Using Machine Learning Techniques to Predict the Dimensional Changes of Low-cost Metal Material Extrusion Fabricated Parts

Abstract

Additive manufacturing (AM) is a widely used layer-by-layer manufacturing process. However, it is limited by material options, different fabrication defects, and inconsistent part quality. Material extrusion (ME) is the most widely used AM technologies. Thus, it is adopted in this research. Low-cost metal ME is a new AM technology used to fabricate metal composite parts using sintering metal infused filament material. Since the materials and the process are relatively new, there is a need to investigate the dimensional accuracy of low-cost metal ME fabricated parts for real-world applications. Each step of the manufacturing process such as 3D printing of the samples and the sintering will affect the dimensional accuracy significantly. By using several machine learning (ML) algorithms, a comprehensive analysis of dimensional changes of metal samples fabricated by low-cost metal ME process is developed in this research. ML methods can assist researchers in sophisticated pre-manufacturing planning and product quality assessment and control. In this study, single linear regression, linear regression with interactions and neural networks were utilized to assess and predict the dimensional changes of components after 3D printing and sintering process. The prediction outcomes using a neural network performed the best with the highest accuracy among the other ML methods. The findings of this study can help researchers and engineers to predict the dimensional variations and optimize the printing and sintering process parameters to obtain high quality metal parts fabricated by the low-cost ME process.