Computational Modeling of Laser Additive Manufacturing Processes

The overall goal of the proposed research is to develop a comprehensive, efficient, and well-validated model for the prediction of the shape and thermal history of components manufactured using Laser Additive Manufacturing (LAM) processes. The proposedmodel will use a combined Eulerian-Lagrangian treatment for analyzing the interactions between the gas and particle phases, and the laser beam. The temperature field in the deposition region will be determined by solving the unsteady energy transportequation that accounts for phase change and free surface effects. Detailed information of temperature gradients will be used for predicting the microstructure and residual stresses. Emphasis is placed on computational efficiency. Thus, the Volume-of-Fluidtechnique will be used to predict the shape evolution in a parent grid. A two-domain approach involving local mesh refinement and multigrid solution will be used for accurate analysis of the thermal interaction between the small-scale deposition region andthe large-scale bulk. The model will be validated using data from experiments on practical LAM units. In order to establish the feasibility of the technical approach, Phase I research will focus on the development, validation, and application of the modelsfor a co-axial deposition nozzle and for LAM processing of a thin-walled structure. The proposed model will offer a scientific approach to process design and optimization of LAM processes involving varying processing scales. The cost savings obtainedthrough improvements in powder utilization, turnaround time, and product quality will facilitate wide-scale commercialization of the LAM processes. A conservative estimate of the annual revenue from the commercial version of the model is $500,000/yr.