MRPC eddy current flaw classification in tubes using deep neural networks

Park, Jinhyun; Han, Seong-Jin; Munir, Nauman; Yeom, Yun-Taek; Song, Sung‐Jin; Kim, Hak-Joon; Kwon, Segon

doi:10.1016/j.net.2019.05.011

Cited by 14 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As it is costly to obtain ECT data from objects with varied materials and dimensions, it is hence necessary to develop a new method that can be trained using much less data and show better robustness. In the realm of image analysis, researchers improved the generalization and robustness of models through data augmentation methods, such as adjusting brightness, random erasing, rotation, and so on (Park et al, 2019; Shorten & Khoshgoftaar, 2019).…”

Section: Related Workmentioning

confidence: 99%

An autonomous robot for shell and tube heat exchanger inspection

Zheng

Xie

et al. 2022

Journal of Field Robotics

View full text Add to dashboard Cite

Shell and tube heat exchangers (STHEs) are critical to energy conversion efficiency of power plants. Eddy current examination is a way to evaluate working conditions of these tubes. However, the current testing apparatus requires human to manually insert an eddy current testing (ECT) probe into and extract it out of individual tubes, and meanwhile monitor measurement results for diagnosis. It is a time‐consuming and labor‐intensive procedure even for an experienced technician. To tackle this challenge, in this study, we developed a robot enabled ECT system for autonomous inspection of STHEs. The robotic platform employs Mecanum wheeled chassis for high mobility, machine vision to locate tube bundle and tube inlets, a rotational Cartesian mechanism to operate at planes with all possible inclinations, and a task‐specific mechanism for ECT probe delivery. Machine vision locates tube bundle and tube inlets by an April tag detection algorithm and a Circle Hough Transform algorithm, respectively. Assisted by a guiding cone, the ECT probe is continuously fed into the tubes with a fill factor of 0.819. During this process, the eddy current data are automatically collected and real‐time analyzed by convolutional neural networks, showing accuracy of nearly 100% for identifying defective and nondefective tubes and 85% for four types of defective tubes and nondefective tubes.

show abstract

Section: Related Workmentioning

confidence: 99%

An autonomous robot for shell and tube heat exchanger inspection

Zheng

Xie

et al. 2022

Journal of Field Robotics

View full text Add to dashboard Cite

show abstract

“…Such an approach is supposed to be widely exploited by the NDT&E community in the near future along with the possibility to embed the explainability of a deep learning method. In [67], the authors studied the performance of a deep neural network for defect classification based on the use of two different ECT probes (i.e., pancake coil and +Point probe) signals. The analysis performed showed that the neural network schema adopted was able to classify, with good performance, the longitudinal, circumferential and no-defect classes.…”

Section: Deep Learning In Electromagnetic Ndtande Applied To the Ener...mentioning

confidence: 99%

Deep learning techniques for non-destructive testing and evaluation

Miorelli¹,

Skarlatos²,

Vienne³

et al. 2022

Applications of Deep Learning in Electromagnetics: Teaching Maxwell's Equations to Machines

View full text Add to dashboard Cite

“…Physical models describing ECs can be quite complex, and estimations of flaw parameters like length and depth is not always possible [8]. As alternative, many approaches rely on Machine Learning (ML) methods, employing Artificial Neural Networks (ANNs) [9][10][11][12][13][14][15][16][17][18][19], as they can perform any function if enough training data are provided [20][21][22], without requiring a physical model. Moreover, due to the flexibility of these algorithms, ANNs represent an interesting candidate for future integration with UT data.…”

Section: Introductionmentioning

confidence: 99%

Data Augmentation and Artificial Neural Networks for Eddy Currents Testing

Cormerais

Longo

Duclos

et al. 2020

Studies in Applied Electromagnetics and Mechanics

View full text Add to dashboard Cite

Eddy Currents (ECs) Non Destructive Testing (NDT) is widely used to determine the position and size of flaws in metal materials. Due to difficulties in estimating these parameters via inverse algorithms based on physical models, approaches focused on Artificial Neural Network (ANN) are nowadays of great interest. The main drawbacks of these techniques still reside in the complexity of the numerical models and the large number of simulated data needed to train and test the ANN, leading to a considerable amount of calculation time and resources. To overcome these limitations, this article proposes a new approach based on a data augmentation procedure via Principal Component Analysis (PCA) applied to numerical simulations.

show abstract

MRPC eddy current flaw classification in tubes using deep neural networks

Cited by 14 publications

References 4 publications

An autonomous robot for shell and tube heat exchanger inspection

An autonomous robot for shell and tube heat exchanger inspection

Deep learning techniques for non-destructive testing and evaluation

Data Augmentation and Artificial Neural Networks for Eddy Currents Testing

Contact Info

Product

Resources

About