A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression

Recai Özcan; Ferhat Koca

doi:10.69560/cujast.1904437

EN TR

A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using KNN Regression

Abstract

This study systematically investigates the effect of feature representation design on performance in Remaining Useful Life (RUL) prediction for lithium-ion batteries. While RUL prediction in existing literature primarily focuses on model architecture and hyperparameter optimization, the role of feature representation design, especially concerning cross-battery generalization, has been addressed to a limited extent. In this context, two different representation approaches, namely absolute feature space-based representation (Absolute Feature Representation (AFR)) and reference cycle-centered representation (Reference-based Feature Representation (RFR)), were developed and evaluated. Experiments were conducted on data from 4 batteries within the NASA PCOE lithium-ion battery dataset, and evaluation was performed using a battery-wise cross-validation strategy. K-Nearest Neighbors (KNN) was used as the regression model, and only the representation effect was analyzed by keeping model parameters constant. The results demonstrate that RFR provides consistent improvements across all performance metrics. In the evaluation metrics, the average MAE value decreased from 0.148 to 0.096, and the RMSE value decreased from 0.173 to 0.120. The R² value increased from 0.637 to 0.825. These improvements correspond to performance increases of approximately 35%, 31%, and 29%, respectively. Correlation analysis revealed that the transformation performed in RFR strengthened the feature-target relationship and made relative degradation dynamics more prominent by suppressing absolute level differences. The diagnostic analyses performed confirm that the error structure is balanced and systematic. The findings indicate that performance improvement in RUL prediction is strongly dependent not only on model selection but also on feature representation design. Especially in cases where initial differences between batteries are high, the RFR approach offers an effective strategy that provides significant and consistent improvements without increasing model complexity.

Keywords

Remaining Useful Life (RUL) prediction;, Lithium- ion batteries;, Reference- based feature representation;, Distance- based regression;, K- Nearest Neighbors (KNN);, Battery- wise cross- validation

A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression

Abstract

Bu çalışmada, lithium-ion bataryalar için Remaining Useful Life (RUL) tahmininde özellik temsili tasarımının performans üzerindeki etkisini sistematik olarak incelenmiştir. Mevcut literatürde RUL tahmini çoğunlukla model mimarisi ve hiperparametre optimizasyonuna odaklanırken, özellik temsili tasarımının özellikle cross-battery generalization üzerindeki rolü sınırlı biçimde ele alınmıştır. Bu bağlamda mutlak özellik uzayına dayalı temsil (Absolute Feature Representation (AFR)) ve referans döngü merkezlemeli temsil (Reference-based Feature Representation (RFR)) olmak üzere iki farklı temsil yaklaşımı oluşturularak değerlendirilmiştir. Deneyler NASA PCoE lithium-ion battery veri seti üzerinde 4 bataryaya ait veriler üzerinde gerçekleştirilmiş ve değerlendirme a battery-wise cross-validation stratejisi ile yapılmıştır. Regresyon modeli olarak K-Nearest Neighbors (KNN) kullanılmış ve model parametreleri sabit tutularak yalnızca temsil etkisi analiz edilmiştir. Sonuçlar, RFR’nin tüm performans metriklerinde tutarlı iyileşme sağladığını göstermektedir. Değerlendirme metriklerinde ortalama MAE değeri 0.148’den 0.096’ya, RMSE değeri 0.173’ten 0.120’ye düşmüştür. R² değeri ise 0.637’den 0.825’e yükselmiştir. Bu iyileşmeler sırasıyla yaklaşık %35, %31 ve %29 oranında performans artışına karşılık gelmektedir. Korelasyon analizi, RFR’de yapılan transformasyonun feature–target ilişkisini güçlendirdiğini ve mutlak seviye farklarını baskılayarak göreli degradasyon dinamiklerini daha belirgin hale getirdiğini ortaya koymuştur. Gerçekleştirilen tanılayıcı analizler hata yapısının dengeli ve sistematik olduğunu doğrulamaktadır. Elde edilen bulgular, RUL tahmininde performans artışının yalnızca model seçimine değil, özellik temsili tasarımına da güçlü biçimde bağlı olduğunu göstermektedir. Özellikle bataryalar arası başlangıç farklılıklarının yüksek olduğu durumlarda, RFR yaklaşımı model karmaşıklığını artırmadan anlamlı ve tutarlı iyileşmeler sağlayan etkili bir strateji sunmaktadır.

Keywords

Kalan Kullanım Ömrü (RUL) tahmini;, Lityum iyon bataryalar;, Referans tabanlı özellik temsili;, Mesafe tabanlı regresyon;, K-En Yakın Komşular (KNN);, Batarya bazlı çapraz doğrulama

References

Chen, D., Hong, W., & Zhou, X. (2022). Transformer Network for Remaining Useful Life Prediction of Lithium-Ion Batteries. IEEE Access, 10, 19621–19628. https://doi.org/10.1109/ACCESS.2022.3151975
Chen, X. (2024). A novel transformer-based DL model enhanced by position-sensitive attention and gated hierarchical LSTM for aero-engine RUL prediction. Scientific Reports, 14(1), 10061. https://doi.org/10.1038/s41598-024-59095-3
Cover, T., & Hart, P. (1967). Nearest neighbor pattern classification. IEEE Transactions on Information Theory, 13(1), 21–27. https://doi.org/10.1109/TIT.1967.1053964
Dragomiretskiy, K., & Zosso, D. (2014). Variational Mode Decomposition. IEEE Transactions on Signal Processing, 62(3), 531–544. https://doi.org/10.1109/TSP.2013.2288675
Guyon, I., & Elisseeff, A. (2003). An introduction to variable and feature selection. Journal of Machine Learning Research, 3(Mar), 1157–1182.
Harris, C. R., Millman, K. J., van der Walt, S. J., Gommers, R., Virtanen, P., Cournapeau, D., Wieser, E., Taylor, J., Berg, S., Smith, N. J., Kern, R., Picus, M., Hoyer, S., van Kerkwijk, M. H., Brett, M., Haldane, A., del Río, J. F., Wiebe, M., Peterson, P., … Oliphant, T. E. (2020). Array programming with NumPy. Nature, 585(7825), 357–362. https://doi.org/10.1038/s41586-020-2649-2
Hastie, T., Tibshirani, R., Friedman, J., & others. (2009). The elements of statistical learning. Springer series in statistics New-York.
Hunter, J. (2007). Matplotlib: A 2D graphics environment Computing in Science & Engineering. 2007; 9 (3): 90–95. DOI: Https://Doi. Org/10.1109/MCSE.
Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: Principles and practice. OTexts.
Kaufman, S., Rosset, S., Perlich, C., & Stitelman, O. (2012). Leakage in data mining: Formulation, detection, and avoidance. ACM Transactions on Knowledge Discovery from Data (TKDD), 6(4), 1–21.

Kohavi, R. (1995). A study of cross-validation and bootstrap for accuracy estimation and model selection. Ijcai, 14(2), 1137–1145.
Lee Rodgers, J., & Nicewander, W. A. (1988). Thirteen ways to look at the correlation coefficient. The American Statistician, 42(1), 59–66.
Li, X., Zhang, L., Wang, Z., & Dong, P. (2019). Remaining useful life prediction for lithium-ion batteries based on a hybrid model combining the long short-term memory and Elman neural networks. Journal of Energy Storage, 21, 510–518. https://doi.org/10.1016/j.est.2018.12.011
Li, Y., Li, J., Wang, H., Liu, C., & Tan, J. (2024). Knowledge enhanced ensemble method for remaining useful life prediction under variable working conditions. Reliability Engineering & System Safety, 242, 109748. https://doi.org/10.1016/j.ress.2023.109748
Liu, K., Li, K., Peng, Q., & Zhang, C. (2019). A brief review on key technologies in the battery management system of electric vehicles. Frontiers of Mechanical Engineering, 14(1), 47–64. https://doi.org/10.1007/s11465-018-0516-8
Lundberg, S. M., & Lee, S.-I. (2017). A Unified Approach to Interpreting Model Predictions. Advances in Neural Information Processing Systems, 30. https://proceedings.neurips.cc/paper/2017/hash/8a20a8621978632d76c43dfd28b67767-Abstract.html
McKinney, W. & others. (2010). Data structures for statistical computing in Python. Scipy, 445(1), 51–56.
Nitta, N., Wu, F., Lee, J. T., & Yushin, G. (2015). Li-ion battery materials: Present and future. Materials Today, 18(5), 252–264. https://doi.org/10.1016/j.mattod.2014.10.040
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., & Duchesnay, É. (2011). Scikit-learn: Machine Learning in Python. J. Mach. Learn. Res., 12(null), 2825–2830.
Reza, M. S., Mannan, M., Mansor, M., Ker, P. J., Mahlia, T. M. I., & Hannan, M. A. (2024). Recent advancement of remaining useful life prediction of lithium-ion battery in electric vehicle applications: A review of modelling mechanisms, network configurations, factors, and outstanding issues. Energy Reports, 11, 4824–4848. https://doi.org/10.1016/j.egyr.2024.04.039
Saha, B., & Goebel, K. (2007). NASA Battery Dataset. https://www.kaggle.com/datasets/patrickfleith/nasa-battery-dataset
Sahdev, G., Agarwal, S., & Bhati, P. (2025). Hybrid Learning Models for Remaining Useful Life Prediction of Lithium-Ion Batteries in Electric Vehicles. IOP Conference Series: Materials Science and Engineering, 1334(1), 012003. https://doi.org/10.1088/1757-899X/1334/1/012003
Sarıkaya, Y., Özcan, R. & Koca, F. (2025). Lityum-İyon Batarya Modülünün Farklı Deşarj Akımlarında ve Hava Hızlarında Sıcaklıklarının Deneysel Olarak İncelenmesi. International Journal of Advanced Natural Sciences and Engineering Researches, 9(5), 210-215.
Saxena, A., Celaya, J., Balaban, E., Goebel, K., Saha, B., Saha, S., & Schwabacher, M. (2008). Metrics for evaluating performance of prognostic techniques. 2008 International Conference on Prognostics and Health Management, 1–17. https://doi.org/10.1109/PHM.2008.4711436
Severson, K. A., Attia, P. M., Jin, N., Perkins, N., Jiang, B., Yang, Z., Chen, M. H., Aykol, M., Herring, P. K., Fraggedakis, D., Bazant, M. Z., Harris, S. J., Chueh, W. C., & Braatz, R. D. (2019). Data-driven prediction of battery cycle life before capacity degradation. Nature Energy, 4(5), 383–391. https://doi.org/10.1038/s41560-019-0356-8
Shu, X., Yang, W., Wei, K., Zhang, B., Yan, R., & Cattley, R. (2025). Remaining useful life prediction of lithium-ion batteries by leveraging end-of-charge characteristics. Journal of Energy Storage, 134, 118135. https://doi.org/10.1016/j.est.2025.118135
Si, X.-S., Wang, W., Hu, C.-H., & Zhou, D.-H. (2011). Remaining useful life estimation – A review on the statistical data driven approaches. European Journal of Operational Research, 213(1), 1–14. https://doi.org/10.1016/j.ejor.2010.11.018
Song, K., Hu, D., Tong, Y., & Yue, X. (2023). Remaining life prediction of lithium-ion batteries based on health management: A review. Journal of Energy Storage, 57, 106193. https://doi.org/10.1016/j.est.2022.106193
Sui, X., He, S., Vilsen, S. B., Meng, J., Teodorescu, R., & Stroe, D.-I. (2021). A review of non-probabilistic machine learning-based state of health estimation techniques for Lithium-ion battery. Applied Energy, 300, 117346. https://doi.org/10.1016/j.apenergy.2021.117346
Torres, M. E., Colominas, M. A., Schlotthauer, G., & Flandrin, P. (2011). A complete ensemble empirical mode decomposition with adaptive noise. 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 4144–4147. https://doi.org/10.1109/ICASSP.2011.5947265
Wang, S., Zhang, Y., Huang, H., Shi, Y., & Si, J. (2025). RUL Prediction of DC Contactor Using CNN-LSTM With Channel Attention and Fusion of Dual Aggregated Features. IEEE Access, 13, 35634–35644. https://doi.org/10.1109/ACCESS.2024.3516200
Willmott, C. J., & Matsuura, K. (2005). Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Climate Research, 30, 79–82. https://doi.org/10.3354/cr030079

Details

Primary Language

Turkish

Subjects

Electrical Energy Storage, Electrical Engineering (Other), Energy Systems Engineering (Other)

Journal Section

Research Article

Authors

Recai Özcan ^*
0000-0003-1708-1225
Türkiye

Ferhat Koca
0000-0001-8849-5295
Türkiye

Publication Date

June 29, 2026

Submission Date

March 6, 2026

Acceptance Date

May 4, 2026

Published in Issue

Year 2026 Volume: 5 Number: 1

DOI

https://doi.org/10.69560/cujast.1904437

IZ

https://izlik.org/JA54EU46YG

APA

Özcan, R., & Koca, F. (2026). A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression. Sivas Cumhuriyet Üniversitesi Bilim Ve Teknoloji Dergisi, 5(1), 25-39. https://doi.org/10.69560/cujast.1904437

AMA

1.Özcan R, Koca F. A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression. CUJAST. 2026;5(1):25-39. doi:10.69560/cujast.1904437

Chicago

Özcan, Recai, and Ferhat Koca. 2026. “A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression”. Sivas Cumhuriyet Üniversitesi Bilim Ve Teknoloji Dergisi 5 (1): 25-39. https://doi.org/10.69560/cujast.1904437.

EndNote

Özcan R, Koca F (June 1, 2026) A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression. Sivas Cumhuriyet Üniversitesi Bilim ve Teknoloji Dergisi 5 1 25–39.

IEEE

[1]R. Özcan and F. Koca, “A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression”, CUJAST, vol. 5, no. 1, pp. 25–39, June 2026, doi: 10.69560/cujast.1904437.

ISNAD

Özcan, Recai - Koca, Ferhat. “A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression”. Sivas Cumhuriyet Üniversitesi Bilim ve Teknoloji Dergisi 5/1 (June 1, 2026): 25-39. https://doi.org/10.69560/cujast.1904437.

JAMA

1.Özcan R, Koca F. A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression. CUJAST. 2026;5:25–39.

MLA

Özcan, Recai, and Ferhat Koca. “A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression”. Sivas Cumhuriyet Üniversitesi Bilim Ve Teknoloji Dergisi, vol. 5, no. 1, June 2026, pp. 25-39, doi:10.69560/cujast.1904437.

Vancouver

1.Recai Özcan, Ferhat Koca. A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression. CUJAST. 2026 Jun. 1;5(1):25-39. doi:10.69560/cujast.1904437

A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using KNN Regression

Abstract

Keywords

A Reference-Based Feature Representation Approach for Battery Remaining Useful Life Prediction Using K-Nearest Neighbors Regression

Abstract

Keywords

References

Details

Primary Language

Subjects

Journal Section

Authors

Publication Date

Submission Date

Acceptance Date

Published in Issue

DOI

IZ

Cite