Perbandingan Support Vector Regression Dan Model Pembanding Untuk Peramalan Umur Pakai (End Of Life) Baterai Lithium-Ion Pada Data NASA PCoE: Kajian Untuk Aplikasi UAV Militer
Comparison Of Support Vector Regression And Baseline Models For Forecasting The End Of Life Of Lithium-Ion Batteries On NASA PCoE Data: A Study For Military UAV Applications
DOI:
https://doi.org/10.59585/bajik.v5i1.2012Keywords:
UAV Militer, Baterai Lithium-Ion, Degradasi Kapasitas, Support Vector Regression, Pemeliharaan PrediktifAbstract
Unmanned Aerial Vehicle (UAV) dalam operasi militer bergantung pada baterai Lithium-Ion (Li-Ion) yang kapasitasnya menurun seiring siklus pengisian dan pengosongan. Penelitian ini mengevaluasi Support Vector Regression (SVR) untuk mengestimasi umur pakai (End of Life, EOL; kapasitas ≤ 1.4 Ah atau 70% dari kapasitas nominal 2.0 Ah) baterai Li-Ion menggunakan data NASA Prognostics Center of Excellence (PCoE) (sel B0005 secara rinci; B0006, B0007, dan B0018 untuk uji kekokohan) sebagai proxy laboratorium degradasi baterai UAV. SVR kernel radial basis function (RBF) dibandingkan dengan regresi linear, regresi polinomial, Random Forest, dan Gaussian Process Regression menggunakan pembagian data kronologis dan validasi silang deret waktu. Pada skenario utama (latih siklus 1-134, uji siklus 135-168), regresi linear memiliki RMSE uji terendah (0.0388 Ah), diikuti SVR (0.0497 Ah); seluruh R-squared uji negatif karena kapasitas pada jendela uji yang pendek sangat sedikit bervariasi. EOL aktual terjadi pada siklus ke-125 dan SVR memprediksi siklus ke-126, tetapi skenario ini pada dasarnya interpolasi. Dengan hanya 100 siklus pertama, SVR, Random Forest, dan Gaussian Process Regression tidak memprediksi perlintasan ambang. Pada evaluasi rolling origin terhadap tiga sel yang mencapai 1.4 Ah, SVR kernel RBF memprediksi perlintasan ambang pada 5 dari 15 percobaan, sedangkan regresi linear dan SVR kernel linear pada 15 dari 15 percobaan, dengan rata-rata galat absolut EOL turun dari sekitar 28-30 siklus pada 50% masa pakai menjadi 5-6 siklus pada 90%. Dengan demikian SVR kernel RBF tidak direkomendasikan untuk peramalan EOL pada data ini. Hasil bersifat awal: data berasal dari uji laboratorium beban konstan, bukan misi UAV.
References
Ahwiadi, M. and Wang, W. (2025) ‘Battery health monitoring and remaining useful life prediction techniques: A review of technologies’, Batteries, 11(1), p. 31. doi: 10.3390/batteries11010031.
Andrioaia, D. A., Gaitan, V. G., Culea, G. and Banu, I. V. (2024) ‘Predicting the RUL of Li-Ion batteries in UAVs using machine learning techniques’, Computers, 13(3), p. 64. doi: 10.3390/computers13030064.
Breiman, L. (2001) ‘Random forests’, Machine Learning, 45(1), pp. 5-32.
Deng, L.-M., Hsu, Y.-C. and Li, H.-X. (2017) ‘An improved model for remaining useful life prediction on capacity degradation and regeneration of lithium-ion battery’, Proceedings of the Annual Conference of the PHM Society, 9(1).
Hasib, S. A. et al. (2021) ‘A comprehensive review of available battery datasets, RUL prediction approaches, and advanced battery management’, IEEE Access, 9, pp. 86166-86193. doi: 10.1109/ACCESS.2021.3089032.
Hastie, T., Tibshirani, R. and Friedman, J. (2009) The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd edn. New York: Springer.
Hu, X., Xu, L., Lin, X., Dong, S. and Pecht, M. G. (2020) ‘Battery lifetime prognostics’, Joule, 4(2), pp. 310-346. doi: 10.1016/j.joule.2019.11.018.
Ma, G., Zhang, Y., Cheng, C., Zhou, B., Hu, P. and Yuan, Y. (2019) ‘Remaining useful life prediction of lithium-ion batteries based on false nearest neighbors and a hybrid neural network’, Applied Energy, 253, p. 113626. doi: 10.1016/j.apenergy.2019.113626.
Ng, S. S. Y., Xing, Y. and Tsui, K. L. (2014) ‘A naive Bayes model for robust remaining useful life prediction of lithium-ion battery’, Applied Energy, 118, pp. 114-123. doi: 10.1016/j.apenergy.2013.12.020.
Patrizi, G. et al. (2024) ‘A review of degradation models and remaining useful life prediction for testing design and predictive maintenance of lithium-ion batteries’, Sensors, 24(11), p. 3382. doi: 10.3390/s24113382.
Pedregosa, F. et al. (2011) ‘Scikit-learn: Machine learning in Python’, Journal of Machine Learning Research, 12, pp. 2825-2830.
Rasmussen, C. E. and Williams, C. K. I. (2006) Gaussian Processes for Machine Learning. Cambridge, MA: MIT Press.
Roman, D., Saxena, S., Robu, V., Pecht, M. and Flynn, D. (2021) ‘Machine learning pipeline for battery state-of-health estimation’, Nature Machine Intelligence, 3(5), pp. 447-456. doi: 10.1038/s42256-021-00312-3.
Saha, B. and Goebel, K. (2007) ‘Battery Data Set’, NASA Prognostics Data Repository, NASA Ames Research Center, Moffett Field, CA.
Severson, K. A. et al. (2019) ‘Data-driven prediction of battery cycle life before capacity degradation’, Nature Energy, 4(5), pp. 383-391. doi: 10.1038/s41560-019-0356-8.
Smola, A. J. and Schölkopf, B. (2004) ‘A tutorial on support vector regression’, Statistics and Computing, 14(3), pp. 199-222.
Wang, Y. et al. (2020) ‘A comprehensive review of battery modeling and state estimation approaches for advanced battery management systems’, Renewable and Sustainable Energy Reviews, 131, p. 110015. doi: 10.1016/j.rser.2020.110015.
Xiong, R., Li, L. and Tian, J. (2018) ‘Towards a smarter battery management system: A critical review on battery state of health monitoring methods’, Journal of Power Sources, 405, pp. 18-29. doi: 10.1016/j.jpowsour.2018.10.019.
Zhang, Y., Xiong, R., He, H. and Pecht, M. G. (2018) ‘Long short-term memory recurrent neural network for remaining useful life prediction of lithium-ion batteries’, IEEE Transactions on Vehicular Technology, 67(7), pp. 5695-5705. doi: 10.1109/TVT.2018.2805189.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Ivan Harendra Yusuf Setyawan, Sovian Aritonang, Muhammad Hafidh Husna Ashshidqi, Muhammad Rizky Dayrobby, Rhaulza Fathir Jabar Hamdi, Ayyash Khoyyir Ahmad Kemal, Afriza Rizky Putra Adjie, Brillian Salman Al Farisi

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.




