Defect Recognition of Resistance Spot Welding Based on Artificial Neural Network

Conference paper
Part of the Advances in Intelligent and Soft Computing book series (AINSC, volume 115)


The nugget size forecast model was built based on BP algorithm of Artificial Neural Network. The input parameters of the model are two characteristic number extract from electrode displacement curve. The output parameter of the model is nugget size. The model has three layers and the hidden layer have five nodes. The transfer function of hidden layer is Sigmoid function and the transfer function of output layer is linear function. Measured value and forecast value was analyzed by comparative method and the difference value between them was calculated. The results showed that 83% of the difference value is less than 1 millimeter. Based on the nugget size forecasted by the model, a method to identify incomplete fusion defect of resistance spot welding was suggested. In this method, 7 millimeter nugget size was regarded as criterion of recognition incomplete fusion defect of resistance spot welding. The result showed that recognition accuracy rate up to 94.3%.


Artificial neural network Nugget size forecast Defect recognition Incomplete fusion defect 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Institute of Welding, Chinese Mechanical Engineering Society. The Theory and Practice of Resistance Welding. Machinery Industry Press, Beijing (1994)Google Scholar
  2. 2.
    Luo, X., Ji, C., Zhang, C., et al.: Quality Control Visualization of Direct Current Spot Welder with Secondary Rectification. Transactions of The China Welding Institution 22(1), 59–61 (2001)Google Scholar
  3. 3.
    Hao, M., Osman, K.A., Boomer, D.R., et al.: Developments in Characterization of Resistance Spot Welding of Aluminum. Welding Journal 75(1), 1s–8s (1996)Google Scholar
  4. 4.
    Xue, H., Li, Y., Cui, C., et al.: Extraction of diagnostic information of expulsion defect in resistance spot welding process by wavelet analysis method. Transactions of The China Welding Institution 28(5), 38–40 (2007)Google Scholar
  5. 5.
    Niu, Y., Xue, H., Zeng, Z., et al.: Data acquisition and defects analysis in resistance spot welding process based on parallel port. Transactions of The China Welding Institution 28(6), 61–64 (2007)Google Scholar
  6. 6.
    Osman, K.A., Higginson, A.M., Kelly, H.R., et al.: Monitoring of resistance spot-welding using multi-layer perceptrons. Intelligent Engineering Systems Through Artificial Neural Networks 4, 1109–1114 (1994)Google Scholar
  7. 7.
    Cho, Y.J., Rhee, S.: Quality Estimation of Resistance Spot Welding by Using Pattern Recognition With Neural Networks. IEEE Transactions on Instrumentation and Measurement 53(2), 330–334 (2004)CrossRefGoogle Scholar
  8. 8.
    Lee, H.-T., Wang, M., Maev, R., et al.: A study on using scanning acoustic microscopy and neural network techniques to evaluate the quality of resistance spot welding. International Journal of Advanced Manufacturing Technology 22(9-10), 727–732 (2003)CrossRefGoogle Scholar
  9. 9.
    Chen, Y., Hu, D., Ma, L., et al.: Neural Network Model for DC Spot Welding of Aluminum-alloy. Transactions of The China Welding Institution 21(4), 20–23 (2000)Google Scholar
  10. 10.
    Fang, P., Tan, Y., Wu, L., et al.: Artificial Neural Networks Applied to the Quality Control in Alternating Current Resistance Spot Welding. Acta Aeronautica Et Astronautica Sinica 21(1), 94–95, 86 (2000)Google Scholar
  11. 11.
    Han, L.: Artificial neural network theory, design and application. Chemical Industry Press, Beijing (2002)Google Scholar

Copyright information

© Springer-Verlag GmbH Berlin Heidelberg 2012

Authors and Affiliations

  • Li Yongyan
    • 1
  • Zhao Weimin
    • 1
  • Xue Haitao
    • 1
  • Ding Jian
    • 1
  1. 1.School of Material Science & EngineeringHebei University of TechnologyTianjinChina

Personalised recommendations