Prediction on the fatigue life of butt-welded specimens using artificial neural network

  • Kim, Kyoung Nam (School of Civil Engineering, Chungbuk National University) ;
  • Lee, Seong Haeng (Department of Civil Engineering, Pusan National University) ;
  • Jung, Kyoung Sup (School of Civil Engineering, Chungbuk National University)
  • Received : 2009.11.22
  • Accepted : 2009.11.17
  • Published : 2009.11.25


Fatigue tests for extremely thick plates require a great deal of manufacturing time and are expensive to perform. Therefore, if predictions could be made through simulation models such as an artificial neural network (ANN), manufacturing time and costs could be greatly reduced. In order to verify the effects of fatigue strength depending on the various factors in SM520C-TMC steels, this study constructed an ANN and conducted the learning process using the parameters of calculated stress concentration factor, thickness and input heat energy, etc. The results showed that the ANN could be applied to the prediction of fatigue life.


prediction;fatigue life;fatigue strength;artificial neural network


  1. AASHTO LRFD Bridge Design Specification (2002), U.S. Units 2002, Interim Revisions.
  2. Al-Jabri, K.S. and Al-Alawi, S.M. (2007), "Predicting the behavior of semi-rigid joints in fire using an artificial neural networks", Int. J. Steel Struct., 7(3), 209-217.
  3. Amanullah, M., Siddiqui, N.A., Umar, A. and Abbas, H. (2002), "Fatigue reliability analysis of welded joints of a TLP tether system", Steel Compos. Struct., 2(5), 331-354.
  4. Ayala-Uraga, E. and Moan, T. (2007), "Fatigue reliability assessment of welded joints applying consistent fracture mechanics formulations", Int. J. fatigue, 29, 444-456.
  5. Bezazi, A., Gareth Pierce, S., Worden, K. and Harkati, E.H. (2007), "Fatigue life prediction of sandwith composite materials under flexural tests using a Bayesian trained artificial neural network", Int. J. fatigue, 29, 738-747.
  6. Do, Y.T., Kim, I.G., Kim, J.W. and Park, C.H. (2001), Artificial intelligence-Concept and Application, Scitech Media, Korea.
  7. Fathi, A. and Aghakouchak, A.A. (2007), "Prediction of fatigue crack growth rate in welded tubular joints using neural network", Int. J. fatigue, 29, 261-275.
  8. Kang, J.Y., Choi, B.I., Lee, H.J., Kim, J.S. and Kim, K.J. (2006), "Neural network application in fatigue damage analysis under multiaxial random loadings", Int. J. fatigue, 28, 132-140.
  9. Kim, J.T., Park, J.H., Koo, K.Y. and Lee, J.J (2008), "Acceleration-based neural networks algorithm for damage detection in structures", Smart Struct. Syst., 4(5), 583-603.
  10. Kim, K.N., Lee, S.H. and Jung, K.S. (2009), "Evaluation of factors affecting the fatigue behavior of buttwelded joints using SM520C-TMC Steel", Int. J. Steel Struct., 9(3), 185-193.
  11. Lee, Y.L., Tjhung, T. and Jordan, A. (2007), "A life prediction model for welded joints under multiaxial variable amplitude loading histories", Int. J. fatigue, 29, 1162-1173.
  12. Li, B., Reis, L. and Freitas, M.D. (2006), "Simulation of cyclic stress/strain evolutions for multiaxial fatigue life prediction", Int. J. fatigue, 28, 451-458.
  13. Majidian, A. and Saidi, M.H. (2007), "Comparison of Fuzzy logic and Neural Network in life prediction of boiler tubes", Int. J. fatigue, 29, 489-498.
  14. McCulloch, W.S. and Pitts, W. (1943), "A Logical Calculus of the Ideas Imminent in Nervous Activity", B. Math. Bio., 5, 115-133.
  15. Nishida, M. (1971), Stress Concentration, Morikita Ink, Japan.
  16. Rumelhart, D.E., Hinton, G.E. and Williams, R.J. (1986), Learning Internal Representation by Error Propagation, Parallel Distributed Processing, Vol. 1, MIT Press.
  17. Yi, J.H., Yun, C.B. and Feng, M.Q. (2003), "Model updating and joint damage assessment for steel frame structures using structural identification techniques", Int. J. Steel Struct., 3(2), 83-94.

Cited by

  1. Prediction of the flexural overstrength factor for steel beams using artificial neural network vol.17, pp.3, 2014,