导航

ACTA AERONAUTICAET ASTRONAUTICA SINICA ›› 2015, Vol. 36 ›› Issue (11): 3586-3594.doi: 10.7527/S1000-6893.2014.0339

• Solid Mechanics and Vehicle Conceptual Design • Previous Articles     Next Articles

Determination of welding residual stress intensity factors by weight function methods

JING Zhi, WU Xueren, TONG Dihua, CHEN Bo   

  1. AVIC Beijing Institute of Aeronautical Materials, Beijing 100095, China
  • Received:2014-10-31 Revised:2014-12-15 Online:2015-11-15 Published:2014-12-19
  • Supported by:

    National Natural Science Foundation of China (11402249)

Abstract:

Stress intensity factor due to complex welding residual stress is a prerequisite for damage tolerance analysis of welded structures. The weight function method is a powerful tool for calculating stress intensity factors under arbitrary loads. However, application of the classical weight function method has been hampered by the crack geometry complexity. In this paper, using a new weight function approach based on complex Taylor series expansion, the first partial derivative of crack opening displacement with respect to crack length is determined from complex finite element computation. Classical series expansion expressions are used to curve-fit the weight functions for three crack geometries:a periodic array of collinear cracks in an infinite sheet, a center crack in a finite width sheet and an edge crack in a finite width sheet. The stress intensity factors for the three crack geometries subjected to typical weld residual stresses are determined. The results are widely compared with the classic weight function method, the finite element method and the well-known results in the literature. It is demonstrated that the weight function complex Taylor series expansion method is highly efficient and accurate for analyzing crack problems under complicate load conditions including weld residual stresses.

Key words: weld residual stresses, weight functions, stress intensity factor, complex Taylor series expansion, damage tolerance

CLC Number: