In the short term, priority will be given to bug fixing. To report a bug, please register/login and post a request in the Q&A section.
Next in priority will be the addition of user-requested features. These will be addressed in order based on the following criteria:
- Interest for all users
- Added value to the application
- Interest/work load ratio
Finally, there is a long list of long-term features to implement. In no particular order, the most important ones are:
- implementation of a foil scripting capability in xflr5 to make it possible to run combined foil + plane scripts
- Improvement of the quality and robustness of the surface triangular mesher
- Optimisation of flow5 for speed – GUI and analysis tasks
- Improvement of the quality of the tessellator
- Implementation of a general triangular panel method for unstructured meshes such as those imported from STL files.
- Implementation of a simplified wake roll-up model. Now that a linear triangular Galerkin model has been added to the analysis capabilities, the flat wake model is probably the next most important thing to improve. The difficulties are considerable though, and this feature will require significant development work.
- Improvement of the 2d IBL differential solver. This is interesting in terms of research but is not essential to an application like flow5. Open-sourcing this part of the code or continuing the development with other parties interested in the subject is an option. It is also something which I intend to continue investigating as a subject of personal research.
- Implementation of Mark Drela’s 8 d.o.f. model for stability analysis, instead of the 2×4 d.o.f of Etkin and Reid. The 8 d.o.f. model makes fewer assumptions than the 2×4 d.o.f. model and may be more accurate.
- Automatic differentiation for the computation of stability derivatives. This could bring improved precision and reduced analysis times to the calculation of stability derivatives. It wasn’t an option in the former version of the code but can now be considered given the refactoring performed for flow5.
- Make flow5 available on Linux using either Flatpak or Snap
- Implementation of a simplified 3d IBL model, which is a huge undertaking. This may require working in partnership with research centres.
- Implementation of compressibility corrections in the 3d analysis
- Update of the documentation, including a theory manual, a user guide, and a reference manual.
- Use of CUDA libraries to perform massive multi-threaded tasks on the GPU, with benefits expected in the construction times of influence matrices and RHS vectors
- Update of the OpenGL related methods for forward compatibility
- Implementation of a specific hydrofoil module, in case the current features of the plane analysis module prove not to be sufficient.
- Implementation of a turbine/propeller module