Open Collections

Fast to run model for thermal fields during metal additive manufacturing simulations

University of British Columbia

Additive Manufacturing simulations for thermal fields are computationally expensive because of the highly disparate length and time scales involved and can sometimes take days to run. Improving the speed of these simulations enables multiple virtual experiments to be run to understand the effects of various process parameters on heat buildup and can even be useful for in situ process control based on sensor measurements from the build area. The goal of this work is to reduce the computational time of such simulations while maintaining sufficient physics fidelity to yield reliable results. The approach taken is to replace the FEM model with a Fast-to-run (FTR) model which exploits the cyclic nature of the process to predict the thermal fields during AM. In this approach, peak temperatures and melt pools dimensions in a substrate melted by a moving heat source are modelled. The dependence of the heat transfer patterns on the heat source location and characteristics and the initial conditions of the substrate is modelled using data from the FEM simulation. Simulation time using the FTR model has been reduced significantly compared to the FEM simulation based on the domain size and time simulated. Finally, the FTR model is run on various complex scenarios. The effects of various hatching strategies are modelled and their maximum temperatures and melt depths are compared. Additionally, a slice of an impeller model is simulated using the FTR model to generate maximum temperature and melt depth maps, allowing the identification of hotspots and undermelted regions.

Text

2020-09-10

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/75970

Materials Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/