Multi-Transform Domain Fusion Architecture for High-Speed Train Bogie Fault Diagnosis in Speed-Varying Conditions
IEEE Transactions on Intelligent Transportation Systems · 2026-10 · 27 (10) · 12279-12288
DOI: 10.1109/tits.2026.3699673 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Intelligent Transportation Systems, 2026-10. Verfasst von Junxiao Ren und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 27 Literaturangaben. Gefördert von: Sichuan Provincial Science and Technology Program, Southwest University of Science and Technology Ph.D. Foundation.
Abstract (Originalsprache)
Fault diagnosis of high-speed train bogies faces unprecedented challenges in speed-varying conditions, primarily manifesting in nonlinear dynamic characteristic variations, velocity-dependent fault signatures, and limited adaptability of existing methodologies. To address these critical challenges, this research proposes a novel multi-transform domain fusion architecture. This innovative framework synergistically integrates temporal domain feature with Fast Wavelet Transform and Gramian Angular Field to capture frequency-domain and time-frequency domain characteristics, respectively, while implementing a bespoke interactive attention mechanism for dynamic multi-domain feature fusion. Empirical validation on our curated dataset demonstrates remarkable performance metrics, achieving 100% diagnostic accuracy under iso-velocity conditions. Notably, the framework maintains 97.00% accuracy in the most challenging high-velocity scenarios. In comprehensive velocity condition testing, our methodology achieves a superior accuracy of 99.00%, significantly outperforming existing state-of-the-art approaches. Feature visualization analytics further substantiate the framework’s robust feature extraction capabilities across diverse velocity conditions. The proposed architecture provides a highly transferable approach for fault diagnosis of high-speed train and other complex industrial systems operating under speed-varying conditions.
Förderung
- Sichuan Provincial Science and Technology Program
- Southwest University of Science and Technology Ph.D. Foundation
Zitieren
APA
Ren, J., Gao, X., Li, Q., Li, J., & Li, Z. (2026). Multi-Transform Domain Fusion Architecture for High-Speed Train Bogie Fault Diagnosis in Speed-Varying Conditions. IEEE Transactions on Intelligent Transportation Systems, 27(10), 12279-12288. https://doi.org/10.1109/tits.2026.3699673
Vancouver
Ren J, Gao X, Li Q, Li J, Li Z. Multi-Transform Domain Fusion Architecture for High-Speed Train Bogie Fault Diagnosis in Speed-Varying Conditions. IEEE Transactions on Intelligent Transportation Systems. 2026;27(10):12279-12288. doi:10.1109/tits.2026.3699673
BibTeX
@article{Ren2026,
author = {Ren, Junxiao and Gao, Xuefeng and Li, Qiang and Li, Junlin and Li, Zengyan},
title = {Multi-Transform Domain Fusion Architecture for High-Speed Train Bogie Fault Diagnosis in Speed-Varying Conditions},
journal = {IEEE Transactions on Intelligent Transportation Systems},
year = {2026},
volume = {27},
number = {10},
pages = {12279--12288},
doi = {10.1109/tits.2026.3699673}
}
EUVIMED