Numerical simulation of part-level temperature fields during selective laser melting of stainless steel 316L

Numerical transient thermal analysis of a part during the selective laser melting process is a great challenge due to huge computational cost. An efficient simulation scheme is proposed to estimate the temperature distribution and history in part level. The building domain is coarsely meshed at the beginning. Then, deposited elements are activated layer by layer with desired material properties. The Gaussian distributed heat flux is applied to the top surface and travels according to the predefined scanning path. With the motion of input heat flux, the mesh is continuously refined and coarsened as the laser beam moves in and out. To improve the solution accuracy, temperature-dependent thermal properties in powder and solid state are incorporated in the finite element system. To reduce the computation cost, the Gaussian line heat source with a large time step length is used to replace the traditional moving Gaussian point heat source. To achieve a compromise between the computational efficiency and the solution accuracy, the hybrid of the Gaussian line heat source and Gaussian point heat source is adopted as heat input and results are compared with that of the Gaussian point heat source. The proposed simulation scheme is validated by comparing the simulated geometry of the melt pool with experimental results. Simulation results show that the proposed simulation scheme has the ability to efficiently and accurately predict the temperature field of a part during the selective laser melting process and can be easily extended to other powder bed fusion processes.