AI-Powered Very-High-Cycle Fatigue Control: Optimizing Microstructural Design for Selective Laser Melted Ti-6Al-4V
Materials · 26.03.2025 · 18 (7) · 1472
DOI: 10.3390/ma18071472 ↗Kurz gefasst
Fachartikel in der Zeitschrift Materials, online erschienen am 26.03.2025. Verfasst von Mustafa Awd und 1 weiteren Autor:innen. Beteiligte Einrichtung: Institute for Informatics and Automation (IIA), Bremen City University of Applied Sciences (HSB), Flughafenallee 10, D-28199 Bremen, Germany. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: DFG, German Research Foundation (Deutsche Forschungsgemeinschaft, DFG).
Abstract (Originalsprache)
Integrating machine learning into additive manufacturing offers transformative opportunities to optimize material properties and design high-performance, fatigue-resistant structures for critical applications in aerospace, biomedical, and structural engineering. This study explores mechanistic machine learning techniques to tailor microstructural features, leveraging data from ultrasonic fatigue tests where very high cycle fatigue properties were assessed up to 1×1010 cycles. Machine learning models predicted critical fatigue thresholds, optimized process parameters, and reduced design iteration cycles by over 50%, leading to faster production of safer, more durable components. By refining grain orientation and phase uniformity, fatigue crack propagation resistance improved by 20–30%, significantly enhancing fatigue life and reliability for mission-critical aerospace components, such as turbine blades and structural airframe parts, in an industry where failure is not an option. Additionally, the machine learning-driven design of metamaterials enabled structures with a 15% weight reduction and improved yield strength, demonstrating the feasibility of bioinspired geometries for lightweight applications in space exploration, medical implants, and high-performance automotive components. In the area of titanium and aluminum alloys, machine learning identified key process parameters such as temperature gradients and cooling rates, which govern microstructural evolution and enable fatigue-resistant designs tailored for high-stress environments in aircraft, biomedical prosthetics, and high-speed transportation. Combining theoretical insights and experimental validations, this research highlights the potential of machine learning to refine microstructural properties and establish intelligent, adaptive manufacturing systems, ensuring enhanced reliability, performance, and efficiency in cutting-edge engineering applications.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Institute for Informatics and Automation (IIA), Bremen City University of Applied Sciences (HSB), Flughafenallee 10, D-28199 Bremen, Germany.
- Testia GmbH, Airbus Group, Cornelius-Edzard-Straße 15, D-28199 Bremen, Germany.
- Chair of Materials Test Engineering (WPT), Faculty of Mechanical Engineering, TU Dortmund University, Baroper Str. 303, D-44227 Dortmund, Germany.
ORCID-Profile: Mustafa Awd, Frank Walther
Förderung
- DFG
- German Research Foundation (Deutsche Forschungsgemeinschaft, DFG)
Zitieren
APA
Awd, M., & Walther, F. (2025). AI-Powered Very-High-Cycle Fatigue Control: Optimizing Microstructural Design for Selective Laser Melted Ti-6Al-4V. Materials, 18(7), 1472. https://doi.org/10.3390/ma18071472
Vancouver
Awd M, Walther F. AI-Powered Very-High-Cycle Fatigue Control: Optimizing Microstructural Design for Selective Laser Melted Ti-6Al-4V. Materials. 2025;18(7):1472. doi:10.3390/ma18071472
BibTeX
@article{Awd2025,
author = {Awd, Mustafa and Walther, Frank},
title = {AI-Powered Very-High-Cycle Fatigue Control: Optimizing Microstructural Design for Selective Laser Melted Ti-6Al-4V},
journal = {Materials},
year = {2025},
volume = {18},
number = {7},
pages = {1472},
doi = {10.3390/ma18071472}
}
EUVIMED