2022
DOI: 10.1088/1361-665x/ac66aa
|View full text |Cite
|
Sign up to set email alerts
|

A transfer learning approach for damage diagnosis in composite laminated plate using Lamb waves

Abstract: Lamb wave-based damage diagnosis systems are widely regarded as a likely candidate for real-time structural health monitoring (SHM), although analysing the Lamb wave response is still a challenging task due to its complex physics. Recently, deep learning (DL) models such as convolutional neural network (CNN) have shown robust classification performance in various structures using Lamb wave based diagnostic strategies. However, these DL models are often designed to address isolated tasks, which means that the m… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
6
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
7
1

Relationship

0
8

Authors

Journals

citations
Cited by 20 publications
(6 citation statements)
references
References 52 publications
0
6
0
Order By: Relevance
“…According to the above theory, the length of the near field of a Lamb wave is the same as the p-wave sound field, which is inversely proportional to the wavelength and directly proportional to the area of the wave source. The length N of the near-field region in the second medium can be deduced from Equation (11).…”
Section: Lamb Wave Sound Field Characteristicsmentioning
confidence: 99%
See 2 more Smart Citations
“…According to the above theory, the length of the near field of a Lamb wave is the same as the p-wave sound field, which is inversely proportional to the wavelength and directly proportional to the area of the wave source. The length N of the near-field region in the second medium can be deduced from Equation (11).…”
Section: Lamb Wave Sound Field Characteristicsmentioning
confidence: 99%
“…Substituting them in Equation ( 10), we could get N = 7.50 mm. From Equation (11), we could get N = 0.03 mm, which we could ignore. At this time, there was almost no s-wave near-field area in the second medium.…”
Section: Selection Of Wedge Thicknessmentioning
confidence: 99%
See 1 more Smart Citation
“…DL needs a lot of data to train. However, it is usually very difficult to collect sufficient structural damage data in real word scenarios [21]. The problem of training a DL network with few labeled samples (e.g.…”
Section: Introductionmentioning
confidence: 99%
“…Inspection methods, such as X-ray tomography [1][2][3] and ultrasonic scanning, [4][5][6] are very effective for detecting defects in composites but require removing the part from the structure and are not suited for real-time monitoring. Nondestructive methods, such as acoustic emissions 3,7 and ultrasonic-guided waves (Lamb waves), 8 offer the advantages of in situ and real-time monitoring of the structures. However, those techniques require additional devices (sensors) attached to the analyzed structure and generally detect damage only at their immediate vicinity, thus limiting the sensing area.…”
Section: Introductionmentioning
confidence: 99%