PANDA: Polytopal Algorithm for Numerical Discretization and Analysis

PANDA is a research-oriented numerical framework for solving partial differential equations using polytopal methods on general polyhedral meshes. The code is designed to work directly with real polytopal meshes and supports problems such as Poisson, Stokes....

Features

• Polytopal discretization methods
• Support for arbitrary polyhedral meshes
• Solvers for:
  • Poisson equation
  • Stokes equations (and extensions)
• Direct reading of MED mesh format
• Output in VTK format for visualization
• Lightweight dependency model

Dependencies

PANDA has a single external dependency:

• MedCoupling (from the SALOME platform)

---

Installation

1. Clone the repository

git clone https://github.com/mohd-afeef-badri/panda.git

2. Download MedCoupling

Download a MedCoupling binary distribution (standalone or via SALOME) compatible with your system.

---

Running the Code

Before running PANDA, you must initialize the SALOME/MedCoupling environment.

1. Load the MedCoupling environment

./salome context

2. Run an example solver

Poisson problem

cd panda/poisson
python main.py

The solver:

• Reads a polytopal mesh in MED format in the panda/poisson/mesh folder
• Assembles and solves the discrete system
• Writes results to VTK files
• Results can be visualized using ParaView

Example:

paraview panda/poisson/solution/solution.vtk

Applications

PANDA is suitable for:

• Research in polytopal / polygonal / polyhedral methods
• Benchmarking PDE solvers on general meshes
• Rapid prototyping of new discretization schemes
• Educational use in numerical PDEs

---

License

@@@@