A Collaborative Denoising and Multisensor Local–Global Fusion Framework for Intelligent Fault Diagnosis of Rotating Machinery Under Noisy Environments
IEEE Transactions on Instrumentation and Measurement · 2026 · 75 · 3522020-3522020
DOI: 10.1109/tim.2026.3720915 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Instrumentation and Measurement, 2026. Verfasst von Shijie Lu und 4 weiteren Autor:innen. Die Arbeit nennt 67 Literaturangaben.
Abstract (Originalsprache)
Accurate measurement and fault diagnosis of rotating machinery are paramount for the safe and reliable monitoring of modern industrial systems. In practical sensing applications, however, accurate state detection is severely challenged by strong background noise and the complexity of physical fault signatures, especially concerning the effective processing and fusion of multisensor measurement data. To tackle this industrial measurement and monitoring challenge, we propose a novel measurement data-processing method, named collaborative denoising and multisensor local–global fusion framework (CDMSLGFF). At its core, the proposed method introduces an innovative dual-branch feature extractor to effectively characterize the mechanical condition: one branch performs deep feature denoising and extracts local physical details from sensor readings via a multiscale refinement denoising module (MRDM), while the other captures global contextual dependencies of the measurement data using a discrete wavelet transform (DWT) and a Transformer architecture. This differential design ensures synergistic noise suppression and complementary feature extraction from multiple sensing perspectives. Subsequently, a hierarchical adaptive fusion module (HAFM) is proposed to intelligently screen, enhance, and reconstruct characteristics from both branches, generating a robust fused representation with high discriminability and noise invariance. Extensive experiments on real-world industrial (RWI) and public measurement datasets demonstrate that the proposed method provides a novel and superior monitoring and detection solution, significantly advancing multisensor information fusion and fault diagnosis reliability in harsh measurement environments.
Zitieren
APA
Lu, S., Qin, B., Zhang, Q., Qiu, Y., & Sun, G. (2026). A Collaborative Denoising and Multisensor Local–Global Fusion Framework for Intelligent Fault Diagnosis of Rotating Machinery Under Noisy Environments. IEEE Transactions on Instrumentation and Measurement, 75, 3522020-3522020. https://doi.org/10.1109/tim.2026.3720915
Vancouver
Lu S, Qin B, Zhang Q, Qiu Y, Sun G. A Collaborative Denoising and Multisensor Local–Global Fusion Framework for Intelligent Fault Diagnosis of Rotating Machinery Under Noisy Environments. IEEE Transactions on Instrumentation and Measurement. 2026;75:3522020-3522020. doi:10.1109/tim.2026.3720915
BibTeX
@article{Lu2026,
author = {Lu, Shijie and Qin, Binbin and Zhang, Qinghua and Qiu, Yongqi and Sun, Guoxi},
title = {A Collaborative Denoising and Multisensor Local–Global Fusion Framework for Intelligent Fault Diagnosis of Rotating Machinery Under Noisy Environments},
journal = {IEEE Transactions on Instrumentation and Measurement},
year = {2026},
volume = {75},
pages = {3522020--3522020},
doi = {10.1109/tim.2026.3720915}
}
EUVIMED