Toward a Simulation-Based Qualification Paradigm for Metal Additive Manufacturing: Support Design for Residual Stress and Process-Grain Structure Modeling: SEMM Seminar
Lecture: Structural Engineering, Mechanics and Materials: Civil and Environmental Engineering | September 23 | 12-1 p.m. | 502 Davis Hall
Albert C. To, Ph.D. Professor, University of Pittsburgh
This seminar will present recent advances in models and methods developed for achieving a simulation-based qualification paradigm for laser-based metal additive manufacturing (AM). First, we will present a support structure design optimization method proposed to reduce residual stress and distortion in an AM build. The key novelty of this method lies in the formulation of the modified inherent strain model which enables fast and accurate prediction of part-scale residual stress and deformation resulting from laser processing. The fast model makes it practical to use topology optimization for AM support structure design. Through both simulation and experimentation, the support structure optimization method developed is shown to not only prevent AM build failure by reducing residual stress significantly, but also reduce the amount of material needed for the support. Next, we will present an efficient model proposed to simulate the competitive growth of epitaxial columnar dendritic grains by geometrically tracking the tip of the constituent stable primary dendrite arms. The grain textures obtained from simulation and experiment for different laser scan strategies are in excellent agreement.