About this Abstract |
Meeting |
TMS Specialty Congress 2024
|
Symposium
|
2nd World Congress on Artificial Intelligence in Materials & Manufacturing (AIM 2024)
|
Presentation Title |
Predicting Mechanical Behavior in Creep Conditions: High-Throughput Protocols With Unconventional Geometries and Digital Image Correlation |
Author(s) |
Seyed Mohammad Ali Seyed Mahmoud, Samantha Mitra, Raj Jung Mahat, Ali Khosravani, Surya Kalidindi |
On-Site Speaker (Planned) |
Seyed Mohammad Ali Seyed Mahmoud |
Abstract Scope |
Creep tests are fundamental for understanding the behavior of materials required to endure specific operational temperatures under load. Metals and alloys exhibit notable property variations at different temperatures. Our research focuses on developing high-throughput protocols employing unconventional sample geometries in tandem with digital image correlation to evaluate monotonic mechanical properties. These non-standard geometries induce stress heterogeneity during testing while maintaining controlled and consistent creep stress on samples. Utilizing finite element simulations and machine learning, we aim to construct a surrogate model predicting monotonic mechanical responses. By leveraging data from non-standard experiments and established standard properties of 6061 aluminum alloys, our methodology combines training and calibrating the model. The protocol's validity will be demonstrated through the analysis, establishing a pathway for understanding and predicting mechanical behaviors in materials subjected to varying stress conditions and temperatures. |
Proceedings Inclusion? |
Definite: Other |