OpenFOAM natural-convection simulation
A completed OpenFOAM v2606 study of buoyancy-driven air flow and heat transfer in a differentially heated square cavity.

Software compatibility
Cases are prepared for the OpenCFD OpenFOAM v2606 distribution. No ANSYS Fluent, COMSOL, STAR-CCM+, or OpenFOAM Foundation v14 files are included.
Project definition
Problem statement
Natural convection couples temperature, density, gravity, momentum, and heat transport. Wall heat transfer and circulation change strongly with Rayleigh number, while high-Rayleigh boundary layers need adequate mesh resolution.
Project objectives
- Generate a standard differentially heated square cavity from declared physical inputs.
- Solve steady laminar air flow and temperature with the Boussinesq approximation in OpenFOAM v2606.
- Compare Rayleigh numbers from 1e3 to 1e6 and three meshes at Ra = 1e6.
- Calculate average Nusselt number, centreline velocities, temperature profiles, heat balance, symmetry, residuals, and continuity.
- Compare the completed results with the de Vahl Davis cavity benchmark.
Project structure
Project components
Case generator
Calculates the hot-wall temperature from the target Rayleigh number and writes complete native OpenFOAM cases.
Buoyant solver
Uses steady laminar buoyantBoussinesqSimpleFoam for coupled velocity, pressure, and temperature.
Result extractor
Reads cell centres and fields to calculate Nusselt, velocity, profile, balance, symmetry, and convergence measures.
Benchmark analyser
Compares four Rayleigh regimes and the three-level Ra = 1e6 mesh response with published values.
Verification
Checks mesh validity, residuals, continuity, heat balance, symmetry, benchmark error, files, and documentation.
Methodology
Project workflow
- 01Generate
Python maps Rayleigh, Prandtl, geometry, gravity, and air properties into a complete case.
- 02Mesh
blockMesh creates the structured square cavity and checkMesh applies strict geometry and topology checks.
- 03Solve
buoyantBoussinesqSimpleFoam calculates the steady coupled flow and temperature fields.
- 04Extract
Native fields are converted into six JSON records, one CSV table, and 17 paired PNG and SVG figures.
- 05Compare
Nusselt values, velocities, mesh response, heat balance, and symmetry are checked and documented.
Demonstration scenario
Four Rayleigh regimes progress from conduction-dominant transfer to strong circulation. The student compares Nusselt number, speed, temperature, and centreline profiles, then explains how the 32 x 32, 48 x 48, and 96 x 96 meshes change the Ra = 1e6 wall-gradient result.
Engineering
Tools and method
- Tools
- The project uses OpenFOAM v2606, Python 3, Matplotlib for subject analysis, simulation, and results.
- Case format
- Native OpenCFD OpenFOAM v2606 dictionaries, structured mesh, fields, solver logs, and cell centres.
- Physics
- Two-dimensional steady laminar air with constant properties and the Boussinesq density approximation.
- Geometry
- A 0.1 m square cavity has a hot left wall, cold right wall, adiabatic horizontal walls, and no-slip physical walls.
- Analysis
- Python preserves six raw result records and regenerates the study table, tests, diagrams, and numerical plots.
Testing
Evaluation
Evaluation measures
- Ra = 1e3, 1e4, and 1e5 Nusselt benchmark errors of 0.051, 0.535, and 1.908 percent
- Ra = 1e6 fine-grid average Nusselt number 8.965779
- Ra = 1e6 fine-grid benchmark error 1.884 percent
- Ra = 1e6 Nusselt error falls from 19.471 to 1.884 percent from 32 x 32 to 96 x 96 cells
- Heat-balance mismatch remains below 0.001 percent in all six cases
- Fine-grid dimensionless horizontal and vertical velocity maxima of 65.689 and 221.631
- Seven automated tests passing
Project boundaries
- Only OpenCFD OpenFOAM v2606 cases are delivered.
- The model is a two-dimensional square-cavity benchmark with one empty cell through the depth.
- The model uses steady laminar air, constant properties, and the Boussinesq approximation.
- Radiation, conjugate wall conduction, variable properties, three-dimensional effects, turbulence, and surface roughness are excluded.
- The three meshes show a clear improving trend but are not presented as a formal grid-convergence-index study.
- No experimental validation is included.
- The project is academic CFD and not certified thermal performance, equipment sizing, fire analysis, or safety evidence.
Included
- 01Six completed steady natural-convection cases
- 02Rayleigh numbers from 1e3 to 1e6
- 03Structured 32 x 32, 48 x 48, and 96 x 96 meshes
- 04Strict geometry and topology checks for every mesh
- 05Nusselt, velocity, heat-balance, symmetry, residual, profile, and benchmark evidence
- 06Six raw JSON records, one CSV study table, and 17 labelled figures in PNG and SVG
- 07Three real natural-convection literature photographs with source and licence records
- 08Complete project files, models, calculations, and analysis material in a private GitHub repository
- 0980-page project documentation in PDF and editable Word formats
- 109-page setup and usage guide
Project record
No information is collected on this page.
- Permanent project ID
- GP-PH-08X8I9X
- Catalogued
- 22 Aug 2026
- Completed
- 23 Aug 2026
- Verified
- 23 Aug 2026
- Demonstration
- Included in repository
Handover
After purchase
- 01Payment is confirmed
The project is marked unavailable and cannot be purchased again.
- 02Repository access is granted
The buyer's submitted GitHub account receives access to the private repository.
- 03The purchase record is delivered
The certification sheet is prepared from the reviewed buyer details and sent privately by email.