Search

Abstract This paper develops a Bayesian inference-based probabilistic crack nucleation model for the Ni-based superalloy René 88DT under fatigue loading. A data-driven, machine learning approach is developed, identifying underlying mechanisms driving crack nucleation. An experimental set of fatigue-loaded microstructures is characterized near crack nucleation sites using scanning electron microscopy and electron backscatter diffraction images for correlating the grain morphology and crystallography to the location of crack nucleation sites. A concurrent multiscale model, embedding experimental polycrystalline microstructural representative volume elements (RVEs) in a homogenized material, is developed for fatigue simulations. The RVE domain is modeled by a crystal plasticity finite element model. An anisotropic continuum plasticity model, obtained by homogenization of the crystal plasticity model, is used for the exterior domain. A Bayesian classification method is introduced to optimally select informative state variable predictors of crack nucleation. From this principal set of state variables, a simple scalar crack nucleation indicator is formulated.

In this work, we study the relationship between slip localizations and local orientation gradients in a polycrystalline material. We employ a 3D spatially resolved crystal plasticity-based micromechanics technique that includes explicit intragranular slip band modeling, called SB-FFT (slip band fast Fourier transform). EBSD analysis reveals the development of slip localizations, with a rare few generating relatively large local orientation gradients in the neighboring grain. Our results find that several conditions need to be met simultaneously for large zone of intense orientation gradients to form within a grain and with this, offer an explanation for their scarcity. These conditions entail the localization of slip to a sufficient high intensity in the grain neighbor, the grain experiencing higher stress levels than the average stress of its nearest grain neighborhood, and a crystallographic orientation suitable for activating secondary slip under the stress field produced by the slip localization. Analysis of the 3D orientation distributions as they evolve in strain indicates that orientation gradient zones cause the distribution to become highly skewed, with the extreme tails extending further as strain increases. The numerical simulations with various grain neighborhoods show that this feature is, however, not a sufficient signature for such orientation gradients associated with slip localizations since triple junctions or quadruple points can also produce intragranular orientation gradients that have a similar impact on the grain orientation distribution. Last, comparison with several previously proposed slip transmission criteria indicates that this phenomenon is not relevant, contrary to current belief.