About this Abstract |
Meeting |
2020 TMS Annual Meeting & Exhibition
|
Symposium
|
Computational Materials Science and Engineering of Materials in Nuclear Reactors
|
Presentation Title |
Reduced Order Modeling of Thermal Creep in 316H Stainless Steel |
Author(s) |
Aaron E. Tallman, M Arul Kumar, Laurent Capolungo |
On-Site Speaker (Planned) |
Aaron E. Tallman |
Abstract Scope |
Hierarchical multiscale modeling has been pursued as a means of bridging the gap between the scale of physics-based material models and the engineering scale. While more efficient than direct numerical simulation, hierarchical methods still retain a numerical expense greater than engineering models. To further reduce the expense of physics-informed simulation, a reduced-order method is proposed. Here, primary and secondary thermal creep of 316H stainless steel are captured using a constitutive model in a viscoplastic self-consistent (VPSC) framework. A reduced-order model (ROM) is proposed which emulates VPSC simulation results using polynomial regression over a reduced number of degrees of freedom, including internal state variables. An application domain is defined, i.e. temperature and applied stress range. A design of experiments is used to ensure fidelity of the ROM to VPSC within the entire chosen domain. Invertible mappings are employed to mitigate compounding errors. The ROM is applied at the engineering scale. |
Proceedings Inclusion? |
Planned: Supplemental Proceedings volume |