural network is 2.68%, which is 3.18% higher than that of BP neural network, and the accuracy is higher. The proposed method has more advantages.
    VL  - 10
    IS  - 6
    ER  - 

    Copy | Download

Author Information
  • School of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China; Key Laboratory of Unconventional Metallurgy, Ministry of Education, Kunming, China

  • School of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China

  • School of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China

  • Sections