A phase-adaptive weighted approach for self-data-driven RUL prediction of machinery
Measurement Science and Technology · 15.05.2026 · 37 (20) · 206105
DOI: 10.1088/1361-6501/ae693c ↗Kurz gefasst
Fachartikel in der Zeitschrift Measurement Science and Technology, online erschienen am 18.05.2026. Verfasst von Runzhong Fang und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 31 Literaturangaben. Gefördert von: Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China, National Natural Science Foundation of China.
Abstract (Originalsprache)
Self-data-driven remaining useful life (RUL) prediction is vital for industrial assets, yet deploying such methods faces certain limitations: the computational inefficiency of conventional batch-update strategies and the inability to capture multi-stage deterioration dynamics. To reconcile high-fidelity tracking with real-time computational constraints, this paper proposes a phase-adaptive, computation-aware prognostic framework. First, to address the initialization bias caused by imbalanced historical data, a phase-adaptive expectation maximization algorithm (PAEM) is introduced. PAEM extracts robust priors by enforcing smoothness and monotonicity constraints, preventing divergence during the offline-to-online handover. Second, for the online stage, an entropy-driven particle filter is developed to recursively estimate the RUL without reprocessing historical data. Utilizing posterior entropy, it adaptively fuses candidate models and dynamically allocates particles to concentrate resources where uncertainty rises. Experimental studies on the XJTU-SY bearing dataset and a customized harmonic drive dataset demonstrate that the proposed framework not only achieved higher prediction accuracy but also required less computation time compared to the existing methods, confirming its suitability for online industrial applications.
Förderung
- Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China
- National Natural Science Foundation of China
Zitieren
APA
Fang, R., Gao, Y., Lei, Y., Yang, B., Li, X., & Li, N. (2026). A phase-adaptive weighted approach for self-data-driven RUL prediction of machinery. Measurement Science and Technology, 37(20), 206105. https://doi.org/10.1088/1361-6501/ae693c
Vancouver
Fang R, Gao Y, Lei Y, Yang B, Li X, Li N. A phase-adaptive weighted approach for self-data-driven RUL prediction of machinery. Measurement Science and Technology. 2026;37(20):206105. doi:10.1088/1361-6501/ae693c
BibTeX
@article{Fang2026,
author = {Fang, Runzhong and Gao, Yang and Lei, Yaguo and Yang, Bin and Li, Xiang and Li, Naipeng},
title = {A phase-adaptive weighted approach for self-data-driven RUL prediction of machinery},
journal = {Measurement Science and Technology},
year = {2026},
volume = {37},
number = {20},
pages = {206105},
doi = {10.1088/1361-6501/ae693c}
}
EUVIMED