Process Parameter Selection and Optimization for the Prediction of Porosity, Hardness and Roughness in Additive Manufactured Aluminum Alloy

Abstract

Despite the advantages of additive manufacturing, its widespread adoption is still hindered by the poor quality of fabricated parts. Advanced machine learning techniques can improve the repeatability and open up additive manufacturing to a wider range of industries by predicting the part quality. This study aims to accurately predict the relative density, surface roughness, and hardness of AlSi10Mg samples produced by selective laser melting with respect to process parameters such as scan speed, layer thickness, laser power, and hatch distance. For this purpose, porosity, surface hardness and roughness data were extracted from literature and additional measurements were conducted on additive manufactured samples in the current work. Feature importance analysis on the compiled dataset using lasso and random forest revealed that laser power, and layer thickness are the most important features affecting relative density (e.g., porosity), while hatch distance and scan speed significantly impact the surface roughness and hardness of the fabricated parts. In this work, five supervised machine learning algorithms are compared including artificial neural network, support vector regression, kernel ridge regression, random forest, and lasso regression. These models are evaluated based on the mean square error and the coefficient of determination. Based on the computational results, artificial neural network outperformed in predicting relative density, surface roughness and hardness. The study identified an optimal laser power and scan speed region that achieves a relative density > 99%, surface roughness < 10 µm, and hardness > 120 HV. The outcomes presented in this study provide significant advantages for additive manufacturing, potentially reducing experimentation costs by identifying the process parameters that optimize the quality of the fabricated parts.

Date of Award	Aug 2023
Original language	American English
Supervisor	Maher Maalouf (Supervisor)