Safdar, Mutahar, Wood, Gentry, Zimmermann, Max, Hannesen, Florian, Paul, Padma, Lamouche, Guy, Wanjara, Priti et Zhao, Yaoyao Fiona. 2025. « Applying artificial intelligence in directed energy deposition additive manufacturing of metal matrix composites: Lessons and experiences for industrial implementation ». Communication lors de la conférence : CSME-CFDSC-CSR 2025 International Congress (Montreal, QC, Canada, May 25-28, 2025).
Prévisualisation |
PDF
123 - Applying artificial intelligence i.pdf - Version publiée Licence d'utilisation : Tous les droits réservés aux détenteurs du droit d'auteur. Télécharger (205kB) | Prévisualisation |
Résumé
Additive manufacturing technologies are prone to issues of process repeatability and part quality. Artificial intelligence offers significant potential to address these quality concerns. Machine learning techniques are increasingly leveraged to support additive manufacturing process development and industrial applications. This study focuses on a real-world industrial project that applied artificial intelligence to support directed energy deposition additive manufacturing of metal matrix composites. The project timeline is divided into major development phases based on the application domain (manufacturing) and digital technology (artificial intelligence). Challenges and insights encountered during the development process are grouped into ten unique themes: five manufacturing themes and five artificial intelligence themes. These include minimizing the knowledge gap (domain understanding, artificial intelligence potential), identifying candidate concerns (application scope, relevant machine learning models), developing data and modeling workflows (data quality, data quantity, model complexity, development resources), deploying models in production (integration, validation), and maintaining artificial intelligence integration with manufacturing systems (continuous update and adaptation). These themes collectively underscore the importance of interdisciplinary collaboration to address the multifaceted challenges of integrating artificial intelligence with advanced manufacturing processes. In this context, the presentation highlights in-situ defect detection through vision-based deep learning models applied to thermal imaging as a case study to explore these challenges and insights. By leveraging real-time data from thermal cameras, the project demonstrates the potential of artificial intelligence to enhance process monitoring and defect detection in directed energy deposition additive manufacturing. This approach exemplifies how in-process data can contribute to expedited development, offering the potential to improve process repeatability, while lowering the need for post-process evaluations. The objective of this work is to support the effective and realistic implementation of artificial intelligence in additive manufacturing projects, while identifying potential research directions at the intersection of these domains. The findings aim to guide the additive manufacturing industry in addressing critical considerations for industrial implementation of artificial intelligence and future advancements.
| Type de document: | Communication (Communication) |
|---|---|
| Informations complémentaires: | Progress in Canadian Mechanical Engineering, Volume 8. Co-chairs: Lucas A. Hof, Giuseppe Di Labbio, Antoine Tahan, Marlène Sanjosé, Sébastien Lalonde and Nicole R. Demarquette. |
| Date de dépôt: | 18 déc. 2025 14:19 |
| Dernière modification: | 18 déc. 2025 14:19 |
| URI: | https://espace2.etsmtl.ca/id/eprint/31970 |
Actions (Authentification requise)
 |
Dernière vérification avant le dépôt |