PREDICTION OF HARDNESS VALUES OF AGED SELECTIVE LASER MELTED AlSi10Mg ALLOY DATA WITH MACHINE LEARNING METHODS

Abstract

Aluminum manufactured with the Selective Laser Melting (SLM) method has been the subject of many research due to the benefits it provides, especially when used in the automotive and aviation industries. Therefore, it is important to examine and improve the mechanical properties of Al parts produced by the SLM method. Many experiments are needed to examine and improve the mechanical properties of SLM Al materials. This situation causes losses in terms of both time and cost. In this study, aims to estimate the hardness values of SLM AlSi10Mg materials that have been aged. For this purpose, aging processes were applied to SLM AlSi10Mg materials at different times and temperatures, and different machine learning methods were used to predict the hardness values using the hardness values obtained because of the process. Random Forest Regression (RFR) algorithm and Artificial Neural Network (ANN) were used in the study. As a result of the study, it was determined that the hardness values estimated by the ANN (R2 0.9276) method were close to the real hardness values. This is proof that it is possible to predict hardness values using the machine learning method.