Research on an efficient prediction for deformations of thread connections based on deep transfer learning
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
Threaded connections are essential components in mechanical assemblies, subjected to complex deformations under various loading conditions (e.g., tension, torque, bending, and shear), influenced by bolt geometry and material properties. Accurate deformation prediction under complex loading conditions is critical for structural safety. This study presents an efficient method for rapid multi-dimensional deformation prediction in threaded connections under combined loading, utilizing deep transfer learning(TL) to address various bolt types and loading scenarios. A simplified static analysis model is first developed to predict deformations under combined loads, validated through comparisons with finite element analysis (FEA) results. A deep neural network (DNN) is pre-trained on a large deformation dataset to learn complex load-deformation relationships. TL is applied to leverage knowledge from the pre-trained model, improving prediction accuracy and generalization across bolt types. Latin Hypercube Sampling (LHS) is used to generate a sparse dataset for bolts with varying geometries and materials. Experimental results demonstrate that this method reduces computational costs and accurately simulates nonlinear deformations. For complex loading conditions, the proposed model requires only 0.14% of the computation time compared to the FEM (4.2 s vs. 3000 s). For different bolt types, the prediction accuracy reaches up to 96.6% after transfer learning. This approach provides a cost-effective alternative to computationally intensive FEA, enabling real-time, large-scale applications in engineering design and structural assessment.
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 | 0.002 | 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