Wavelet coherence-aware multi-branch deep ensemble for fault identification in centrifugal pumps
- 1Department of Electrical, Electronic and Computer Engineering, University of Ulsan
Abstract
Reliable fault diagnosis of centrifugal pumps is challenging due to the nonstationary nature of vibration signals, weak early-stage laboratory fault signatures, and overlapping characteristics among different mechanical defects. This study proposes a wavelet coherence-aware multi-branch deep ensemble framework that integrates physically meaningful time–frequency coupling with complementary deep feature learning. Multi-channel vibration signals are transformed into two-dimensional wavelet coherence maps to emphasize localized inter-sensor phase-consistent structures induced by mechanical processes. Three lightweight and architecturally diverse convolutional neural networks are trained in parallel to extract fine-scale, global, and compact structural features. Their outputs are fused through a probabilistic soft-voting strategy to improve robustness and decision stability. The framework is evaluated on vibration datasets collected from a PMT-4008 centrifugal pump test bench under three operating pressures (3.0, 3.5, and 4.0 bar). The results demonstrate consistent and reliable fault discrimination across all investigated conditions, with strong class separability confirmed by Receiver Operating Characteristic analysis and feature-space visualization. These findings demonstrate the effectiveness of the proposed framework for centrifugal pump fault diagnosis within the investigated experimental setup and operating conditions.
Related
- Application of a multimodal model integrating optical coherence tomography angiography (OCTA) and radiomics for the identification of high-risk Nonarteritic anterior ischaemic optic neuropathy (NAION) populations and for distinguishing its acute phase.
- Development of a risk identification model for sPTB in patients with GDM based on ensemble learning algorithms.
- Central Centrifugal Cicatricial Alopecia in Black Women: Diagnostic Challenges and Molecular Insights.
- Design and verification of an interlocked centrifugal exhaust system for hatch-type boiling disinfectors
- Tricuspid annular plane systolic excursion/pulmonary vascular resistance index predicts right ventricular assist device requirement and mortality after magnetically levitated centrifugal-flow left ventricular assist device implantation.
BibTeX
@article{saleem2026wavelet,
title = {{Wavelet coherence-aware multi-branch deep ensemble for fault identification in centrifugal pumps}},
author = {Faisal Saleem and Muhammad Umar and Jong-Myon Kim},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62260-5},
url = {https://doi.org/10.1038/s41598-026-62260-5},
note = {CC BY},
}
RIS
TY - JOUR
TI - Wavelet coherence-aware multi-branch deep ensemble for fault identification in centrifugal pumps
AU - Faisal Saleem
AU - Muhammad Umar
AU - Jong-Myon Kim
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62260-5
UR - https://doi.org/10.1038/s41598-026-62260-5
AB - Reliable fault diagnosis of centrifugal pumps is challenging due to the nonstationary nature of vibration signals, weak early-stage laboratory fault signatures, and overlapping characteristics among different mechanical defects. This study proposes a wavelet coherence-aware multi-branch deep ensemble framework that integrates physically meaningful time–frequency coupling with complementary deep feature learning. Multi-channel vibration signals are transformed into two-dimensional wavelet coherence maps to emphasize localized inter-sensor phase-consistent structures induced by mechanical processes. Three lightweight and architecturally diverse convolutional neural networks are trained in parallel to extract fine-scale, global, and compact structural features. Their outputs are fused through a probabilistic soft-voting strategy to improve robustness and decision stability. The framework is evaluated on vibration datasets collected from a PMT-4008 centrifugal pump test bench under three operating pressures (3.0, 3.5, and 4.0 bar). The results demonstrate consistent and reliable fault discrimination across all investigated conditions, with strong class separability confirmed by Receiver Operating Characteristic analysis and feature-space visualization. These findings demonstrate the effectiveness of the proposed framework for centrifugal pump fault diagnosis within the investigated experimental setup and operating conditions.
SN - 2045-2322
ER -