A Study on Fatigue Damage Modeling Using Neural Networks

  • Lee Dong-Woo (Department of Mechanical Engineering, Dong-A University) ;
  • Hong Soon-Hyeok (Cooperative Laboratory Center, Pukyong National University) ;
  • Cho Seok-Swoo (Department of Vehicle Engineering, Samcheok National University) ;
  • Joo Won-Sik (Department of Mechanical Engineering, Dong- A University)
  • Published : 2005.07.01

Abstract

Fatigue crack growth and life have been estimated based on established empirical equations. In this paper, an alternative method using artificial neural network (ANN) -based model developed to predict fatigue damages simultaneously. To learn and generalize the ANN, fatigue crack growth rate and life data were built up using in-plane bending fatigue test results. Single fracture mechanical parameter or nondestructive parameter can't predict fatigue damage accurately but multiple fracture mechanical parameters or nondestructive parameters can. Existing fatigue damage modeling used this merit but limited real-time damage monitoring. Therefore, this study shows fatigue damage model using backpropagation neural networks on the basis of X -ray half breadth ratio B / $B_o$, fractal dimension $D_f$ and fracture mechanical parameters can estimate fatigue crack growth rate da/ dN and cycle ratio N / $N_f$ at the same time within engineering limit error ($5\%$).

Keywords

References

  1. Coffin, L. F., Jr., 1954, 'A Study of the effects of Cyclic Thermal Stresses in a Ductile Metal,' ASME, Transactions, Vol. 16, pp. 931-950
  2. Elber, W., 1970, The significance of Fatigue Crack Closure, Damage Tolerance in Aircraft Structures, STP-486, ASTM, pp. 230-243
  3. Hironobu Nisitani and Masahir Goto, 1985, 'Relation between Small-crack Growth Law and Fatigue Life of Machines,' Journal of JSME, Vol. 51, No. 462, pp. 332-341
  4. Hiroshi OKUDA, Hiroshi MIYZAKI, Genki YAGAWA, 1996, 'Model of Inelastic Response using Neural Networks,' Int. J. of JSME (A), Vol. 62, No. 597, pp. 1284-1291
  5. Jang, D. Y., Cho, S. S. and Kim, D. J., 1999, A Study on Fractal Property of Surface Microcrack Under Plane Bending Load, Journal of Samcheok National University (Korea), Vol. 31, pp. 35-49
  6. Joo, W. S. and Cho, S. S., 1996a, 'A Study on High Temperature Low Cycle Fatigue Crack Growth Modeling by Neural Networks,' Journal of the Korean Society of Mechanical Engineering (A), Vol. 20, No. 9, pp. 2752-2759
  7. Joo, W. S., Oh, S. W., Cho, S. S. and Hue, C. W., 1994a, 'A Study on Fatigue Crack Growth Behavior at a Creep Temperature Region in SUS 304 Stainless Steel,' Journal of the Korean Society of Mechanical Engineering, Vol. 18, No. 3, pp. 548-554
  8. Joo, W. S., Park, S. Y., Kim, D. J. Cho, S. S. and Jang, D. Y., 1998a, 'A Study on Relationship of between X-ray Half-value Breadth and Cycle ratio in Al 2024-T3 Alloy Using Average gradient method,' 1998 Fall Conference Proceeding (II), KSPE, pp. 881-886
  9. Kim, D. S., 1992, Neural networks-Theory and Application, HiTech publisher, Seoul, pp. 97-144
  10. Kim, M. C., 1998, 'A Study on Integrated Fatigue Damage Modeling Using Back-propagation Neural Networks,' M.S. Thesis, Dong-A University (Korea), pp. 11-22
  11. Mandelbert, B. B., 1983, The Fractal Geometry of Nature, Freeman, Sanfrancisco, pp. 25-29
  12. Paris, P. C. and Erdogan, F., 1963, 'A Critical Analysis of Crack Propagation Laws,' Trans. ASME, Basic Eng., Vol. 85, p. 528 https://doi.org/10.1115/1.3656900
  13. Park, S. Y., Kim, D. J., Cho, S. S., Joo, W. S. and Hong, S. H. 1998, 'A Study on Estimation of Fatigue Life in SPCC Steel using X-ray Halfvalue Breadth,' 2003 Spring Conference Proceeding (II), KSAE, pp. 768-774
  14. Tanaka. K., Hishide, T. and Maekawa, O, 1982, 'Surface-crack Propagation in Plane Bending Fatigue of Smooth Specimen of Low-Carbon Steel,' Eng. Frac. Mech., Vol. 16, No. 2, p. 207 https://doi.org/10.1016/0013-7944(82)90150-3
  15. Wohler, A., 1860, Versuche uber die Festigkeit der Eisenbahnwagen-Achsen, Zeitchrift fur Bauwesen
  16. X-Wu, J. Ghabousi, 1993, 'Modeling The Cyclic Behavior of Concrete Using Adaptive Neural Network,' Computational Mechanics, Vol. 1, pp. 1319-1329