Multiscale Approach to Fast ModSim for Laser Processing of Metals for Future Nuclear Deterrence Environments
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Predicting performance of parts produced using laser-metal processing remains an out- standing challenge. While many computational models exist, they are generally too computationally expensive to simulate the build of an engineering-scale part. This work develops a reduced order thermal model of a laser-metal system using analytical Green's function solutions to the linear heat equation, representing a step towards achieving a full part performance prediction in an "overnight" time frame. The developed model is able to calculate a thermal history for an example problem 72 times faster than a traditional FEM method. The model parameters are calibrated using a non-linear solution and microstructures and residual stresses calculated and compared to a non-linear case. The calibrated model shows promising agreement with a non-linear solution.
- Research Organization:
- Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA)
- DOE Contract Number:
- AC04-94AL85000; NA0003525
- OSTI ID:
- 1669201
- Report Number(s):
- SAND--2020-9870; 690955
- Country of Publication:
- United States
- Language:
- English
Similar Records
Securing machine learning models
Multi-scale computational modeling of residual stress in selective laser melting with uncertainty quantification