Current Signature-Based Bearing Fault Severity Classification Using a Robust Multilevel Cascaded Framework
Bibliographic record
Abstract
Bearing faults are a critical concern in electrical machines, particularly permanent magnet synchronous motors (PMSMs), commonly used in electric vehicles. Early and accurate classification of bearing fault severity is essential for predictive maintenance, as it enhances cost-effectiveness, ensures safety, and extends product life. Although vibration-based monitoring offers rich diagnostic information, it remains costly and requires excess modifications. In contrast, current-based non-invasive techniques offer advantages in cost and integration but face challenges with accuracy due to operational complexities. This study presents six distinct artificial neural networks (ANNs)-based cascaded classification schemes for bearing fault severity classification. Discrete wavelet transform (DWT) with Symlet (Sym) is used for multi-resolution feature extraction in currents, combined with motor speed data to generate multi-channel features. These features are fed into an ANN-based level I algorithm using various fusion techniques, offering a more interpretable algorithmic framework. One approach employs a multi-input ANN for level I, integrated with an ANN-based level II for refined severity classification. This two-level cascaded approach achieves an accuracy over 99% on the Paderborn University dataset in various operational scenarios. The model is trained and analyzed using MATLAB. The proposed cascaded algorithms outperform single-stage models, and enhanced signal preprocessing improves accuracy and noise resilience. Additionally, the proposed risk-based performance indicator offers insights into maintenance strategies, while an optimum algorithm selection framework identifies an algorithm by considering a trade-off between computational complexity and accuracy.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".