Flaw detection in non-destructive testing, especially for complex signals like ultrasonic data, has thus far relied heavily on the expertise and judgement of trained human inspectors. While automated systems have been used for a long time, these have mostly been limited to using simple decision automation, such as signal amplitude threshold. The recent advances in various machine learning algorithms have solved many similarly difficult classification problems, that have previously been considered intractable. For non-destructive testing, encouraging results have already been reported in the open literature, but the use of machine learning is still very limited in NDT applications in the field. Key issue hindering their use, is the limited availability of representative flawed data-sets to be used for training. In the present paper, we develop modern, deep convolutional network to detect flaws from phased-array ultrasonic data. We make extensive use of data augmentation to enhance the initially limited raw data and to aid learning. The data augmentation utilizes virtual flaws—a technique, that has successfully been used in training human inspectors and is soon to be used in nuclear inspection qualification. The results from the machine learning classifier are compared to human performance. We show, that using sophisticated data augmentation, modern deep learning networks can be trained to achieve human-level performance.

No abstract

Modern ultrasonic inspections utilize ever-richer data-sets made possible by phased array equipment. A typical inspection may include tens of channels with different refraction angle, that are acquired at high speed. These rich data sets allow highly reliable and efficient inspection in complex cases, such as dissimilar metal or austenitic stainless steel welds. The rich data sets allow human inspectors to detect cracks with low signal-to-noise ratio from the wider signal patterns. There’s a clear trend in the industry to even richer data sets with full matrix capture (FMC) and related techniques. Convolutional neural networks have recently shown capability to detect flaws with human level accuracy in ultrasonic signals at the B-scan level. To enable automated flaw detection at human-level accuracy for critical applications, these neural networks need be developed to take advantage of today’s rich phased array data-sets. In the present paper, we extend previous work and develop convolutional neural networks that perform highly reliable flaw detection on typical multi-channel phased array data on austenitic welds. The results show, that the modern neural networks can accommodate the rich ultrasonic data and display high flaw detection performance.

Previous research (Li et al., Understanding the disharmony between dropout and batch normalization by variance shift. CoRR abs/1801.05134 (2018). http://arxiv.org/abs/1801.05134arXiv:1801.05134) has shown the plausibility of using a modern deep convolutional neural network to detect flaws from phased-array ultrasonic data. This brings the repeatability and effectiveness of automated systems to complex ultrasonic signal evaluation, previously done exclusively by human inspectors. The major breakthrough was to use virtual flaws to generate ample flaw data for the teaching of the algorithm. This enabled the use of raw ultrasonic scan data for detection and to leverage some of the approaches used in machine learning for image recognition. Unlike traditional image recognition, training data for ultrasonic inspection is scarce. While virtual flaws allow us to broaden the data considerably, original flaws with proper flaw-size distribution are still required. This is of course the same for training human inspectors. The training of human inspectors is usually done with easily manufacturable flaws such as side-drilled holes and EDM notches. While the difference between these easily manufactured artificial flaws and real flaws is obvious, human inspectors still manage to train with them and perform well in real inspection scenarios. In the present work, we use a modern, deep convolutional neural network to detect flaws from phased-array ultrasonic data and compare the results achieved from different training data obtained from various artificial flaws. The model demonstrated good generalization capability toward flaw sizes larger than the original training data, and the effect of the minimum flaw size in the data set affects the $$a_{90/95}$$ a 90 / 95 value. This work also demonstrates how different artificial flaws, solidification cracks, EDM notch and simple simulated flaws generalize differently.