| Title |
UT data analysis steps for development of automated detection technique of bonding defects in multi-layered structures / |
| Authors |
Smagulova, D ; Jasiuniene, E ; Yilmaz, B ; Mazeika, L |
| Full Text |
|
| Is Part of |
e-Journal of nondestructive testing: 13th European conference on non-destructive testing (ECNDT 2023), 3-7 July 2023, Lisbon, Portugal.. Mayen : NDT.net. 2023, vol. 28, spec. iss. 8, p. 1-2.. ISSN 1435-4934 |
| Keywords [eng] |
data post-processing ; automatic analysis ; ultrasonic ; adhesive joints ; interface defects |
| Abstract [eng] |
Multi-layered structures of adhesively bonded similar and dissimilar materials have a high interest in aerospace and automotive industries. Adhesive joining is more advanced technology which allows to join different types of materials, provide uniform load distribution and resistance to corrosion. Adhesive joining and composite materials are the key technology and materials which provide lightening of the structure as well as reduce CO2 emission. Adhesively bonded similar and dissimilar materials are usually used in secondary load carriers due to their advanced mechanical characteristics. However, this structures are still limited in use due to the absence of reliable method suitable for investigation of multi-layered structures and, especially, dissimilar material joints. It is problematic to detect bonding defects in multi-layered structures without complex modelling of sound propagation performed by numerical methods. This factor limits practical application of the technique. Ultrasonic testing (UT) is the widely used non-destructive method for inspection of different materials and structures. A lot of research has been carried out to evaluate adhesion quality using ultrasonic inspection. This technique is one of the most suitable techniques for investigation of multi-layered structures. However, complex modelling of wave propagation is almost always required for the detection of bonding defects due to uncertainty factors such as inspection execution, component geometry, surface preparation, thickness of adhesive and adherents and signal reverberation which influence on the detection of interface defects. Therefore, the complex modelling and numerical evaluations can be eliminated by developing a robust method for automatic analysis of ultrasonic NDT data for detection of bonding defects in multi-layered structures. As a result, the objective of this work is to create steps of analysis of UT data for development of automated defect detection technique. The experimental investigation of adhesively bonded materials was performed using ultrasonic inspection. The collected data was analyzed. Influential factors which have an impact on the interface defect detection were determined. The valuable features which are capable to increase probability of interface defect detection and features related to the defects for development of data post-processing technique were determined and extracted. The basic steps of analysis and algorithm of automated detection of interface defects are provided in this work. The reliability of the technique was evaluated and detected defects were sized. As a result, application of multi-dimensional data fusion of different non-destructive methods has provided better evaluation results and increased the reliability of the technique. |
| Published |
Mayen : NDT.net |
| Type |
Conference paper |
| Language |
English |
| Publication date |
2023 |
| CC license |
|
