Journal of Lanzhou University of Technology ›› 2024, Vol. 50 ›› Issue (5): 32-38.

• Mechanical Engineering and Power Engineering • Previous Articles     Next Articles

Multiaxial fatigue life modeling of notched parts based on Weibull distribution

LIU Jian-hui*, HE Ying-bao, WEI Yao-bing   

  1. School of Mechanical and Electronical Engineering, Lanzhou Univ. of Tech., Lanzhou 730050, China
  • Received:2022-04-06 Online:2024-10-28 Published:2024-10-31

Abstract: In order to investigate the influence of stress gradient on the fatigue life of notched parts and to assess the fatigue life of notched parts with different failure probabilities, a new fatigue life prediction method based on Weibull distribution is proposed. First, the maximum strain energy density is defined as the damage parameter, and the location of the critical plane reflecting the crack initiation and expansion is determined by combining the coordinate transformation principle. Secondly, the stress distribution of a specific path on the critical plane is extracted by using the finite element method, which is then normalized to obtain the multiaxial equivalent stress gradient factor. Meanwhile, based on the GSP strain energy density model, an appropriate strain energy density model is selected, and the linear relationship between the strain energy density and the experimental life can be obtained. Finally, a new fatigue life prediction method is proposed based on the Von-Mises criterion by combining the equivalent stress gradient factor, the strain energy density model, and the Weibull distribution, which can be used to find the fatigue life at 10%, 50%, and 90% failure probability respectively. In order to verify the accuracy of the proposed model, the calculation results of this paper are compared with those of the SWT model, the FS model, and the Manson-Coffin equation, demonstrating a high degree of precision in the predictions.

Key words: multiaxial notched parts, stress gradient effect, Weibull distribution, critical plane method

CLC Number: