DATOFoam - Discrete Adjoint Topology Optimization

Topology optimization software with the discrete adjoint method from DAFoam.

Pre-requisite

Need to build datofoam_base docker image before.

• Clone DATOFOam_base:

git clone https://github.com/ckiyono/datofoam_base.git

• cd into folder:

cd datofoam-base

• Generate docker image of DATOFoam_base:

sudo docker build -t datofoam:base .

• Leave folder:

cd ..

Instalation

• Clone this repository:

git clone https://github.com/ckiyono/datofoam.git

• cd into folder:

cd datofoam

• Generate docker image of DATOFoam:

sudo docker build -t datofoam .

Modifications to the code

If any modification is made to the DATOFoam code, just build the image again.

Running DATOFoam

• cd into an example folder:

cd /path/to/example/folder

• Run a container with datofoam image:

docker run -it --rm -u dafoamuser --mount "type=bind,src=$(pwd),target=/home/dafoamuser/mount" -w /home/dafoamuser/mount datofoam bash

Inside the container, scripts may not have persmission to run. In this case:

• Leave mount folder:

cd ..

• Change permissions:

chmod -R 777 mount

• cd to mount folder

cd mount

Now, to run the optimization:

• Run the preProcessing.sh script to generate the mesh:

./preProcessing.sh

• Change runScript_v2.py file by adding the following code at task options

elif args.task == "runAdjVector":

   optFuncs.runAdjVector()

• Then, use the following command to run the optimization with 4 CPU cores:

mpirun -np 4 python runScript_v2.py --task="runPrimal" 2>&1 | tee logOpt.txt

The option '--task=' can be used after 'runScript_v2.py' to choose the task:

• "opt" (default)
• "runPrimal"
• "runAdjoint"
• "runAdjVector"