Machine Learning Prediction of Porosity in Low-Pressure-Cast AlSi7Mg Alloy with Varying Recycled Aluminum Content
Materials · 06.11.2026 · 19 (21) · 4729
DOI: 10.3390/ma19214729 ↗Kurz gefasst
Fachartikel in der Zeitschrift Materials, online erschienen am 06.11.2026. Verfasst von Jaroslaw Piatkowski und 6 weiteren Autor:innen. Beteiligte Einrichtung: Materials Innovation Laboratory, Faculty of Materials Science and Industrial Digitalization, Silesian University of Technology, Krasinskiego 8, 40-01…. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of Silesia in Katowice.
Abstract (Originalsprache)
Industrial AlSi7Mg produced by low-pressure casting was evaluated with 50–95 wt.% recycled material under two conditions: no intentional Mn correction and Mn correction targeted at Mn/Fe ≈ 0.5. The porosity dataset contained 200 observations; for 60–95 wt.% recycled material, 160 records from 16 experimental conditions also included sample-specific chemical composition. Without Mn correction, MeanPor rose from 0.160% to 6.111%, whereas Mn correction reduced porosity under most conditions. Eight hypothesis-driven candidate feature architectures and 13 model families were assessed by grouped Leave-One-Scrap-Level-Out cross-validation, with hyperparameters tuned only on training data. A second-order regression with predefined interactions using recycled material content, Mn-correction variant, and sample-specific Fe gave the lowest out-of-fold (OOF) error for MeanPor (coefficient of determination (R2) = 0.9910, root mean square error (RMSE) = 0.1691, mean absolute error (MAE) = 0.1508). The reduced Sludge Factor descriptor, SFCr0 = (1 × %Fe) + (2 × %Mn), also improved prediction and reached R2 ≈ 0.986. SHAP (SHapley Additive exPlanations) was used to describe model attribution without causal interpretation. At 90 and 95 wt.% recycled material, Mn correction coincided with higher V values and more regular pore cross-sections. Sample-specific chemistry improved porosity prediction within the investigated process window; independent validation is required before use outside that range.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Materials Innovation Laboratory, Faculty of Materials Science and Industrial Digitalization, Silesian University of Technology, Krasinskiego 8, 40-019 Katowice, Poland.
- Department of Metallurgy and Recycling, Faculty of Materials Engineering and Industrial Digitalization, Silesian University of Technology, Krasińskiego 8, 40-019 Katowice, Poland.
- Department of Production Engineering, Faculty of Materials Science, Silesian University of Technology, Krasinskiego 8, 40-019 Katowice, Poland.
- Environmental Research Department, Institute of Technology and Business in České Budějovice, Okružní 517/10, 370 01 České Budějovice, Czech Republic.
- Institute of Physics-Centre for Science and Education Silesian, University of Technology, Krasińskiego 8, 40-019 Katowice, Poland.
- Faculty of Mining, Safety Engineering and Industrial Automation, Silesian University of Technology, Ul. Akademicka 2, 44-100 Gliwice, Poland.
ORCID-Profile: Jaroslaw Piatkowski, Grzegorz Siwiec, Tomasz Matula, Beata Oleksiak, Karel Gryc, Andrzej Grabowski, Katarzyna Nowinska
Förderung
- University of Silesia in Katowice
Zitieren
APA
Piatkowski, J., Siwiec, G., Matula, T., Oleksiak, B., Gryc, K., Grabowski, A., & Nowinska, K. (2026). Machine Learning Prediction of Porosity in Low-Pressure-Cast AlSi7Mg Alloy with Varying Recycled Aluminum Content. Materials, 19(21), 4729. https://doi.org/10.3390/ma19214729
Vancouver
Piatkowski J, Siwiec G, Matula T, Oleksiak B, Gryc K, Grabowski A, et al. Machine Learning Prediction of Porosity in Low-Pressure-Cast AlSi7Mg Alloy with Varying Recycled Aluminum Content. Materials. 2026;19(21):4729. doi:10.3390/ma19214729
BibTeX
@article{Piatkowski2026,
author = {Piatkowski, Jaroslaw and Siwiec, Grzegorz and Matula, Tomasz and Oleksiak, Beata and Gryc, Karel and Grabowski, Andrzej and Nowinska, Katarzyna},
title = {Machine Learning Prediction of Porosity in Low-Pressure-Cast AlSi7Mg Alloy with Varying Recycled Aluminum Content},
journal = {Materials},
year = {2026},
volume = {19},
number = {21},
pages = {4729},
doi = {10.3390/ma19214729}
}
EUVIMED