Skip to content

Manual Installation Guide

Jump to bottom
RayanDhib edited this page Aug 16, 2024 · 8 revisions

This guide provides instructions on how to manually install and run CFD-HOWL without using Docker. This approach is suitable for users who prefer to manage dependencies directly on their system.

Prerequisites

Before you begin, ensure that your system has the following installed:

  • Python 3.8 or higher
  • pip3 (Python package installer)
  • git (for cloning the repository)

Step 1: Install Python Dependencies

First, install the required Python packages using pip3. These include numpy and scipy:

pip3 install numpy scipy

Step 2: Install pyCGNS

For installing pyCGNS and its dependencies (e.g., VTK, HDF5), please follow the official installation guide provided by the pyCGNS project:

This guide provides detailed instructions for various platforms, including the necessary system dependencies and installation steps.

Step 3: Clone the CFD-HOWL Repository

Next, clone the CFD-HOWL repository to your local machine:

git clone https://github.com/RayanDhib/CFD-HOWL.git
cd CFD-HOWL

Step 4: Verify config.json

The config.json file in the root directory is already configured with the correct paths to run the provided example. The example is the result of the simulation of an inviscid subsonic flow through a channel with a sinusoidal bump in 2D using Flux Reconstruction (P3). For more details, check the Usage examples page.

You generally do not need to modify this file unless you move the example files to different locations or you are using different inputs. However, it's a good idea to verify that the paths match your local setup.

Step 5: Run CFD-HOWL

To run CFD-HOWL, execute the main.py script:

python main.py

This script will process the input data specified in config.json and generate a high-order CGNS file in the output directory.

Step 6: Review the Output

After running the script, the output CGNS file will be available in the directory specified in config.json. You can open this file with visualization tools such as ParaView or Tecplot to inspect the high-order data.

Troubleshooting

If you encounter any issues, please refer to our Troubleshooting page for help.

Clone this wiki locally