Abstract
The computer code uses a parallel simulated annealing framework with embedded machine learning components to solve multi-constrained optimization problems. The software automatically balances the execution of low and high fidelity physics models within the optimization procedure. The low fidelity model is used to rapidly explore the design space while the high fidelity physics model is executed sparingly to account for complex design constraints that are not resolved by the quickly executing low fidelity model.
- Developers:
-
Gurecky, William [1] ; Collins, Benjamin ; Kropaczek, Dave [1] ; Pandya, Tara [1] ; Laiu, Ming Tse [1]
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
- Release Date:
- 2022-01-10
- Project Type:
- Open Source, Publicly Available Repository
- Software Type:
- Scientific
- Programming Languages:
-
Python
- Licenses:
-
Apache License 2.0
- Sponsoring Org.:
-
USDOEPrimary Award/Contract Number:AC05-00OR22725
- Code ID:
- 69252
- Research Org.:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
- Country of Origin:
- United States
- Keywords:
- simulated annealing, optimization, multifidelity
Citation Formats
Gurecky, William, Collins, Benjamin, Kropaczek, Dave, Pandya, Tara, and Laiu, Ming Tse P.
ML-PSA.
Computer Software.
https://code.ornl.gov/ml-psa/ml-psa.
USDOE.
10 Jan. 2022.
Web.
doi:10.11578/dc.20220110.2.
Gurecky, William, Collins, Benjamin, Kropaczek, Dave, Pandya, Tara, & Laiu, Ming Tse P.
(2022, January 10).
ML-PSA.
[Computer software].
https://code.ornl.gov/ml-psa/ml-psa.
https://doi.org/10.11578/dc.20220110.2.
Gurecky, William, Collins, Benjamin, Kropaczek, Dave, Pandya, Tara, and Laiu, Ming Tse P.
"ML-PSA." Computer software.
January 10, 2022.
https://code.ornl.gov/ml-psa/ml-psa.
https://doi.org/10.11578/dc.20220110.2.
@misc{
doecode_69252,
title = {ML-PSA},
author = {Gurecky, William and Collins, Benjamin and Kropaczek, Dave and Pandya, Tara and Laiu, Ming Tse P.},
abstractNote = {The computer code uses a parallel simulated annealing framework with embedded machine learning components to solve multi-constrained optimization problems. The software automatically balances the execution of low and high fidelity physics models within the optimization procedure. The low fidelity model is used to rapidly explore the design space while the high fidelity physics model is executed sparingly to account for complex design constraints that are not resolved by the quickly executing low fidelity model.},
doi = {10.11578/dc.20220110.2},
url = {https://doi.org/10.11578/dc.20220110.2},
howpublished = {[Computer Software] \url{https://doi.org/10.11578/dc.20220110.2}},
year = {2022},
month = {jan}
}