Robotization of eddy current surface inspection in the aerospace industry
Bibliographic record
Abstract
Service-Induced cracks in metallic aircraft structures caused by fatigue are a common problem. If fatigue cracks remain undetected long enough and therefore untreated, they can cause a catastrophic failure of the structure, which threatens the safety of flight. To prevent failure from happening and to detect these possible fatigue cracks, periodic inspections are carried out. Eddy current testing is the primary non-destructive testing method used in the aerospace industry to detect such surface opening fatigue cracks in mostly aluminium and titanium structures. Traditionally the eddy current inspection is conducted manually by non-destructive testing inspector. Automating non-destructive inspection, which is necessary but expensive, time consuming and prone to human factors process, could provide many benefits over the manual inspection in certain inspection areas that are well suited for automation. The most important benefit would be the consistency of the inspections. When the inspection process is consistent, the quality of it can be measured and therefore the reliability can be assessed. Time consumption of the inspection would not be necessarily considerably reduced, unless the batch sizes are large but the time during which the inspection is conducted could be assigned for example during off business hours when it does not interfere with other work tasks. The human factors, such as fatigue and stress would not be factoring to the results of the inspection, which would increase the consistency and reliability of the inspection. Thanks to the lesser workload on the non-destructive testing inspectors and interference to other maintenance tasks the costs of eddy current inspection would be reduced. The aim of this thesis was to study how eddy current inspection of the CASE specimens could be automated. Literature review was conducted to find out how non-destructive methods have been automated previously in the industry. It was discovered that the use of industrial robots to conduct the inspection was a viable solution in many cases with different non-destructive methods, including ultrasonic testing and eddy current testing. It was suspected that additional sensors were necessary for the robot to carry out the inspection. A force torque sensor was used to ensure that the robot was able to conduct the inspection properly. Based on the findings of the literature review an experimental robotized eddy current surface inspection was conducted on the two CASE specimens provided by the client company. Preparation for the experiment included manufacturing tool holder for the robot used in the experiment. Also fixtures for the CASE specimens were manufactured. The most significant results when analysing the viability of applying robotized inspection for the CASE specimen are the quality of eddy current inspection signal and the inspection coverage. The inspection signal quality was analysed against a reference calibration signal received from a standard calibration block. The inspection coverage on the other hand was analysed and measured visually, based on videos, photographs and physical measurements taken from the experiment.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 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 | 0.001 |
| 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 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".