Topic outline

  • Simulation project

    General

    You are free to choose the project topic. However, the topic must be such that similar simulations can be found from journal or conference papers. You should also note that the focus on the course is on fluid flow modelling. Topics which include e.g. chemical reactions or complex thermodynamics are beyond the scope of the course. If you still want to do a project on such a topic, you have to find a named instructor for the project.

    Think about the topic from the point of view of

    • the expert profile that you are developing through your studies
    • possible master's thesis project, as the course gives you an opportunity to do a pre-study related to a master's thesis (do note that work, which will be assessed as part of your thesis, cannot be included in the course project, as you would get credits twice for the same work)

    The research groups might also have some proposals for project topics. If I get such proposals, I will let you know about these. You can also be active and ask the staff at the university for possible topics.

    The project can be done individually or in a pair. I expect the projects done in a pair to be more ambitious than individual projects. You should start thinking about the topic of the project and the research question as soon as possible. When your project idea starts to be more clear, send me an email in which you specify the topic, the research question, your motivation for the project, the scope of the simulations, simplifications of the problem and links to references with similar simulations. In case of a project in a pair include the planned division of work as well. For these projects it is sufficient that just one member of the pair send me an email, but put the other member of the pair in the cc field. If I consider the project to be too challenging, we will discuss, how the project could be simplified, so that it would still be interesting to you, but more manageable considering that the course is only 5 ECTS points. The topic should be confirmed by the 21st of September. Do not prolong this process. I am happy to discuss any ideas you have at any point.

    If you would like to do your project in a pair, but do not have a colleague yet, you can advertise your project ideas in the general forum on the course pages.

    Scope of research

    You should choose a research question that you are personally interested in. Your goal is to answer this research question using computational fluid dynamics as the tool. The research question can be of physical nature (e.g. influence of some physical parameter on the flow field or some derived quantity) or more focused on CFD methodology. For the latter some relevant options could be the influence of one or several of the following on the simulation results

    • boundary conditions
    • turbulence models
    • discretisation schemes
    • local features of the mesh

    However, even in this case, the simulation project should not be too trivial as one of the main goals of the course is to train your ability to do simulations in more practical cases. Regardless of your choice of the research question you have to perform a grid dependency study to estimate the uncertainty of your numerical prediction. Note also that unsteady problems are significantly more time consuming and require significantly more storage capacity than steady problems.

    High-performance computing option (HPC)

    I will offer to do a limited amount of simulations on the super-cluster of CSC. The idea is that if your resources are limited and prevent you to do e.g. a proper grid dependence study, I can perform simulation for a higher resolution. This is based upon request. In the request you have to justify the need for additional resources and specify an estimate of the required resources. I will base the assessment of the need on the complexity of your case. If the need is justified, you will need to provide a tested and working case via the Funet Filesender service. I will not check the case, but I will simply submit it to the queue and return the results to you. The important dates for the HPC simulations are the following (these are subject to the availability and load of the CSC services).

    26.10.Deadline for the resource application
    5.11.Deadline for the case files
    13.11.Simulation results available for analysis

◄ Previous section
Tutorial
Next section ►
Working diary