[bigfootedrockmidget]

The SU2 Foundation is applying for Google Summer of Code 2024! From all the possible projects, we have made a selection of 5 topics that could be used in a GSoC project. The document can be viewed here:


https://docs.google.com/document/d/1...it?usp=sharing


edit: please also see the other post on How To Apply:
GSoC Applications - How to Apply

[bigfootedrockmidget]

For completeness, the 5 projects that were selected are given below:



Project 1: Differentiated Meshing for CFD-based Design

• Project Description (max. 5 Sentences)

The objective is to write a code to generate a discretized mesh which has gradient information for one of the response surfaces. To achieve the goals of this assignment three main tasks are:

• Develop meshing code /algorithm and differentiate it using Python packages (for example JAX).
• Couple the mesh code with the open-source CFD code SU2.
• Run optimization to benchmark the new method against the old method of deforming the grid.

• Expected Outcome (deliverables): Python extension
• Skills Required: Python3 (Numpy, Jax), Computational Methods, C++ (Basic), SU2
• Possible Mentors: Nitish Anand (lead)
• Expected Project Size: 175 hrs/medium
• Difficulty rating: medium-hard (some experience/background)

Project 2: Extending Python Wrapper with Tabulated Thermodynamics and Chemistry

• Project Description (max. 5 Sentences)

Objective: enable thermochemical data interpolation for data-driven fluid models. This interface shall have the following features:
1: The reference data is interpolated through a (structured) look-up table, presented through the python wrapper of SU2.
2: The table format and look-up feature shall support an arbitrary number of table dimensions.
3: Emphasis lies on high interpolation accuracy with robust and efficient performance in terms of memory usage and query time.

• Expected Outcome (deliverables): C++ code and python code to enable python access to tabulated thermodynamics and chemistry. Example code with testcase (provided by mentor) to demonstrate functionality. Documentation and tutorials.
• Skills Required: Python, C++, python, SU2
• Possible Mentors: Evert Bunschoten (lead)
• Expected Project Size: 175 hrs/medium
• Difficulty rating: easy-medium

Project 3: Graphical User Interface: Core functionality and user friendly installer

• Project Description (max. 5 Sentences)

We started writing a GUI for the SU2 solver to help new/inexperienced users with setting up CFD problems in SU2. Most core functionality is available but the installation needs to be easy to make it truly user friendly. Finalizing core functionality regarding initialization of problems and visualization of results will be the focus.

• Expected Outcome (deliverables): SU2-GUI (python+trame library), availability on github, installation through an easy installation process
• Skills Required: Python, Paraview, SU2, Trame
• Possible Mentors: Nijso Beishuizen (lead)
• Expected Project Size: 175 hrs/medium
• Difficulty rating: easy-medium

Project 4: Python wrapper extension and documentation of user defined boundary conditions

• Project Description (max. 5 Sentences)

The python wrapper can be used to modify some but not all boundary conditions and properties. The goal of this project is to extend the capabilities of the python wrapper and be able to modify all transported variables on all boundary types. Additionally, user documentation and tutorials have to be written.

• Expected Outcome (deliverables) modified c++ and python code to enable spatially and temporally varying boundary conditions, tutorial, user documentation
• Skills Required: C++, python
• Possible Mentors: Nijso Beishuizen (lead)
• Expected Project Size (90 hrs/ small , 175 hrs/medium, 350 hrs/large): 175hrs
• Difficulty rating (easy (little experience/background), medium (some experience/background), hard (experienced)): easy-medium

Project 5: Feasibility of GPU acceleration in SU2

• Project Description (max. 5 Sentences)

The SU2 code relies heavily on matrix-based algebra. In this area, there is significant speed-up potential with the adoption of GPU-based processing. The goal of this project is to explore a straightforward adaptation of the existing SU2 code using NVBLAS, verify the potential speed-up in existing test cases, identify bottlenecks, and explore the potential for future work.

• Expected Outcome (deliverables) A NVBLAS-enabled SU2 version.
• Skills Required C++
• Possible Mentors Leonardo Cavanha (lead)
• Expected Project Size 175 hrs
• Difficulty rating easy / medium

Other ideas can also be supported if you have a novel idea that matches your specific knowledge and interests (improving specific turbulence models for instance), but you need to find a mentor to supervise the project.