Domain adapted deep-learning for improved ultrasonic crack characterization using limited experimental data

Pyle, Richard J. and Bevan, Rhodri L. T. and Hughes, Robert R. and Ali, Amine Ait Si and Wilcox, Paul D. (2022) Domain adapted deep-learning for improved ultrasonic crack characterization using limited experimental data. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 69 (4). 1485 - 1496. (https://doi.org/10.1109/TUFFC.2022.3151397)

[thumbnail of Pyle-etal-IEEE-TUFFC-2022-Domain-adapted-deep-learning-for-improved]
Preview
Text. Filename: Pyle_etal_IEEE_TUFFC_2022_Domain_adapted_deep_learning_for_improved.pdf
Accepted Author Manuscript
License: Strathprints license 1.0

Download (3MB)| Preview

Abstract

Deep learning is an effective method for ultrasonic crack characterization due to its high level of automation and accuracy. Simulating the training set has been shown to be an effective method of circumventing the lack of experimental data common to nondestructive evaluation (NDE) applications. However, a simulation can neither be completely accurate nor capture all variability present in the real inspection. This means that the experimental and simulated data will be from different (but related) distributions, leading to inaccuracy when a deep learning algorithm trained on simulated data is applied to experimental measurements. This article aims to tackle this problem through the use of domain adaptation (DA). A convolutional neural network (CNN) is used to predict the depth of surface-breaking defects, with in-line pipe inspection as the targeted application. Three DA methods across varying sizes of experimental training data are compared to two non-DA methods as a baseline. The performance of the methods tested is evaluated by sizing 15 experimental notches of length (1–5 mm) and inclined at angles of up to 20° from the vertical. Experimental training sets are formed with between 1 and 15 notches. Of the DA methods investigated, an adversarial approach is found to be the most effective way to use the limited experimental training data. With this method, and only three notches, the resulting network gives a root-mean-square error (RMSE) in sizing of 0.5 ± 0.037 mm, whereas with only experimental data the RMSE is 1.5 ± 0.13 mm and with only simulated data it is 0.64 ± 0.044 mm.

ORCID iDs

Pyle, Richard J. ORCID logoORCID: https://orcid.org/0000-0002-5236-7467, Bevan, Rhodri L. T., Hughes, Robert R., Ali, Amine Ait Si and Wilcox, Paul D.;