Installation

This page provides instructions for installing WEIS and its dependencies.

Pre-requisites

Before installing WEIS, ensure you have the following prerequisites:

  • Python Environment: WEIS requires Python >=3.9 with the Anaconda or Miniforge3 package managers

  • Operating Systems: WEIS is supported on:

    • Linux

    • macOS

    • Windows Subsystem for Linux (WSL)

    • Native Windows

  • Compiler Requirements:

    • Windows: m2w64-toolchain and libpython

    • Linux/Mac: compilers

  • Git: Required for downloading the source code

  • Additional Packages: Some optional features require additional packages:

    • petsc4py (Linux/Mac only)

    • mpi4py

    • pyoptsparse (for optimization capabilities)

For those working behind company firewalls, you may need to change the conda authentication with:

conda config --set ssl_verify no

Proxy servers can also be set with:

conda config --set proxy_servers.http http://id:pw@address:port
conda config --set proxy_servers.https https://id:pw@address:port

Installation Instructions

WEIS can be installed following the instructions below. The installation process differs slightly depending on your system.

Standard Installation (Linux, macOS, Windows)

  1. Setup and activate the Anaconda environment:

    conda config --add channels conda-forge
conda install git
git clone https://github.com/NLRWindSystems/WEIS.git
cd WEIS
git checkout branch_name                         # (Only if you want to switch branches, say "develop")
conda env create --name weis-env -f environment.yml
conda activate weis-env                          # (if this does not work, try source activate weis-env)

  2. Add final packages and install the software:

    conda install -y petsc4py mpi4py pyoptsparse     # (Mac / Linux only, sometimes Windows users may need to install mpi4py)
pip install -e .

  3. If you want to use the more advanced meshing capability for BEM modeling, install the following after you install WEIS:

    pip install pygmsh==7.1.17
pip install https://github.com/LHEEA/meshmagick/archive/refs/tags/3.4.zip
pip install trimesh                              # (Only if you want to use the internal function to plot the mesh)

Installation on Kestrel (DOE HPC System)

  1. Purge existing modules and load conda:

    module purge
module load conda

  2. Setup and activate the Anaconda environment:

    conda config --add channels conda-forge
conda install git
git clone https://github.com/NLRWindSystems/WEIS.git
cd WEIS
git checkout branch_name                         # (Only if you want to switch branches, say "develop")
conda env create --name weis-env -f environment.yml
conda activate weis-env                          # (if this does not work, try source activate weis-env)

  3. Load required modules and install:

    module load comp-intel intel-mpi mkl
module unload gcc
pip install --no-deps -e . -v

Using WEIS After Installation

To use WEIS after installation is complete, you will always need to activate the conda environment first:

conda activate weis-env   # (or source activate weis-env)

On Kestrel, make sure to reload the necessary modules.

For Windows users, we recommend installing git and the m264 packages in separate environments as some of the libraries appear to conflict such that WISDEM cannot be successfully built from source. The git package is best installed in the base environment.

Installing SNOPT for use within WEIS

SNOPT is a commercial optimization solver that can be used with WEIS through the pyOptSparse package. If you wish to use SNOPT, you’ll need to:

  1. Purchase SNOPT: SNOPT is available for purchase from Stanford Business Software Inc..

  2. Install SNOPT with pyOptSparse: There are two approaches to install SNOPT:

    Option 1: Before installing WEIS (Recommended)

    Before running the final pip install -e . command for WEIS:

    1. Clone pyOptSparse and build it from source:

      git clone https://github.com/mdolab/pyoptsparse.git
cd pyoptsparse

    2. Create a folder called pyoptsparse/pyoptsparse/pySNOPT/source if it doesn’t exist

    3. Place the SNOPT source files in that folder: - Copy all files from the src folder in your SNOPT package - Do not include snopth.f file

    4. Install pyOptSparse from source:

      pip install -e .

    5. Continue with WEIS installation (pip install -e .)

    Option 2: After installing WEIS

    If you’ve already installed WEIS:

    1. Uninstall the conda-installed pyOptSparse:

      conda remove --force pyoptsparse

    2. Follow the same steps as in Option 1 to install pyOptSparse from source with SNOPT

Additional details on installing SNOPT with pyOptSparse can be found in the pyOptSparse documentation.

Note

SNOPT is particularly useful for constrained optimization problems in WEIS. It is not required but provides enhanced capabilities compared to the open-source optimizers that come with WEIS.