Artificial Neural Network Based Prediction Techniques for Torch Current Deviation to Produce Defect-Free Welds in GTAW Using IR Thermography

  • N. M. Nandhitha
Conference paper
Part of the Smart Innovation, Systems and Technologies book series (SIST, volume 43)


In recent years, on-line weld monitoring is the potential area of research. In this work, torch current deviation prediction systems are developed with Artificial Neural Networks to produce welds free from Lack of Penetration. Lack of penetration is deliberately introduced by varying the torch current. Thermographs are acquired during welding and hotspots are extracted using Euclidean Distance based segmentation and are quantitatively characterized using the second order central moments. Exemplars are then created with central moments as input parameters and deviation in torch current as the output parameter. Radial Basis Networks (RBN) and Generalized Regressive Neural Networks (GRNN) are then trained and tested to assess the suitability for torch current prediction. GRNN outperforms RBN in predicting the torch current deviation with 98.95 % accuracy.


GTAW Lack of penetration RBN GRNN Torch current deviation 


  1. 1.
    Sreedhar, U., Krishnamurthy, C.V., Balasubramaniam, K., Raghupathy V.D., Ravisankar S.: Automatic defect identification using thermal image analysis for online weld quality monitoring. J. Mater. Process. Tech. 212(7), 1557–1566 (2012)Google Scholar
  2. 2.
    Vasudevan, M., Chandrasekhar, N., Maduraimuthu, V., Bhaduri, A.K., Raj, B.: Real-time monitoring of weld pool during GTAW using infra-red thermography and analysis of infra-red thermal images. Weld. World 55(7–8), 83–89 (2012)Google Scholar
  3. 3.
    Leksir, Y.L.D., Bouhouche, S., Boucherit, M.S., Bast, J.: Submerged arc welding online quality evaluation using infrared thermography based fuzzy reasoning. In: 13th International Symposium on Nondestructive Characterization of Materials (2013)Google Scholar
  4. 4.
    Swiderski, W., Hlosta, P.: Pulsed eddy current thermography for defects detection in joints of metal sheets. In: 11th European Conference on Non-Destructive Testing (2014)Google Scholar
  5. 5.
    De La Yedra, A.G., Echeverria, A., Beizama, A., Fuente, R., Fernández, E.: Infrared thermography as an alternative to traditional weld inspection methods thanks to signal processing techniques. In: 11th European Conference on Non-Destructive Testing (2014)Google Scholar
  6. 6.
    Lancaster, J.: Handbook of Structural Welding, Processes, Materials and methods used in the Welding of Major Structures, pipelines and process plants, vol. 260. Abington Publishing, (1997)Google Scholar
  7. 7.
    Halmshaw, R.: Industrial Radiology, Theory and Practice, vol. 230. Chapman & Hall Publications, London (1995)Google Scholar
  8. 8.
    Gonzalez, R.C., Woods, R.E.: Digital Image Processing. Prentice Hall, New Delhi (2005)Google Scholar
  9. 9.
    Selvarasu, N., Nachiappan, A., Nandhitha, N.M.: Abnormality detection from medical thermographs in human using Euclidean distance based color image segmentation. In: Proceedings of 2010 International Conference on Signal Acquisition and Processing, pp. 73–75, (2010)Google Scholar
  10. 10.
    Freeman, J.A., Skapura, D.M.: Neural Networks Algorithms. Applications and Programming Techniques, Pearson Education (1997)Google Scholar

Copyright information

© Springer India 2016

Authors and Affiliations

  1. 1.Deparment of Electronics and Communications EngineeringSathyabama UniversityChennaiIndia

Personalised recommendations