Progress towards a Complete Model of Metal Additive Manufacturing

Abstract

Metal additive manufacturing based on powder bed fusion processes is increasingly important. However, highly transient physical phenomena that occur in these processes at different length scales are difficult to observe. Challenging and costly experiments are usually needed to obtain data for process understanding and improvement. Computational modelling of powder-bed fusion processes is therefore important from several points of view. These include better process understanding, optimisation of process parameters and component designs, prediction of component properties, qualification of components and to assist process control. Several physical processes have to be treated to develop a complete model, namely the raking of the powder bed surface, the transfer of energy from the laser or electron beam to the metal, the melting and solidification of the powder, the flow of liquid metal in the melt pool, the heat transfer from the melt pool to the surrounding powder and solid metal, the evolution of the microstructure, and the residual stress and deformation of the component. These processes occur at very different scales, and have to be treated using several different computational techniques. In addition, the interdependency of some of the processes has to be accounted for. This paper discusses the rationale for developing a complete model, progress in developing sub-models of the different physical processes, and the framework that is envisaged to combine the sub-models into a predictive model of the additive manufacturing process.