Process modelling for selective laser melting additive manufacturing

Selective laser melting (SLM), a typical additive manufacturing technique, selectively fuses metallic powders to build products via a layer-wise method from computer-aided-design models. The SLM technique has shown great potential for aerospace, automobile and biomedicine industries. However, SLM-built products suffer from drawbacks such as poor surface finish, thermal distortion, cracks and unsatisfactory mechanical strength. In the SLM additive manufacturing process, there exist complex multi-scale and multi-physics phenomena including laser interaction with powders, heat and mass transfer, thermal distortion, multi-phase flow and microstructural evolution. The representative parameters of these phenomena are difficult to measure through experiments. Alternatively, numerical simulation provides a feasible and efficient way to predict these process parameters and reveal the underlying mechanism of defect formation within metals. Therefore, this Ph.D. research aims to develop numerical models to study the thermal, mechanical and fluid-dynamic behaviours of stainless steel 316L and Ti6Al4V in the SLM additive manufacturing process. Firstly, a three-dimensional thermo-mechanical coupling model based on the finite element method has been developed to simulate the multi-layer and multi-track SLM process of stainless steel 316L. The model considers temperature-dependent material properties and the powder-liquid-solid phase transition which includes melting, vaporization, solidification, shrinkage and cooling phenomena. Temperature and stress fields in the SLM-built specimen can be predicted by solving the heat conduction and stress equilibrium equations. The thermal analysis reveals that both increasing laser power and reducing scanning speed can lead to the increase of the melt pool size. Furthermore, the mechanical analysis suggests that the stress component along the scanning direction is larger than the stress components in the other two directions. Large residual stresses are distributed in the first two layers of a multi-layer printed product due to the constraints from the substrate. Secondly, a thermo-metallurgical-mechanical coupling model has been developed to predict temperature, solid-state phase and residual stress fields in Ti6Al4V for the SLM additive manufacturing process. The thermal analysis is based on the transient heat conduction problem with a volumetric heat source describing the laser absorption and scattering in the powder bed. The volume fraction of metallurgical phases is determined by the temperature field and used to obtain the volumetric change strain due to the solid-state phase transformation (SSPT). An elasto-plastic constitutive law considering the strains induced by thermal gradients and SSPT has been proposed to evaluate stress fields. Modelling results conclude that the consideration of SSPT leads to the decrease of tensile residual stresses and the increase of compressive residual stresses. Finally, a powder-scale multi-physics model has been developed to investigate the melt pool dynamics and porosity for stainless steel 316L additively manufactured by SLM. The effects of the laser power, laser scanning speed and ambient pressure on the thermo-fluid behaviour and pore formation in the melt pool are analyzed. Simulation results reveal that excessive laser energy may cause the formation of keyhole-induced pores, while inadequate laser intensity may give rise to irregular pores at the interlayer. Additionally, the reduced ambient pressure can increase the melt pool depth, whereas the increase is not noticeable below a critical ambient pressure. As compared with the atmospheric pressure, the sub-atmospheric pressure contributes to a lower average temperature and larger average pressure on the keyhole surface. Consequentially, the sub-atmospheric pressure can increase the keyhole stability and thus minimize the pores in SLM-built products. In this Ph.D. research, thermal, mechanical and fluid dynamic models have been developed to predict the temperature, stress and fluid flow fields during the SLM additive manufacturing process. The numerical models can provide insight into the physical understanding of the metal additive manufacturing process. The simulation results can be further utilized to optimize the processing parameters for tailoring the mechanical properties of SLM-built products.Doctor of Philosoph