Problem
Advanced flow solvers can be difficult to learn from because core ideas are hidden behind large production codebases. This project was built as a compact but capable environment for testing lattice Boltzmann methods and teaching simulation workflows.
Approach
- Implement a modular 2D LBM solver in C/Python with a practical tooling workflow.
- Support multiple collision models, boundary conditions, and turbulence options.
- Include coupled transport and biomedical-oriented add-ons (e.g., margination models).
- Keep geometry setup simple by allowing image-based domain painting.
Key finding
A lightweight solver can still expose many advanced LBM concepts and serve as a productive bridge between education, method prototyping, and early research experiments.
Why it matters
Fast iteration in an understandable codebase helps students and researchers validate ideas before scaling to larger production frameworks.
Outputs
- Source code: medFlow2D on GitHub
- Includes theory-oriented and practical usage documentation in the repository.