A Bearing Fault Diagnosis Method Based on ECA-ConvNeXt and BWO-VMD
Bibliographic record
Abstract
In order to address the challenges of extracting meaningful features and enhancing the noise immunity and robustness of diagnostic models in rolling bearing fault diagnosis, a novel method is proposed.This method combines the variational mode decomposition (VMD) features extraction algorithm with the Beluga Whale Optimization Algorithm (BWO) and utilizes an improved ConvNeXt network featuring an efficient channel attention mechanism (ECA).Firstly, The BWO algorithm optimizes the number of mode decomposition and penalty factors in VMD, seeking the optimal parameter combination based on the highest permutation entropy fitness.It then performs VMD decomposition on the fault signal to extract the most representative sample features.Secondly, the Gram angle field (GAF) encoding method transforms the extracted one-dimensional feature signals into twodimensional features.At the same time, the ECA-Block module is specifically designed to update the Block module in the ConvNeXt network.ECA is introduced into the ConvNeXt network, the two-dimensional feature signal after GAF conversion is input into the ECA-ConvNeXt network fault diagnosis model for training, identification, and classification.Finally, the verification process for the original signal loading noise of the bearing vibration data set from Case Western Reserve University has been conducted.The results indicate that the ECA-ConvNeXt model, trained with BWO-VMD, demonstrates high accuracy in bearing fault diagnosis.The accuracy is 99.78% for identifying the original fault vibration signals, and the classification accuracy after loading different signal-to-noise ratio Gaussian white noise is above 99.57%.Experimental conditions varied across different datasets, including load and noise, in the model transfer experiments.The average recognition rate for load model transfer was 96.22%, while for noise model transfer, it was 98.22%, exceeding that of the comparative algorithm.Additionally, the recognition rate exceeded expectations and bit the least fluctuation.The experiments demonstrate that the proposed method possesses excellent noise resilience and robustness.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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 teacher head, 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".