Supervised Machine Learning and Physics Machine Learning approach for prediction of peak temperature distribution in Additive Friction Stir Deposition of Aluminium Alloy
PLOS ONE · 04.04.2025 · 20 (4) · e0309751
DOI: 10.1371/journal.pone.0309751 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 04.04.2025. Verfasst von Akshansh Mishra und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Industrial and Information Engineering, Politecnico Di Milano, Milan, Italy. Hauptthemen laut MEDLINE: Alloys, Aluminum, Supervised Machine Learning. Die Arbeit wurde bisher 18-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: King Saud University.
Abstract (Originalsprache)
Additive friction stir deposition (AFSD) is a novel solid-state additive manufacturing technique that circumvents issues of porosity, cracking, and properties anisotropy that plague traditional powder bed fusion and directed energy deposition approaches. However, correlations between process parameters, thermal profiles, and resulting microstructure in AFSD still need to be better understood. This hinders process optimization for properties. This work employs a framework combining supervised machine learning (SML) and physics-informed neural networks (PINNs) to predict peak temperature distribution in AFSD from process parameters. Eight regression algorithms were implemented for SML modeling, while four PINNs leveraged governing equations for transport, wave propagation, heat transfer, and quantum mechanics. Across multiple statistical measures, ensemble techniques like gradient boosting proved superior for SML, with the lowest MSE of 165.78. The integrated ML approach was also applied to classify deposition quality from process factors, with logistic regression delivering robust accuracy. By fusing data-driven learning and fundamental physics, this dual methodology provides comprehensive insights into tailoring microstructure through thermal management in AFSD. The work demonstrates the power of bridging statistical and physics-based modeling for elucidating AM process-property relationships.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- School of Industrial and Information Engineering, Politecnico Di Milano, Milan, Italy.
- Department of Mechanical Engineering, School of Engineering and Applied Sciences, Bennett University, Noida, Uttar Pradesh, India.
- Department of Mechanical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan.
- Department of Biosciences, Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences, Chennai, Tamil Nadu, India.
- Department of Mechanical Engineering, Gazi University Faculty of Engineering, Maltepe, Ankara, Turkey.
- Department of Machining, Assembly and Engineering Metrology, Faculty of Mechanical Engineering, VSB-Technical University of Ostrava, Ostrava, Czech Republic.
ORCID-Profile: Akshansh Mishra, Sachin Salunkhe, Robert Cep
Förderung
- King Saud University
Zitieren
APA
Mishra, A., Jatt, V., Sefene, E. M., Salunkhe, S., Cep, R., & Abouel Nasr, E. (2025). Supervised Machine Learning and Physics Machine Learning approach for prediction of peak temperature distribution in Additive Friction Stir Deposition of Aluminium Alloy. PLOS ONE, 20(4), e0309751. https://doi.org/10.1371/journal.pone.0309751
Vancouver
Mishra A, Jatt V, Sefene EM, Salunkhe S, Cep R, Abouel Nasr E. Supervised Machine Learning and Physics Machine Learning approach for prediction of peak temperature distribution in Additive Friction Stir Deposition of Aluminium Alloy. PLOS ONE. 2025;20(4):e0309751. doi:10.1371/journal.pone.0309751
BibTeX
@article{Mishra2025,
author = {Mishra, Akshansh and Jatt, Vijaykumar and Sefene, Eyob Messele and Salunkhe, Sachin and Cep, Robert and Abouel Nasr, Emad},
title = {Supervised Machine Learning and Physics Machine Learning approach for prediction of peak temperature distribution in Additive Friction Stir Deposition of Aluminium Alloy},
journal = {PLOS ONE},
year = {2025},
volume = {20},
number = {4},
pages = {e0309751},
doi = {10.1371/journal.pone.0309751}
}
EUVIMED