Physics-informed gradient-boosting prognostics of remaining useful life for PV inverters and telecom rectifiers under voltage-unbalance exposure: a simulation study
Keywords:
kuchlanish nosimmetrikligi, qolgan xizmat muddati (RUL), XGBoost, fizika asosidagi mashinaviy o‘rganish, konformal bashorat, ma’lumotlar sizib chiqishi, baholash protokoli, fotoelektrik invertor, telekommunikatsiya elektr ta’minoti, prognozli texnik xizmat ko‘rsatishAbstract
Sustained voltage unbalance shortens the life of three-phase equipment at distributed sites - grid-connected photovoltaic (PV) inverters and telecommunication rectifier stations - through additional losses and accelerated thermal ageing. Because no run-to-failure dataset links long-term unbalance exposure to failures of such assets, we evaluate a hybrid pipeline in which a cumulative-damage physical model generates degradation trajectories and an XGBoost regressor learns remaining useful life (RUL) from unbalance statistics, threshold-exposure durations, thermal-load descriptors and asset context. The principal result is methodological. Fleet degradation data are clustered - 20 498 observations from 520 objects - so an observation-level split places 100 % of test rows in objects already seen in training and reports = 0.812. Under a fully object-disjoint protocol the model reaches = 0.726 ± 0.033 over 12 generator seeds, the margin over a budget-matched Random Forest falls to 2.1 ± 2.5 RMSE percentage points, and removing the physical damage index costs 0.087 ± 0.016 of rather than the 1.03 measured under the leaky split. Conformal intervals calibrated on held-out objects attain 84.3 % coverage against a nominal 80 %, with conditional coverage of 61.5–94.5 % across RUL quintiles. A three-level balancer-recommendation layer reaches 91.6 % accuracy ( = 0.771) against 63.9 % for a no-model threshold rule. We report the study’s simulation circularity openly and treat the split-protocol comparison as the transferable finding.
References
GOST 32144-2013. Electric energy. Electromagnetic compatibility of technical equipment. Power quality limits in public power supply systems. Interstate Council for Standardization, Metrology and Certification, 2013.
EN 50160:2010. Voltage characteristics of electricity supplied by public electricity networks. CENELEC, 2010.
IEC 61000-4-30:2015. Electromagnetic compatibility (EMC) — Part 4-30: Testing and measurement techniques — Power quality measurement methods. IEC, 2015.
J. P. G. de Abreu and A. E. Emanuel, “Induction motor thermal aging caused by voltage distortion and imbalance: Loss of useful life and its estimated cost,” IEEE Transactions on Industry Applications, vol. 38, no. 1, pp. 12–20, 2002.
P. Gnaciński, “Windings temperature and loss of life of an induction machine under voltage unbalance combined with over- or undervoltages,” IEEE Transactions on Energy Conversion, vol. 23, no. 2, pp. 363–371, 2008.
V. M. Montsinger, “Loading transformers by temperature,” Transactions of the American Institute of Electrical Engineers, vol. 49, no. 2, pp. 776–790, 1930.
H. Wang and F. Blaabjerg, “Reliability of capacitors for DC-link applications in power electronic converters — An overview,” IEEE Transactions on Industry Applications, vol. 50, no. 5, pp. 3569–3578, 2014.
A. Saxena, K. Goebel, D. Simon, and N. Eklund, “Damage propagation modeling for aircraft engine run-to-failure simulation,” in Proc. International Conference on Prognostics and Health Management (PHM), Denver, CO, 2008, pp. 1–9.
X.-S. Si, W. Wang, C.-H. Hu, and D.-H. Zhou, “Remaining useful life estimation — A review on the statistical data driven approaches,” European Journal of Operational Research, vol. 213, no. 1, pp. 1–14, 2011.
Y. Lei, N. Li, L. Guo, N. Li, T. Yan, and J. Lin, “Machinery health prognostics: A systematic review from data acquisition to RUL prediction,” Mechanical Systems and Signal Processing, vol. 104, pp. 799–834, 2018.
T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, 2016, pp. 785–794.
L. Grinsztajn, E. Oyallon, and G. Varoquaux, “Why do tree-based models still outperform deep learning on typical tabular data?,” in Advances in Neural Information Processing Systems 35 (NeurIPS 2022), Datasets and Benchmarks Track, 2022.
R. Shwartz-Ziv and A. Armon, “Tabular data: Deep learning is not all you need,” Information Fusion, vol. 81, pp. 84–90, 2022.
S. Zheng, K. Ristovski, A. Farahat, and C. Gupta, “Long short-term memory network for remaining useful life estimation,” in Proc. IEEE International Conference on Prognostics and Health Management (ICPHM), Dallas, TX, 2017, pp. 88–95.
Y. Romano, E. Patterson, and E. J. Candès, “Conformalized quantile regression,” in Advances in Neural Information Processing Systems 32 (NeurIPS 2019), 2019, pp. 3538–3548.
V. Vovk, A. Gammerman, and G. Shafer, Algorithmic Learning in a Random World, 2nd ed. Cham, Switzerland: Springer, 2022.
M. A. Miner, “Cumulative damage in fatigue,” Journal of Applied Mechanics, vol. 12, no. 3, pp. A159–A164, 1945.
S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” in Advances in Neural Information Processing Systems 30 (NeurIPS 2017), 2017, pp. 4765–4774.
ANSI/NEMA MG 1-2024. Motors and Generators, §14.36. National Electrical Manufacturers Association, 2024.
IEC 60034-26:2026. Rotating electrical machines — Part 26: Effects of unbalanced voltages on the performance of three-phase cage induction motors, 2nd ed. IEC, 2026.
H. Wang, P. Davari, H. Wang, D. Kumar, F. Zare, and F. Blaabjerg, “Lifetime estimation of DC-link capacitors in adjustable speed drives under grid voltage unbalances,” IEEE Transactions on Power Electronics, vol. 34, no. 5, pp. 4064–4078, 2019.
IEEE Std C57.91-2025. IEEE Guide for Loading Mineral-Oil-Immersed Transformers and Step-Voltage Regulators. IEEE, 2026.
IEC 60216-1:2025. Electrical insulating materials — Thermal endurance properties — Part 1: Ageing procedures and evaluation of test results, 7th ed. IEC, 2025.
M. Arias Chao, C. Kulkarni, K. Goebel, and O. Fink, “Fusing physics-based and deep learning models for prognostics,” Reliability Engineering & System Safety, vol. 217, art. no. 107961, 2022.
M. Arias Chao, C. Kulkarni, K. Goebel, and O. Fink, “Aircraft engine run-to-failure dataset under real flight conditions for prognostics and diagnostics,” Data, vol. 6, no. 1, art. no. 5, 2021.
L. Liao and F. Köttig, “Review of hybrid prognostics approaches for remaining useful life prediction of engineered systems, and an application to battery life prediction,” IEEE Transactions on Reliability, vol. 63, no. 1, pp. 191–207, 2014.
S. Peyghami, P. Palensky, and F. Blaabjerg, “An overview on the reliability of modern power electronic based power systems,” IEEE Open Journal of Power Electronics, vol. 1, pp. 34–50, 2020.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
