Numerical Model based Reliability Estimation of Selective Laser Melting Process

Selective laser melting is developing into a standard manufacturing technology with applications in various sectors. However, the process is still far from being at par with conventional processes such as welding and casting, the primary reason of which is the unreliability of the process. While various numerical modelling and experimental studies are being carried out to better understand and control the process, there is still a lack of research into establishing the reliability of the process.In this paper, a combined modelling-experimental approach is introduced to establish the reliability of the selective laser melting process. A validated 3D finite-volume alternating-direction-implicit numerical technique is used to model the selective laser melting process, and is calibrated against results from single track formation experiments. Correlation coefficients are determined for process input parameters such as laser power, speed, beam profile, etc. Subsequently, uncertainties in the processing parameters are utilized to predict a range for the various outputs, using a Monte Carlo method based uncertainty analysis methodology, and the reliability of the process is established.