Physics-Guided Dual-Stream Fusion for Extreme Few-Shot Fault Diagnosis Under Massive Domain Shifts
Processes · 2026-06-20 · 14 (12) · 2012
DOI: 10.3390/pr14122012 ↗At a glance
Journal article in Processes, published online on 2026-06-20. Written by Shiqian Wu and 4 co-authors. 48 references. The full text is freely available (open access, CC BY 4.0). Funded by: Science and Technology Research Project of Jiangxi Provincial Department of Education.
Abstract (original language)
Reliable fault diagnosis of rotating machinery is critical for averting serious failures in modern industrial systems. While data-driven deep learning has advanced condition monitoring, its success is fundamentally predicated on the availability of independent and identically distributed (I.I.D.) datasets. In realistic operational environments, machinery frequently experiences massive domain shifts induced by varying rotational speeds. Concurrently, acquiring high-fidelity fault instances is limited compared to abundant healthy baseline data, often resulting in a long-tailed distribution. Under such data-starved conditions, conventional few-shot domain adaptation (FSDA) methodologies often may be affected by distributional erasure; global alignment objectives are mainly driven by the healthy majority, causing sparse fault signatures to be erroneously absorbed as noise and leading to severe diagnostic performance degradation. To address this setting, this study develops a physics-guided dual-stream fusion framework for extreme few-shot cross-domain fault diagnosis. The method does not treat the Laplace wavelet, STFT, CNNs, or AdaBN as newly introduced techniques. Instead, it integrates these components into a unified diagnostic pipeline designed for long-tailed target support sets under large speed shifts. A learnable Laplace wavelet convolution is used in the temporal branch to emphasize transient impact responses, while STFT spectrograms provide a complementary time-frequency representation for the two-dimensional branch. The two feature streams are then fused for target fault classification. For domain adaptation, a Strict AdaBN strategy is applied using only the target support set, rather than the target test data or a large unlabeled target pool. Under the evaluated 50 healthy + 12 fault support condition, the healthy samples provide target-domain operating-background statistics for BN recalibration, while the limited fault samples are used for supervised classifier adjustment. Experiments on the HUSTbearing and Torino DIRG datasets show that the proposed integrated framework achieves stable performance under the evaluated few-shot cross-speed settings. These results suggest that combining physics-guided Laplace convolution, time-frequency representations, and support-set-restricted BN recalibration can be useful for bearing fault diagnosis when target fault samples are limited.
Funding
- Science and Technology Research Project of Jiangxi Provincial Department of Education
Cite
APA
Wu, S., Zhang, W., Liu, H., Lu, Y., & Zhang, Y. (2026). Physics-Guided Dual-Stream Fusion for Extreme Few-Shot Fault Diagnosis Under Massive Domain Shifts. Processes, 14(12), 2012. https://doi.org/10.3390/pr14122012
Vancouver
Wu S, Zhang W, Liu H, Lu Y, Zhang Y. Physics-Guided Dual-Stream Fusion for Extreme Few-Shot Fault Diagnosis Under Massive Domain Shifts. Processes. 2026;14(12):2012. doi:10.3390/pr14122012
BibTeX
@article{Wu2026,
author = {Wu, Shiqian and Zhang, Weiming and Liu, Huiyu and Lu, Yuchen and Zhang, Yuxuan},
title = {Physics-Guided Dual-Stream Fusion for Extreme Few-Shot Fault Diagnosis Under Massive Domain Shifts},
journal = {Processes},
year = {2026},
volume = {14},
number = {12},
pages = {2012},
doi = {10.3390/pr14122012}
}
EUVIMED