Data-driven approach for predicting abnormal grain growth in sintered binder jet steels

Abstract

Austenitic 316 stainless steel printed by binder jetting requires sintering to high densities to minimize porosity and have the corrosion resistance and strength required for most applications. While sintering can achieve densities above 99%, this paper reports the occurrence of abnormal grain growth (AGG) in this high-density region. A comprehensive process map is proposed, integrating key parameters to predict and inform grain sizes using regression model and machine learning approaches. Additionally, a clear relationship is identified between surface roughness, density, and grain size, offering a potential strategy for quality monitoring in serial production.