You are currently viewing a new version of our website. To view the old version click .
Machines
  • This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
  • Article
  • Open Access

13 October 2025

Phase-Aware Complex-Spectrogram Autoencoder for Vibration Preprocessing: Fault-Component Separation via Input-Phasor Orthogonality Regularization

,
,
,
,
and
1
Department of Ocean System Engineering, College of Marine Science, Gyeongsang National University, 11-dong, 2, Tongyeonghaean-ro, Tongyeong-si 53064, Gyeongsangnam-do, Republic of Korea
2
Department of Naval Architecture and Ocean Engineering, College of Marine Science, Gyeongsang National University, 11-dong, 2, Tongyeonghaean-ro, Tongyeong-si 53064, Gyeongsangnam-do, Republic of Korea
*
Author to whom correspondence should be addressed.
Machines2025, 13(10), 945;https://doi.org/10.3390/machines13100945 
(registering DOI)
This article belongs to the Section Machines Testing and Maintenance

Abstract

We propose a phase-aware complex-spectrogram autoencoder (AE) for preprocessing raw vibration signals of rotating electrical machines. The AE reconstructs normal components and separates fault components as residuals, guided by an input-phasor phase-orthogonality regularization that defines parallel/orthogonal residuals with respect to the local signal phase. We use a U-Net-based AE with a mask-bias head to refine local magnitude and phase. Decisions are based on residual features—magnitude/shape, frequency distribution, and projections onto the normal manifold. Using the AI Hub open dataset from field ventilation motors, we evaluate eight representative motor cases (2.2–5.5 kW: misalignment, unbalance, bearing fault, belt looseness). The preprocessing yielded clear residual patterns (low-frequency floor rise, resonance-band peaks, harmonic-neighbor spikes), and achieved an area under the receiver operating characteristic curve (ROC-AUC) = 0.998–1.000 across eight cases, with strong leave-one-file-out generalization and good calibration (expected calibration error (ECE) ≤ 0.023). The results indicate that learning to remove normal structure while enforcing phase consistency provides an unsupervised front-end that enhances fault evidence while preserving interpretability on field data.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.