A TCD-based statistical method to assess the impact of surface roughness and pores on the fatigue strength of LPBF Inconel 718 specimens

The aim of this study is to model the impact of surface roughness and pores on the fatigue strength of plain and V-notched specimens made of Inconel 718 under as-built and machined conditions and produced by laser powder bed fusion (LPBF). Combining fractographic analyses with the Gumbel and the exponential distribution functions, the statistical analyses of the diameters of the pores and of their distances from the external surfaces were implemented. Surface roughness scans were performed with the optical profilometer. The finite element (FE) method was used to simulate a sample of pores generated by the identified probability distributions and the surface profiles obtained with the scans. The theory of critical distances (TCD) was implemented combining the blunt and sharp V-notched specimens in the machined condition, and it was combined with the Gumbel or the generalized extreme values distributions to calculate the fatigue strength concentration factors provided by the pores and the surface roughness at 99% of probability. Finally, the proposed model was used to predict the fatigue strength of the blunt V-notched specimens in the as-built conditions and of the plain specimens in the as-built and machined conditions resulting appreciably similar to the experimental data.