Home¶

Test: Discrete Adjoint with OpenFoam¶
Test contains a suite of discrete adjoint solvers for OpenFOAM. These adjoint solvers run as standalone executives to compute derivatives. Test also has a Python interface that allows the adjoint solvers to interact with external modules for high-fidelity design optimization. Test has the following features:
- It implements an efficient discrete adjoint approach with competitive speed, scalability, accuracy, and compatibility.
- It allows rapid discrete adjoint development for any steady-state OpenFOAM solvers with modifying only O(100) lines of source codes.
- It supports design optimizations for a wide range of disciplines such as aerodynamics, heat transfer, structures, hydrodynamics, and radiation.
The Test repository comprises of five main directories:
- applications: adjoint solvers and utilities
- doc: documentation
- python: python interface to other optimization packages
- src: the core Test libraries
- tutorials: sample optimization setup for each adjoint solver
Contents:
Installation¶
Test runs on Linux systems and is based on OpenFOAM-v1812. You must install OpenFOAM and verify that it is working correctly. You also need to install the 3rd party and MDOLab packages before using Test for optimization. Other dependencies include:
- C/C++ compilers (gcc/g++ or icc/icpc)
- Fortran compiler (gfortran or ifort)
- MPI software (openmpi, mvapich2, or impi)
- Swig
- cmake
To compile the documentation, you also need: