Quantifying Uncertainty in the Mechanical Performance of Additively Manufactured Parts Due to Material and Process Variation

Additive manufacturing (AM) technology offers rapid prototyping and large design freedom. It can potentially reduce costs and improve quality by providing precise control on microstructure and density for the production of complex components. However, the part quality and mechanical performance of components fabricated by current AM technology are not comparable to that produced by traditional methods. Furthermore, the final quality of AM product varies depending on variation between and within machines, raw material, laser beam characteristics, builds parameters, and thermal processing. Current practices, which often place more emphasis on post hoc qualitative approaches, are not able to quantify the performance uncertainty due to multiple input variables. Therefore, a quantitative computational approach for uncertainty analysis is proposed here to deal with uncertainties and provide higher confidence in performance properties of parts fabricated by SLM process. In particular, the potential impact of all variables on the performance characteristics will be addressed by our simulation/surrogate modeling approach. In Phase I, we have successfully demonstrated the use of simulation and surrogate models to provide fast uncertainty quantification of thermal history, porosity and microstructure evolution. The framework will be extended and validated at coupon- and part-level with in-situ monitoring data and experiments.