Employing a U-net convolutional neural network for segmenting impact damages in optical lock-in thermography images of CFRP plates

Seibert, Artur Antonio; Borges, Vicente K.; Albertazzi, Armando; Schmitt, Robert

doi:10.1080/10589759.2020.1758099

Cited by 27 publications

(7 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the duration of the excitation and its shape, active thermography is divided into short pulse, long pulse, and phase thermography, with a sinusoidal shape, also known as the lock-in method. Lock-in thermography was used for segmenting impact damage of carbon fiber-reinforced plastics, in fields such as aerospace [ 17 ]. Some studies compared the use of various forcings, trying to identify areas of use.…”

Section: Applied Diagnostic Methodsmentioning

confidence: 99%

A Diagnostic Method Based on Active Thermography for the Degradation Assessment of Power Plant Boiler Tubes

Zator

Tomaszewski

Lasar

2022

Sensors

View full text Add to dashboard Cite

Proper maintenance management of power infrastructure requires inspections, in order to gather knowledge about the facility’s current condition. For this purpose, periodic diagnostic tests are carried out, not only to determine the current state, but to also predict future conditions, and subsequently plan for the scope of necessary repair work. Currently, in the case of heat screens of power boilers, the diagnostic process takes many days, is very expensive, and usually does not cover the entire screen area. Therefore, it is necessary to develop new, noninvasive diagnostic methods. This study presents the concept and research for an alternative method of locating places with excessive energy boiler screen degradation that require replacement. It was assumed that the new method should be fast, require no scaffolding assembly, and permit checking of the entire screen surface; this is unlike the methods used currently, which require manual checking of selected screen tubes. The proposed method is based on active thermography, in which heat flux is forced by the liquid flowing inside the screen. Tests were carried out based on a model of an axial-symmetric system in the form of a tube, with controlled reductions in the wall thickness. An experiment was carried out many times by recording the pipe surface temperature in many characteristic places (different thicknesses of the tube walls) with a thermal imaging camera. A temperature change was forced by a controlled flow of hot or cold water. The methods of analysis were proposed and verified, allowing firstly, detection of places with a reduced wall thickness, and secondly, estimations of the wall thickness (i.e., excessive degradation). For the best-proposed model (one of the four analyzed), all of the thickness changes were detected, and the limit error of thickness obtained was 0.3 mm.

show abstract

Section: Applied Diagnostic Methodsmentioning

confidence: 99%

A Diagnostic Method Based on Active Thermography for the Degradation Assessment of Power Plant Boiler Tubes

Zator

Tomaszewski

Lasar

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…In the last decade, several groups studied the use of artificial intelligence tools for NDT&E. Oliveira et al presented a transfer learning case in [ 24 ] with the application of a U-Net convolutional neural network, which was optimised for processing medical images, for segmenting impact damages in infrared phase images of carbon fibre reinforced plastic plates, which were acquired using optical lock-in thermography. Bang et al [ 25 ] proposed a framework for identifying defects in composite materials by integrating a thermography test with a deep learning technique.…”

Section: Methodsmentioning

confidence: 99%

A Deep Learning Method for the Impact Damage Segmentation of Curve-Shaped CFRP Specimens Inspected by Infrared Thermography

Wei

Fernandes

Herrmann

et al. 2021

Sensors

View full text Add to dashboard Cite

Advanced materials such as continuous carbon fiber-reinforced thermoplastic (CFRP) laminates are commonly used in many industries, mainly because of their strength, stiffness to weight ratio, toughness, weldability, and repairability. Structural components working in harsh environments such as satellites are permanently exposed to some sort of damage during their lifetimes. To detect and characterize these damages, non-destructive testing and evaluation techniques are essential tools, especially for composite materials. In this study, artificial intelligence was applied in combination with infrared thermography to detected and segment impact damage on curved laminates that were previously submitted to a severe thermal stress cycles and subsequent ballistic impacts. Segmentation was performed on both mid-wave and long-wave infrared sequences obtained simultaneously during pulsed thermography experiments by means of a deep neural network. A deep neural network was trained for each wavelength. Both networks generated satisfactory results. The model trained with mid-wave images achieved an F1-score of 92.74% and the model trained with long-wave images achieved an F1-score of 87.39%.

show abstract

“…erefore, the document records that in the distributed control mode, the control structure of the high-speed communication network will form a distributed computing economic system [8]. e document records that the roughly independent units can be differentiated according to the application system, and their master-slave relationship has autonomy for these units [9].…”

Section: Related Workmentioning

confidence: 99%

Implementation of Dynamic Load Balancing in Distributed System Based on Improved Algorithm

Zhou

2022

Mobile Information Systems

View full text Add to dashboard Cite

The digitalization and networking of the operating status of manufacturing equipment and facilities are still carried out on the basis of the automatic measurement system. Combining the embedded microprocessor AD/DA device and the network controller into an IC chip not only can solve the technical problems between the embedded microcontroller and the Internet but also can reduce the cost of this connection to a minimum limit. Embedded intelligent technology, especially the development and application of agent technology, makes the monitoring system change from a central computing model to a distributed model. With the application of computing from centralized development to distributed development. CBA has also transitioned from static central control to dynamic distributed control. The system load balancing method, distributed in the system processor, can enhance the capabilities of all the instruments and equipment in the system; eliminate the imbalance between busy and idle, and improve the overall performance of the system. The purpose of this article is to analyze and experimentally verify the parallel computing methods of computer distributed systems, and it can be applied to parallel computing for a variety of environmental support in computer systems. An effective combination of distributed object technology and embedded system technology can build a model of a distributed parallel computing system. After the distributed parallel computing system retains the advantages of the previous system, the distributed availability of parallel computing systems has been greatly improved.

show abstract

Employing a U-net convolutional neural network for segmenting impact damages in optical lock-in thermography images of CFRP plates

Cited by 27 publications

References 21 publications

A Diagnostic Method Based on Active Thermography for the Degradation Assessment of Power Plant Boiler Tubes

A Diagnostic Method Based on Active Thermography for the Degradation Assessment of Power Plant Boiler Tubes

A Deep Learning Method for the Impact Damage Segmentation of Curve-Shaped CFRP Specimens Inspected by Infrared Thermography

Implementation of Dynamic Load Balancing in Distributed System Based on Improved Algorithm

Contact Info

Product

Resources

About