What does this research mean for the field?

Integrating physics-informed fault characterization with image-based machine learning on motor current spectrograms provides an effective, interpretable, and cost-efficient method for detecting and localizing rolling bearing faults. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To develop an interpretable, sensorless diagnostic framework for early detection and localization of rolling bearing faults using motor current signals.

August 24, 2026International Journal of Computational Intelligence SystemsOpen Access

Hierarchical Approach to Rolling Bearing Failure Detection Using Spectrogram Images Classification

View Full Paper

Ask AI

Bookmark

Authors

AFArantzazu FlórezUniversity of the Basque Country GEGoretti EchegarayUniversity of the Basque Country AFAne M. Florez-TapiaVicomtech

Discussion

Member takes

Implication

Validation study demonstrates hierarchical spectrogram classification detects bearing faults in rotating machinery, highlighting cost-effective industrial monitoring.

Key Points

To develop an interpretable, sensorless diagnostic framework for early detection and localization of rolling bearing faults using motor current signals.
Converted raw stator current signals into 2D time-frequency spectrogram images to capture transient mechanical defect signatures.
Constructed a three-stage hierarchical framework incorporating domain-specific characteristic fault frequencies to guide health assessment and fault localization.
Extracted global image descriptors from spectrograms to classify conditions as healthy or faulty and pinpoint internal versus external fault locations on a benchmark bearing dataset.
Successfully differentiated healthy from damaged bearings and accurately localized internal versus external faults without requiring dedicated vibration sensors.
Achieved competitive diagnostic accuracy while improving model interpretability by integrating rule-based physical fault frequencies with image-based machine learning.

Cite This Study

Flórez et al. (2026) studied this question.

synapsesocial.com/papers/6a8c00d0bca056c88e6dfd62 https://doi.org/10.1007/s44196-026-01528-0

View Full Paper

Ask AI

Bookmark

Implication

Validation study demonstrates hierarchical spectrogram classification detects bearing faults in rotating machinery, highlighting cost-effective industrial monitoring.

Hierarchical Approach to Rolling Bearing Failure Detection Using Spectrogram Images Classification

Discussion

Member takes

Implication

Key Points

Cite This Study

Hierarchical Approach to Rolling Bearing Failure Detection Using Spectrogram Images Classification

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider