PSI - Issue 57

ScienceDirect Structural Integrity Procedia 00 (2022) 000 – 000 Structural Integrity Procedia 00 (2022) 000 – 000 Available online at www.sciencedirect.com Available online at www.sciencedirect.com ScienceDirect Available online at www.sciencedirect.com ScienceDirect

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

Procedia Structural Integrity 57 (2024) 73–78

© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0 ) Peer-review under responsibility of the scientific committee of the Fatigue Design 2023 organizers Abstract Additively manufactured metal lattice structures have significant advantages in weight reduction, thermal insulation, high specific strength, and energy damping over conventional manufactured materials, showing great potentials in the applications of aerospace, biomedical, and transportation industries. However, the surface defects, produced by the multiple phenomena during the additive manufacturing process, are the main origins of premature crack initiation and lead to early fatigue failure under cyclic loading. Strut-based lattice structures are composed of micro-sized struts, the fatigue properties of these struts have great influences on the entire components. A thorough understanding of the fatigue behaviors of additively manufactured micro-sized parts requires extensive full-scale fatigue testing, which is costly and time-consuming. This study focuses on estimating the fatigue life of Electron Beam Melting (EBM) manufactured Ti-6Al-4V micro-sized parts using a combination of machine learning and Finite Element Modelling (FEM) approach. To this end, a Generative Adversarial Network (GAN) is trained to generate 2D surface profiles of the EBM manufactured micro-sized parts. Next, the regenerated 2D surface profiles are randomly used to create 2D finite element models to detect the critical notches and get the stress gradients at the hot spot. Finally, Continuous Damage Mechanics (CDM) and Theory of Critical Distance (TCD) are implemented to estimate the fatigue life. This way, hundreds of simulations are performed using regenerated synthetic surface profiles. The obtained results show that using both GAN and FEM makes it possible to numerically reproduce the scattered fatigue data, which is the inherent characteristic of additively manufactured materials. © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the Fatigue Design 2023 organizers Keywords: Lifetime Prediction, Additive Manufacturing, Electron Beam Melting, Machine Learning Abstract Additively manufactured metal lattice structures have significant advantages in weight reduction, thermal insulation, high specific strength, and energy damping over conventional manufactured materials, showing great potentials in the applications of aerospace, biomedical, and transportation industries. However, the surface defects, produced by the multiple phenomena during the additive manufacturing process, are the main origins of premature crack initiation and lead to early fatigue failure under cyclic loading. Strut-based lattice structures are composed of micro-sized struts, the fatigue properties of these struts have great influences on the entire components. A thorough understanding of the fatigue behaviors of additively manufactured micro-sized parts requires extensive full-scale fatigue testing, which is costly and time-consuming. This study focuses on estimating the fatigue life of Electron Beam Melting (EBM) manufactured Ti-6Al-4V micro-sized parts using a combination of machine learning and Finite Element Modelling (FEM) approach. To this end, a Generative Adversarial Network (GAN) is trained to generate 2D surface profiles of the EBM manufactured micro-sized parts. Next, the regenerated 2D surface profiles are randomly used to create 2D finite element models to detect the critical notches and get the stress gradients at the hot spot. Finally, Continuous Damage Mechanics (CDM) and Theory of Critical Distance (TCD) are implemented to estimate the fatigue life. This way, hundreds of simulations are performed using regenerated synthetic surface profiles. The obtained results show that using both GAN and FEM makes it possible to numerically reproduce the scattered fatigue data, which is the inherent characteristic of additively manufactured materials. © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the Fatigue Design 2023 organizers Keywords: Lifetime Prediction, Additive Manufacturing, Electron Beam Melting, Machine Learning Fatigue Design 2023 (FatDes 2023) Fatigue life prediction of electron beam melted micro-sized parts based on regenerated surfaces using machine learning approach Yixuan Hou a , Steve Kench b ,Tony Wouters c , Reza Talemi a, * Fatigue Design 2023 (FatDes 2023) Fatigue life prediction of electron beam melted micro-sized parts based on regenerated surfaces using machine learning approach Yixuan Hou a , Steve Kench b ,Tony Wouters c , Reza Talemi a, * a KU Leuven, Department of Materials Engineering, 3001 Leuven, Belgium b Dyson School of Design Engineering, Imperial College London, London, UK c KU Leuven, Department of Computer Science, 3001 Leuven, Belgium a KU Leuven, Department of Materials Engineering, 3001 Leuven, Belgium b Dyson School of Design Engineering, Imperial College London, London, UK c KU Leuven, Department of Computer Science, 3001 Leuven, Belgium

* Corresponding author. Tel.: +32 9331 6501. E-mail address: reza.talemi@kuleuven.be * Corresponding author. Tel.: +32 9331 6501. E-mail address: reza.talemi@kuleuven.be

2452-3216 © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the Fatigue Design 2023 organizers 2452-3216 © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the Fatigue Design 2023 organizers

2452-3216 © 2024 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the Fatigue Design 2023 organizers 10.1016/j.prostr.2024.03.009