Application of machine learning methods to optimize quality of conductive surface and permeability of porous structures produced by selective laser melting

The problem of optimizing the surface quality and permeability of porous structures produced by selective laser melting (SLM) is considered. Modern research highlights the importance of surface roughness control and accurate porosity assessment in order to achieve high product quality standards, especially in high-tech industries such as rocket-and-space technology, aviation and medicine. The influence of various factors, such as the parameters of the printing and the conditions of heat treatment, on the characteristics of the products has been examined. The obtained results show that the printing parameters do not significantly affect the roughness of the surface, which indicates the importance of other factors, such as the dynamics of the melting pool and the layered additive design. Additionally, using mathematical methods and Darcy's law, an analysis of the water permeability of the samples was carried out, which made it possible to identify patterns between the printing parameters, porosity and conductivity of the liquid. The results obtained can be useful to engineers and technologists involved in the implementation and optimization of SLM printing processes to improve the functionality and quality of products.