Welding Defect Classification from Simulated Ultrasonic Signals

  • Raphaella H. F. Murta
  • Flávison de A. Vieira
  • Victor O. Santos
  • Elineudo P. de MouraEmail author


Nondestructive testing is widely used to detect and to size up discontinuities embedded in a material. Among the several ultrasonic techniques, time of flight diffraction (TOFD) combines high speed inspection, high sizing reliability and low rate of incorrect results. However, the classification of defects through ultrasound signals acquired by the TOFD technique depends heavily on the knowledge and experience of the operator and thus, this classification is still frequently questioned. Besides, this task requires long processing time due to the large amount of data to be analyzed. Nevertheless, computational tools for pattern recognition can be employed to analyze a high amount of data with large efficiency. In the present work, simulation of ultrasound propagation in two-dimensional media containing, each one, different kinds of modeled discontinuities which mimic defects in welded joints were performed. Clustering (k-means) and classification (principal component analysis and k-nearest neighbors) algorithms were employed to associate each simulated ultrasound signal with its corresponding modeled defects. The results for each method were analyzed, discussed and compared. The results are very promising.


Ultrasound TOFD Welding defects K-NN Principal component analysis K-means 


  1. 1.
    Kaufman, Leonard, Rousseeuw, Peter J.: Finding Groups in Data: An Introduction to Cluster Analysis, vol. 344. Wiley, New York (2009)zbMATHGoogle Scholar
  2. 2.
    Haykin, S.S.: Neural Networks and Learning Machines, vol. 3. Pearson, Upper Saddle River, NJ (2009)Google Scholar
  3. 3.
    Webb, Andrew R.: Statistical Pattern Recognition. Wiley, New York (2003)zbMATHGoogle Scholar
  4. 4.
    Silk, M.G., Lidington, B.H.: Defect sizing using an ultrasonic time-delay approach. Br. J. Non-Destr. Test. 17(2), 33–36 (1975)Google Scholar
  5. 5.
    Wooh, Shi-Chang, Shi, Yijun: A simulation study of the beam steering characteristics for linear phased arrays. Non-Destr. Eval. 18(2), 39–57 (1999)CrossRefGoogle Scholar
  6. 6.
    Ogilvy, J.A., Temple, J.A.G.: Diffraction of elastic waves by cracks: application to time-of-flight inspection. Ultrasonics 21(6), 259–269 (1983)CrossRefGoogle Scholar
  7. 7.
    Temple, J.A.G.: Predicted ultrasonic responses for pulse-echo inspections. Br. J. Non-Destr. Test. 28(3), 145–154 (1986)Google Scholar
  8. 8.
    Baskaran, G., Lakshmana Rao, C., Balasubramaniam, K.: Simulation of the tofd technique using the finite element method. Insight-Non-Destr. Test. Cond. Monit. 49(11), 641–646 (2007)CrossRefGoogle Scholar
  9. 9.
    Ghose, B., Balasubramaniam, K., Krishnamurthy, C.V.,Rao, A.S.: Two dimensional fem simulation of ultrasonic wave propagation in isotropic solid media using comsol. In COMSOL Conference (2010)Google Scholar
  10. 10.
    Orfanidis, S.J.: Introduction to Signal Processing. Prentice-Hall, Inc., Upper Saddle River (1995)Google Scholar
  11. 11.
    de Moura, E.P., Siqueira, M.H.S., da Silva, R.R., Rebello, J.M.A.: Welding defect pattern recognition in tofd signals part 2. Non-linear classifiers. Insight-Non-Destr. Test. Cond. Monit. 47(12), 783–787 (2005)CrossRefGoogle Scholar
  12. 12.
    Jolliffe, Ian: Principal Component Analysis. Wiley Online Library, New York (2002)zbMATHGoogle Scholar
  13. 13.
    Varella, C.A.A.: Análise de componentes principais. Universidade Federal Rural do Rio de Janeiro, Rio de Janeiro (2008)Google Scholar
  14. 14.
    de Moura, E.P., Vieira, A.P., Gonçalves, L.L.: Fluctuation analyses for pattern classification in nondestructive materials inspection. EURASIP J. Adv. Signal Process. 1, 262869 (2010)Google Scholar
  15. 15.
    Rousseeuw, P.J.: Rousseeuw. Silhouettes: a graphical aid to the interpretation and validation of cluster analysis. J. Comput. Appl. Math. 20, 53–65 (1987)CrossRefGoogle Scholar
  16. 16.
    Davies, D.L., Bouldin, D.W.: A cluster separation measure. IEEE Trans. Pattern Anal. Mach. Intell. 2, 224–227 (1979)CrossRefGoogle Scholar
  17. 17.
    Anil, K.: Jain. Data clustering: 50 years beyond k-means. Pattern recognition letters 31(8), 651–666 (2010)CrossRefGoogle Scholar
  18. 18.
    Petrovic, S.: A comparison between the silhouette index and the davies-bouldin index in labelling ids clusters. In Proceedings of the 11th Nordic Workshop of Secure IT Systems, pp 53–64 (2006)Google Scholar
  19. 19.
    Dimitriadou, E., Dolničar, S., Weingessel, A.: An examination of indexes for determining the number of clusters in binary data sets. Psychometrika 67(1), 137–159 (2002)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Departament of Metallurgical and Materials EngineeringFederal University of CearáFortalezaBrazil

Personalised recommendations