Fault Diagnosis of Centrifugal Pumps under Hydraulic Excitation and Flow Noise: A Critical Review of Vibration, Motor-Current, Acoustic-Emission, and Data-Driven Methods
Journal of Engineering · 01.10.2026 · 32 (10) · 174-197
DOI: 10.31026/j.eng.2026.10.08 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Engineering, online erschienen am 01.10.2026. Verfasst von Sharad R. Bhadane und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Centrifugal-pump diagnosis is complicated by flow-induced excitation from pressure pulsation, recirculation, rotating stall, cavitation, and turbulence, which can produce periodic, modulated, impulsive, and broadband components that overlap the signatures associated with bearing, shaft, impeller, and seal faults. This review covers 103 traceable publications on vibration, motor-current, audible-sound, acoustic-emission, signal-processing, and data-driven methods. The included evidence is classified as direct centrifugal-pump evidence, transferable rotating-machinery evidence, or contextual review evidence. It shows that signal processing is not equally robust across all pump configurations and operating conditions, and that the frequency position alone is insufficient to support a diagnostic opinion. Harmonics, sidebands, envelope content, time-domain waveform behavior, speed, and flow context each provide important information. Methods such as envelope analysis, spectral kurtosis, mode decomposition, modulation signal bispectrum, motor current analysis, acoustic emission, and deep learning may provide meaningful results with varying reliability depending on the fault-transmission path, sensor location, operational range, and validation scheme. Higher-order spectral methods and data-driven approaches show promise but have not been iterated widely among pumps or different operating conditions. An example workflow is to perform healthy baselines covering the intended envelope, and diagnostic reports with uncertainty boundaries and evidence boundaries. Limitations include non-exhaustive literature discovery, heterogeneity in test rigs and labels, restricted access to some primary data, and reliance on transferable evidence where pump-specific validation is unavailable. The results therefore support condition-aware, multi-indicator diagnosis rather than universal thresholds or a single preferred algorithm.
Zitieren
APA
Bhadane, S. R., Rao, K. R., & Purkar, S. T. (2026). Fault Diagnosis of Centrifugal Pumps under Hydraulic Excitation and Flow Noise: A Critical Review of Vibration, Motor-Current, Acoustic-Emission, and Data-Driven Methods. Journal of Engineering, 32(10), 174-197. https://doi.org/10.31026/j.eng.2026.10.08
Vancouver
Bhadane SR, Rao KR, Purkar ST. Fault Diagnosis of Centrifugal Pumps under Hydraulic Excitation and Flow Noise: A Critical Review of Vibration, Motor-Current, Acoustic-Emission, and Data-Driven Methods. Journal of Engineering. 2026;32(10):174-197. doi:10.31026/j.eng.2026.10.08
BibTeX
@article{Bhadane2026,
author = {Bhadane, Sharad R. and Rao, Kasi Raja and Purkar, Sanjay T.},
title = {Fault Diagnosis of Centrifugal Pumps under Hydraulic Excitation and Flow Noise: A Critical Review of Vibration, Motor-Current, Acoustic-Emission, and Data-Driven Methods},
journal = {Journal of Engineering},
year = {2026},
volume = {32},
number = {10},
pages = {174--197},
doi = {10.31026/j.eng.2026.10.08}
}
EUVIMED