Issue 53

P. Ferro et alii, Frattura ed Integrità Strutturale, 53 (2020) 252-284; DOI: 10.3221/IGF-ESIS.53.21

a

c

b

Figure 25: Inherent strain definition: a) without residual stress, b) with residual stress; c) removal of residual stresses by cutting the part. ε * is the inherent strain.

Figure 26: Comparison between experimental and numerical distortion of a cantilever specimen obtained by using the inherent strain method [158]. The second kind of approach used to solve the full-scale model is the thermo-mechanical simulation of the entire part but using a layer agglomeration strategy to speed up the calculation. Each macro layer (or block), when activated is either heated at one time [158,159] or scanned by a heat source with fictious parameters in order to reach the full penetration of the block itself [160,161]. A ‘multiscale approach’ is also adopted in literature [147,162] where the equivalent body heat flux is first derived from a micro-scale laser-scanning model and input to a meso-scale model for the simulation of the in-plane stress and finally into the full-scale model to obtained the final residual stress state. Li et al. [163] looked at the effect of macro layer thickness on the residual stress and distortion. They found that a scale-up of the layer thickness from 30 μ m to 1.5 mm dramatically reduces the computational cost by more than 3 times compared to a 0.75 mm macro layer thickness. They confirmed that orthogonal scanning pattern between two adjacent block layers is beneficial for reducing stress and distortion. In general, thermo-mechanical computation of the whole part, even when employing an agglomeration approach, is more time consuming compared to the inherent strain method but even more reliable since the effects of part geometry on the thermal history are at least partially taken into account.

276