Generative adversarial network for power sources’ noise characteristics modeling
https://doi.org/10.26583/vestnik.2025.4.6
EDN: VUYXVI
Abstract
The article presents a new approach for modeling voltage noise of chemical power sources for the purpose of datasets augmentation. For the first time known machine learning methods were applied for modelling of the voltage noises of lithium-ion batteries: a generative adversarial neural network based on LSTM layers was designed for this task. A brief statistical and spectral analysis of experimental voltage fluctuations is given. A qualitative and quantitative study of synthetic noise signals is carried out based on the performed analysis of real data. It is shown how the classification of generated data by a deep neural network results in generation of noise characteristics for a given state of charge of the battery. It is recommended how to apply the proposed technique to improve precision of interpretation of voltage fluctuations in power sources. An experimental assessment of the method’s effectiveness is given: a decrease in the determination error of the battery’s state of charge from its noise went from 6.8% to 4.9%.
About the Author
A. G. Popov
Moscow Institution of Physics and Technology;
Central Scientific Research Institute of Chemistry and Mechanics
Russian Federation
Russian Federation
References
1. Li X., Jiang J., Zhang C., Wang L. Y., Zheng L. Robustness of SOC Estimation Algorithms for EV Lithium-Ion Batteries against Modeling Errors and Measurement Noise. Mathematical Problems in Engineering, 2015. pp. 1–14. doi:10.1155/2015/719490
2. Zhang L., Shen Z., Sajadi S.M., Prabuwono A.S., Mahmoud M.Z., Cheraghian G., ElSayed M., Tag El Din. The machine learning in lithium-ion batteries: A review. Engineering Analysis with Boundary Elements, 2022. vol.141. pp. 1-16. doi:10.1016/j.enganabound.2022.04.035.
3. Eleftheriadis P., Gangi M., Leva S., Rey A.V., Groppo E., Grande L. Comparative study of machine learning techniques for the state of health estimation of Li-Ion batteries. Electric Power Systems Research, 2024. vol.235. Article ID: 110889. doi:10.1016/j.epsr.2024.110889
4. Faragher R. Understanding the Basis of the Kalman Filter Via a Simple and Intuitive Derivation [Lecture Notes]. IEEE signal processing magazine, 2012. vol.29. iss.5. pp.128-132. doi:10.1109/MSP.20212.2203621
5. Xie J., Wei X., Bo X., Zhang P., Chen P., Hao W., Yuan M. State of charge estimation of lithium-ion battery based on extended Kalman filter algorithm. Frontiers in Energy Research, 2023. vol.11. Article ID: 1180881. doi: 10.3389/fenrg.2023.1180881
6. Tian Y., Lai R., Li X., Xiang L., Tian J. A combined method for state-of-charge estimation for lithium-ion batteries using a long short-term memory network and an adaptive cubature Kalman filter. Applied Energy, 2020 vol.265. Article ID: 114789. doi: 10.1016/j.apenergy.2020.114789
7. Astafiev E.A. Izmerenie elektrohimicheskih shumov tverdopolimernogo toplivnogo elementa pod nagruzkoj [Measurement of electrochemical noise of a solid polymer fuel cell under load]. Elektrohimiya, 2018. vol. 54. no.6. pp. 638-645. (in Russian)
8. Astafiev E.A. Izmerenie i analiz elektrohimicheskih shumov litij-ionnogo akkumulyatora [Measurement and analysis of electrochemical noise of a lithium-ion battery]. Elektrohimiya, 2019. vol. 55. no.6. pp. 654-662. (in Russian)
9. Astafev E.A. Electrochemical noise measurement of a lithium iron(II) phosphate (LiFePO4) rechargeable battery. Instrumentation Science and Technology, 2019. vol.10. doi: 10.1080/10739149.2019.1658601
10. Astafev E.A. Sravneniye razlichnikh podhodov v analize electrokhimicheskikh shumov vodorodno-vozdushnogo toplivnogo elementa [Comparison of different approaches towards electrochemical noise analysis of air-hydrogen fuel cell]. Electrochemistry, 2020. vol. 56. no. 2. pp. 167-174. (in Russian). doi: 10.1134/S0424857020020036.
11. Astafev E.A. State-of-charge determination of Li/SOCl2 primary battery by means of electrochemical noise measurement. Journal of Solid State Electrochemistry, 2019. vol. 23. pp. 1493-1504.
12. Popov A.G., Mozgovoy N.A., Sushchenya G.N., Pidzhakov V.A., Ulyanov S.A. Primenenie metodov glubokogo obucheniya dlya analiza fluktuacij napryazheniya litievyh istochnikov toka [Application of deep learning methods for the analysis of voltage fluctuations of lithium power sources]. Trudy MFTI, 2023. vol.15. no. 4 (60). pp. 150-161. (in Russian)
13. Martemianov S., Maillard F., Thomas A., Lagonotte P., Madier L. Noise Diagnosis of Commercial Li-Ion Batteries Using High-Order Moments. Russian Journal of Electrochemistry, 2016, vol. 52, no. 12, pp. 1122–1130.
14. Sarker I.H. Machine learning: Algorithms, real-world applications and research directions. SN Computer Science, 2021. no. 2. pp. 1–21.
15. Popov A.G., Eremin N.V., Mozgovoy N.A. Spektral'noe issledovanie elektrohimicheskih shumov litievyh istochnikov toka [Spectral study of electrochemical noise in lithium power sources]. Trudy MFTI, 2021. vol. 13. no. 2 (50). pp 23–31. (in Russian)
16. Goodfellow I.J., Pouget-Abadie J., Mirza M., Xu B., Warde-Farley D., Ozair S., Courville A., Bengio Y. Generative Adversarial Networks . arXiv 2014, arXiv:1406.2661
17. Karras T., Aila T., Laine S., Lehtinen J. Progressive growing of GANs for improved quality, stability, and variation. arXiv preprint, 2017. arXiv:1710.10196.
18. Ledig C., Theis L., Husz´ar F., Caballero J., Cunningham A., Acosta A., Aitken A.P., Tejani A., Totz J., Wang Z. et al. Photo-realistic single image super-resolution using a generative adversarial network. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. doi:10.1109/CVPR.2017.19.
19. Fedus W., Goodfellow I., Dai A.M. MaskGAN: Better text generation via filling in the ____ . arXiv preprint, 2018. arXiv:1801.07736. doi:10.48550/arXiv.1801.07736
20. Liang X., Hu Z., Zhang H., Gan C., Xing E.P. Recurrent topictransition GAN for visual paragraph generation . arXiv preprint, 2017. arXiv:1703.07022.
21. Mogren O. C-RNN-GAN: Continuous recurrent neural networks with adversarial training. arXiv preprint, 2016. arXiv:1611.09904.
22. Yang L.-C., Chou S.-Y., Yang Y.-H. Midinet: A convolutional generative adversarial network for symbolic-domain music generation. arXiv preprint, 2017. arXiv:1703.10847.
23. Yang G., Ma Q., Sun H., Zhang X. State of Health Estimation Based on GAN-LSTM-TL for Lithium-ion Batteries. International Journal of Electrochemical Science, 2022. vol.17. no. 221128. doi: 10.20964/2022.11.07
24. Wong K.L., Chou K.S., Tse R., Tang S.-K., Pau G. A. Novel Fusion Approach Consisting of GAN and State-of-Charge Estimator for Synthetic Battery Operation Data Generation. Electronics, 2023. vol.12. no.657. doi:10.3390/ electronics12030657
25. Glorot X., Bengio Y. Understanding the Difficulty of Training Deep Feedforward Neural Networks. Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, 2010. Sardinia, Italy, AISTATS, 2010. pp.249–356. https://proceedings.mlr.press/v9/glorot10a/glorot10a.pdf
26. van der Maaten L., Hinton G. Visualizing Data using t-SNE. Journal of Machine Learning Research, 2008. vol.9. iss. 86. pp. 2579-2605.
27. Sidorov G., Gelbukh A., Gómez-Adorno H., Pinto D. Soft Similarity and Soft Cosine Measure: Similarity of Features in Vector Space Model. Computación y Sistemas, 2014. vol.18. no.3. pp. 491-504. doi: 10.13053/CyS-18-3-2043
Supplementary files
Review
For citations:
Popov A.G. Generative adversarial network for power sources’ noise characteristics modeling. Vestnik natsional'nogo issledovatel'skogo yadernogo universiteta "MIFI". 2025;14(4):339-351. (In Russ.) https://doi.org/10.26583/vestnik.2025.4.6. EDN: VUYXVI
Views:
18
Email this article 