Gear Tooth Crack Detection Method Using Bayesian Neural Network
Bibliographic record
Abstract
Gearbox fault detection plays a crucial role in implementing proactive maintenance strategies and reducing economic losses. Fault detection can be addressed by modeling baseline monitoring data and subsequently detecting faults through deviations between the baseline model and newly acquired monitoring data. In the field of deep learning, Long Short-Term Memory (LSTM) have been widely applied to nonlinear time series modeling. This paper attempts to combine Bayesian Neural Networks (BNN) with LSTM, aiming to fully exploit the ability of LSTM to capture complex long-term dependencies as well as the strong regularization and generalization capabilities of BNN. Specifically, we propose two fault detection methods: the fault detection method based on Bayesian LSTM (BLSTM) and the fault detection method based on traditional LSTM combined with Bayesian fully connected layer (LSTM-BFC). Comparative and anti-noise experiments were conducted using data collected from a gearbox test rig. Experimental results demonstrate that the proposed Bayesian Neural Network-based approaches outperform traditional LSTM models in terms of fault detection performance. Among them, the LSTM-BFC method shows particularly outstanding performance in training time, time series prediction capability, fault detection accuracy, and noise 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 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 | 0.003 |
| 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 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".