Application of S-Relu activation function and adaptive dual-threshold noise reduction in fault diagnosis of RV reducer rolling bearings
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Abstract In many complex operating environments of mechanical equipment, the collected vibration signals are easily contaminated by noise, which increases difficulty in fault diagnosis. To overcome the challenge of accurately diagnosing rolling bearing faults in rotate vector (RV) reducer under strong noise conditions, a fault diagnosis model with smooth Rectified Linear Unit (S-Relu) activation function and adaptive dual threshold noise reduction is proposed. Firstly, the receptive field of the neural network model is expanded by the dilated causal convolution, while preserving the temporal relationships within the data. Then, the S-Relu activation function is proposed to solve the limitations existing in the traditional activation function. Finally, an adaptive dual-threshold noise reduction method is proposed to mitigate the influence of a strong noise environment. The method adaptively adjusts the threshold according to the dynamic characteristics of the signal, and continuously reduces the noise through the shortcut connection method to filter out the noise thus enhancing the key fault feature information. The proposed method is validated with vibration signals collected from experiments of an RV reducer in the laboratory. Under the strong noise with signal-to-noise ratio of −8 dB, the diagnostic accuracy of the proposed method under different loads (0 N∙m, 625 N∙m, and 1250 N∙m) reaches 94.4%, 84.5%, and 80.3%, respectively. The classification accuracy of the proposed method is more than 99.38% without noise. The results show that the accuracy and reliability of the proposed method are significantly higher than that deduced by other commonly used advanced methods.
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.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it