Assessment of Geometrical Features of Internal Flaws with Artificial Neural Network

In nondestructive testing (NDT), geometrical features of a flaw embedded in the material such as its location, length, and orientation are critical factors to assess the severity of the flaw and make post-manufacturing decisions to improve the design. In this study, artificial intelligence (AI) based NDT approach was applied to the ultrasonic oscillograms obtained from virtual ultrasonic NDT to estimate geometrical features of a flaw. First, a numerical model of NDT specimen was constructed using acoustic finite element analysis (FEA) to produce the ultrasonic signals. The model was validated by comparing the simulated signals produced from the numerical model with the experimental data from actual NDT tests. Then, 750 numerical models containing flaws with different locations, lengths, and orientation angles were generated by FEA. Next, the oscillograms produced by the models were divided into 3 datasets: 525 for training, 113 for validation, and 112 for testing. Training inputs of the network were parameters extracted from ultrasonic signals by fitting them to sine functions. Lastly, to evaluate the network performance, outputs of the network including flaw’s location, length, and angle were compared with the desired values for all datasets. Deviations of the outputs from desired values were calculated by a regression analysis. Statistical analysis was also performed by measuring root mean square error (RMSE) and efficiency. RMSE in x-location, y-location, length, and angle estimations are 0.09 mm, 0.19 mm, 0.46 mm, and 0.75°, with efficiencies of 0.9229, 0.9466, 0.9140, and 0.9154, respectively for the testing dataset. Results suggest that the proposed AI-based method has the potential to interpret the oscillograms from ultrasonic NDT to estimate geometrical features of flaws embedded in the material.

[1]  S. Olutunde Oyadiji,et al.  Finite element analysis and experimental measurement of acoustic wave propagation for leakage detection in an air-filled pipe , 2017 .

[2]  Sung-Jin Song,et al.  Investigation of deep neural network with drop out for ultrasonic flaw classification in weldments , 2018 .

[3]  Tom Page,et al.  Automatic detection of defects in ultrasonic testing using artificial neural network , 2010 .

[4]  Xueliang Zhang,et al.  Numerical simulation of ultrasonic testing reliability of civil aircraft considering the influence of the angle between the sound beam axis and the crack orientation , 2020, IOP Conference Series: Materials Science and Engineering.

[5]  P. C. M. A. Farias,et al.  Flexible decision support system for ultrasound evaluation of fiber–metal laminates implemented in a DSP , 2016 .

[6]  Thanh-Phong Dao,et al.  An efficient hybrid approach of improved adaptive neural fuzzy inference system and teaching learning-based optimization for design optimization of a jet pump-based thermoacoustic-Stirling heat engine , 2019, Neural Computing and Applications.

[7]  Xia Liu,et al.  Detection of longitudinal cracks with a serrated columnar phased array transducer: A simulation study , 2015, 2015 International Conference on Control, Automation and Robotics.

[8]  Thanh-Phong Dao,et al.  A multi-response optimal design of bistable compliant mechanism using efficient approach of desirability, fuzzy logic, ANFIS and LAPO algorithm , 2020, Appl. Soft Comput..

[9]  P. Ufimtsev.  Fundamentals of the Physical Theory of Diffraction , 2007 .

[10]  Romesh Nagarajah,et al.  Feature selection for neural network based defect classification of ceramic components using high frequency ultrasound. , 2015, Ultrasonics.

[11]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[12]  S Chatillon,et al.  A system model for ultrasonic NDT based on the Physical Theory of Diffraction (PTD). , 2016, Ultrasonics.

[13]  Sheng-Wei Chi,et al.  Wavelet based harmonics decomposition of ultrasonic signal in assessment of plastic strain in aluminum , 2017 .

[14]  Wen Tong Chong,et al.  A Comparative Study of Activation Functions of NAR and NARX Neural Network for Long-Term Wind Speed Forecasting in Malaysia , 2019, Mathematical Problems in Engineering.

[15]  Maria C. S. Albuquerque,et al.  Decision support system for ultrasound inspection of fiber metal laminates using statistical signal processing and neural networks. , 2013, Ultrasonics.

[16]  Jan Drewes Achenbach,et al.  Quantitative nondestructive evaluation , 2000 .

[17]  Nobuyuki Toyama,et al.  Computerized Ultrasonic Imaging Inspection: From Shallow to Deep Learning , 2018, Sensors.

[18]  Sylvain Chatillon,et al.  Recent advances in semi-analytical scattering models for NDT simulation , 2011 .

[19]  Charles Hellier,et al.  Handbook of Nondestructive Evaluation , 2001 .

[20]  Tomasz Rymarczyk,et al.  A Non-Destructive System Based on Electrical Tomography and Machine Learning to Analyze the Moisture of Buildings , 2018, Sensors.

[21]  Chunguang Xu,et al.  Study on the Thick-Walled Pipe Ultrasonic Signal Enhancement of Modified S-Transform and Singular Value Decomposition , 2015 .

[22]  F C Cruz,et al.  Efficient feature selection for neural network based detection of flaws in steel welded joints using ultrasound testing , 2017, Ultrasonics.

[23]  Satish S. Udpa,et al.  Feature extraction techniques for ultrasonic signal classification , 2002 .

[24]  Min Meng,et al.  Ultrasonic signal classification and imaging system for composite materials via deep convolutional neural networks , 2017, Neurocomputing.

[25]  Neb I. Uzelac,et al.  How Do Defect Assessment Methods Influence the Choice and Construction of In-Line Inspection Tools , 2002 .

[26]  Laurence J. Jacobs,et al.  Review of Second Harmonic Generation Measurement Techniques for Material State Determination in Metals , 2015 .