[T9] Innovative Technology

[17E-T9-05] Exploration of Steam Oxide in Powder Metallurgy — An Evaluation through Image Recognition

*W. P. Yung¹, F. J. Yan¹, C. Y. Wei ¹, Z. T. Jun¹, L. C. Yi², L. F. Ning ³, H. C. Wei¹ (1.Feng Chia University, Taiwan, 2.Auroral Sinter Metals Co., Taiwan, 3.Brillant Knowbot Machines Co., Taiwan)

キーワード：Steam Oxide, Image Recognition, Powder Metallurgy, Porosity

Recently, the use of AI-assisted methods for assessing the microstructures of metal powders has become increasingly popular. Especially within metal processing forming with powder metallurgy, the thermal steam oxide process plays a crucial role in determining specific mechanical properties for particular purposes. This work estimates the varied oxidation layer thickness using different parameters gleaned from tensile and hardness test results and confirms the reliability of our experimental data by cross-referencing it with BET density measurements. The placement of the test specimen within the water vapor oxidation cage directly influences both the thickness and density of the oxide layer. Moreover, the comprehensive examination encompassed the correlation between the overall surface porosity and hardness of the test sample, which showed a difference in porosity between 13% and 25% for Rockwell hardness of 75 to 98 HRB. We explore the implications of these variations in hardness, porosity, and thickness, leveraging the outcomes for image recognition purposes. These results bolster product optimization, while the in-depth discussion of this identification method offers a novel analysis of the powder metallurgy preparation process.