Investigation of Inconel 718 powder flowability for laser beam powder bed fusion using physics-informed machine learning framework.
Saved in:
| Title: | Investigation of Inconel 718 powder flowability for laser beam powder bed fusion using physics-informed machine learning framework. |
|---|---|
| Authors: | Hossain, Mohammad Shahadath1 (AUTHOR), Silva, Daniel F1,2 (AUTHOR), Vinel, Alexander1,2 (AUTHOR), West, Brian M3 (AUTHOR), Shamsaei, Nima2,4 (AUTHOR), Liu, Jia1,2 (AUTHOR) lzj0040@auburn.edu |
| Source: | Powder Metallurgy. Jun2025, Vol. 68 Issue 3, p197-209. 13p. |
| Subjects: | Feature extraction, Flow coefficient, Laser beams, Inconel, Statistical correlation, Random forest algorithms |
| Abstract: | Powder flowability plays a significant role in powder layering for laser beam powder bed fusion (LB-PBF), which could affect the quality of LB-PBF-fabricated parts. This study aims to investigate the impact of powder features on the flowability of Inconel 718 powder. Powder features, such as size, shape, and other important features, were extracted from 11 Inconel 718 powder samples. With a physics-based understanding of Inconel 718 and correlation analysis, seven powder features were extracted and selected from 133 original powder features for flowability analysis and prediction using machine learning methods. Specifically, least absolute shrinkage and selection operator, random forest regression, and support vector regression were used to predict powder flowability indicated by the angle of repose and flow function coefficient. It is found that the seven selected powder features from physics-informed method and correlation analysis can be excellent predictors of powder flowability. While predicting flowability using different machine learning methods, in random forest regression, the mean absolute percentage of error was 3.61% for angle of repose and 7.73% for flow function coefficient. The outcomes of this physics-informed machine learning framework enhance prior studies and generate a new understanding of powder features and potential impact on powder flowability. [ABSTRACT FROM AUTHOR] |
| Copyright of Powder Metallurgy is the property of Sage Publications Inc. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Powder flowability plays a significant role in powder layering for laser beam powder bed fusion (LB-PBF), which could affect the quality of LB-PBF-fabricated parts. This study aims to investigate the impact of powder features on the flowability of Inconel 718 powder. Powder features, such as size, shape, and other important features, were extracted from 11 Inconel 718 powder samples. With a physics-based understanding of Inconel 718 and correlation analysis, seven powder features were extracted and selected from 133 original powder features for flowability analysis and prediction using machine learning methods. Specifically, least absolute shrinkage and selection operator, random forest regression, and support vector regression were used to predict powder flowability indicated by the angle of repose and flow function coefficient. It is found that the seven selected powder features from physics-informed method and correlation analysis can be excellent predictors of powder flowability. While predicting flowability using different machine learning methods, in random forest regression, the mean absolute percentage of error was 3.61% for angle of repose and 7.73% for flow function coefficient. The outcomes of this physics-informed machine learning framework enhance prior studies and generate a new understanding of powder features and potential impact on powder flowability. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 00325899 |
| DOI: | 10.1177/00325899251339973 |
