Defect Sizing Using Convolution Neural Network Applied to Guided Wave Imaging
Bibliographic record
Abstract
The lightweight aluminum alloys are extensively used in the aerospace industry. These materials are used for constructing complex structures such as aircraft fuselage due to their excellent strength-to-weight ratio, stiffness, and corrosion resistance. How- ever, defects such as corrosion or fractures can appear because of thermo-mechanical aging in a hostile working environment or impact forces due to the improper use of these structures. In light of this, Guided Waves (GWs)-based Structural Health Monitoring (SHM) system can be considered as a promising solution for the structural integrity screening, maintenance costs reduction and prolongation of the service time of these materials. In general, a sparse array of PZT transducers can be used for GWs exciting and sensing, and GWs Imaging (GWI) algorithms, such as Excitelet, can be used to process the measured signals. This imaging technique allows computing high-resolution images that represent the integrity of the structure, where each pixel of the image is mapped to the elementary portion of the structure and carries a Damage Index (DI) value. While defect presence and location can be determined from visual inspection of the image by naked eye, the defect sizing is a more complex problem due to non-linear behaviour of DI values regarding the defect size and its location. This paper proposes an approach for defect size evaluation. It relies on the extensive GWI database generated by means of Spectral Finite Elements modelling method implemented in CIVA and Convolution Neural Network (CNN) trained on numerical data. The CNN is used to build an accurate inversion model that takes GWI sample as input and determines the size of the defect. The model is tested on the simulated data and validated by means of experiment in aluminum plate.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| 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".