Additively manufactured (AM) materials are prone to porosity, which limits their widespread adoption in fatigue-limited applications. Experimental campaigns are vital in understanding the effects of porosity on fatigue resistance but are costly and time consuming. This work presents a modeling framework to predict fatigue life debits in AM materials due to representative pore defects including keyhole, lack of fusion, and linearly aligned/planar pores. Materials characterization data is leveraged from specimens of alloy 718 (also known as IN718) that have been intentionally seeded with pore defect structures via a systematic modification of the process parameters. Statistically equivalent microstructure (SEM) models are generated as analogues to the experimental specimens and undergo crystal plasticity simulations. A Bayesian inference framework is used to calibrate a critical value of a damage parameter, which is used to predict the fatigue lives of SEMs seeded with pores. A strategy based on the macro-scale treatment of notches is proposed to regularize micromechanical field variables in large stress gradients in the presence of pores, or more generally, geometric discontinuities, of various sizes and morphology. The regularization volume is inversely proportional to the local pore radius of curvature and adequately smooths micromechanical fields. The framework presented here can be used to augment experimental data and rapidly evaluate the detrimental effects of porosity on fatigue resistance in AM components.
Read full abstract