| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Verification, Calibration, and Validation Approaches in Modeling the Mechanical Performance of Metallic Materials
|
| Presentation Title |
Establishing Temperature-Based Relationships for Mechanical Properties and Crystal Plasticity Parameters of Additively Manufactured Haynes-214 Alloy |
| Author(s) |
Mohammad M Keleshteri, Mehrdad Pourjam, Subhadip Sahoo, Jason Mayeur, Kavan Hazeli |
| On-Site Speaker (Planned) |
Mohammad M Keleshteri |
| Abstract Scope |
Crystal plasticity (CP) simulations are practical tools for understanding microstructure-property relationships under high temperatures and stress, reducing the need for costly mechanical tests. However, CP model calibration is time-consuming and computationally expensive. To address this, we developed a machine learning–differential evolution (ML-DE) CP framework to predict the tensile behavior of an alloy over a wide temperature range (ambient to 870°C). We used electron backscatter diffraction (EBSD) data to create microstructural volume elements for CP simulations. Subsequently, we conducted 1000 CP simulations to train three machine learning regression algorithms: linear, extra-trees, and multi-layer perceptron. These models were independently evaluated to compare their efficiency. The ML-DE optimization model was then used to calibrate CP parameters using experimental temperature data. This allowed us to formulate temperature-dependent mechanical properties and CP parameters. The framework's effectiveness and efficiency were confirmed through validation against experimental results. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
High-Temperature Materials, Machine Learning, Other |