A statistical volume element based approach to multiscale modeling of fatigue crack formation in aa 2024-t351

The objective of this paper is to develop an energy- and slip-based multiscale damage criterion to study the formation of fatigue cracks in crystalline metallic aerospace structural components. The formation of the initial crack can be decomposed into two stages: nucleation of micro cracks and coalescence of micro cracks into a major crack. In the first stage, a crack extends from within intermetallic particles into a surrounding grain of the alloy. Fatigue damage increments in four dependent slip planes are calculated and then measured to nucleate micro cracks. In the second stage, the micro cracks are seen to grow and coalesce, leading to the formation of a major crack. A novel meso-statistical volume element model is constructed to implement the simulation and improve computational efficiency compared to traditional representative volume element models. Fatigue tests of lug joint samples are performed to validate this multiscale damage criterion. The crack growing rate and direction show a good correlation between experimental data and simulation results.