Mathematical Modeling and Optimization of Process Parameters in 3D Printing of AlFeSi10Mg Components Using Neural Network and RSM

Abstract

Selected laser melting (SLM) products made of aluminum have been widely employed in biomedical, industries, and aerospace. However,SLM’s unique process parameters make it difficult to efficiently print desired objects.SLM can be used to print high strength aluminum alloys that can be optimized for processing. For parameter optimization of Scanning Speed, hatching distance, Layer height, and laser speed, D-Optimal design of experiments approach is utilized. We develop parameter windows for these three parameters (LED, SED, VED) about part density using 36 samples. The density data collected from the samples via analysis software agrees well with the numerical model calculated.     

SLM printing of Al products requires a pre-processing optimization system because SLM demands so much time, money, and professional understanding of the process and materials. An SLM optimization system based on a supervised Artificial Neural Network is created in this research. The ideal SLM process parameters, which may be employed to manufacture a product that meets a user's need, are the outputs of this optimization method. An SLM operator does not require a lot of knowledge or a lot of time to experiment with this optimization system to print a suitable result. This system is a very important element in the pre-processing of SLM printing.