.png){kind=icon}
.png){kind=icon}
Return
MOOSE StochasticControl: A python interface for black-box optimization with MOOSE-based simulations
.png){kind=icon}
PRE
AI
DOI：10.1016/j.softx.2026.102682.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
StochasticControl is a Python interface designed to bridge high-fidelity Multiphysics Object Oriented Simulation Environment (MOOSE) simulations with modern data science and optimization workflows. The interface automatically generates MOOSE stochastic tools module inputs, manages parallel execution, and retrieves quantities of interest as NumPy arrays through a concise function call. This approach simplifies the integration of multiphysics simulations with Python libraries for optimization and active learning that require iterative function calls. An illustrative example demonstrates shape optimization of an annulus under thermal loading using SciPy, highlighting performance gains through state retention and parallelism.
Keywords:
MOOSE
Optimization
Python interface
Stochastic
Journal
IF:
2.4
Papers:
325
Citations:
7.3K