An interactive 2D airfoil aerodynamics simulator built from scratch in Python.
It solves the incompressible potential-flow field around an airfoil with a
Hess–Smith panel method and lets you explore lift, pressure, and streamlines
in real time in your browser. The solver is the Python package aerosim, in
src/aerosim/.
You need uv; on first run it fetches Python 3.12 and every dependency for you.
git clone https://github.com/markccchiang/Aerodynamic-Simulation-2D.git
cd Aerodynamic-Simulation-2D
uv run python src/web.py # launch the web UI -> http://127.0.0.1:8000
uv run python src/validate.py # run the physics validation suiteDrag the sliders to change the angle of attack, Reynolds number, and the NACA
4-digit shape (MPXX = camber %, camber position, thickness %), or load your own
airfoil (see Loading airfoils). The flow field, surface
pressure plot, and the lift/drag/moment coefficients all update live.
To drive the solver from your own script or notebook instead, see Using the solver from Python.
| Guide | What it covers |
|---|---|
| Web UI | how to use the browser front-end, control by control, with screenshots |
| Using the solver from Python | importing aerosim in scripts and notebooks, a worked example, and what a solve returns |
| Loading airfoils | Selig and Lednicer .dat files, the bundled samples, and the circle / golf-ball pair |
| How it works | the panel method, the viscous drag correction, and two-way coupling |
| Architecture | the three layers, how data flows through them, and the design decisions behind them |
| Validation | what src/validate.py checks, group by group |
| Development | project layout and possible next steps |
| References | the textbooks, papers and data sources behind each method |
The aerodynamic theory behind all of it — the governing equations, why a panel
method solves them, the integral boundary-layer method, the viscous–inviscid
coupling, and how the validation cases are chosen — is written up in full as
Sphinx documentation under docs/. Sphinx and its theme are in the
project's dev dependency group, so uv sync installs them:
uv sync # installs the docs toolchain (dev group)
uv run sphinx-build -M html docs docs/_build # build -> docs/_build/html/index.html
./open-html.sh # open it in a browser (macOS or Linux)cd docs && uv run make html does the same build through the Makefile. The
figures are generated by running the solver at build time, so they always
match the code — which also means that after changing the physics you should
rebuild from clean (rm -rf docs/_build first), or figures cached by the
previous build survive. A clean build should finish with zero warnings.
Released under the MIT License — free to use, modify and redistribute, including in commercial and closed-source work, provided the copyright notice travels with it. Every dependency is permissively licensed as well (NumPy, SciPy and uvicorn under BSD, FastAPI and pydantic under MIT, matplotlib under the PSF licence, and Plotly.js under MIT), so there is no copyleft obligation anywhere in the stack.