Enhancing surface quality and tool life in SLM-machined components with Dual-MQL approach

Abstract

Selective laser melting (SLM) can produce complex metal components with high densities, thereby surpassing the limitations of traditional machining methods. However, achieving accurate dimensions, geometries, and acceptable surface states in parts fabricated through SLM remains a concern as they often fall short compared to traditionally machined components. As a solution, a hybrid additive-subtractive manufacturing (HASM) method was developed to effectively utilize the advantages of both techniques. In this study, SLM-made 316 L stainless steel was machined under distinct cooling conditions to investigate the effects of roughness and tool wear. After a thorough investigation, the dual-MQL strategy was evaluated and compared with dry and MQL cutting strategies. The findings showed that the dual-MQL condition led to a significant reduction in flank wear by 54–56% and 29–34%, respectively, associated with dry and MQL cutting techniques, making it a highly promising key for machining SLM-made steel components. Machine learning techniques are potential tools for prediction and classification capabilities in machining processes. For milling SLM-made 316 L SS, multilayer perceptron (MLP) proved to be the most effective prediction model and for classification MLP and Random forest performed better.