Chris

and 1 more

-Plan definition of the Numerical Flow Simulation project-This project deals with a numerical flow simulation of waves around a Wigley Hull profile. The first step in our project will be to identify clearly the goals and objectives of our study. For this purpose, we will first meet and discuss about the different tasks that have to be undergone in this study. We will then specify the time needed to perform the individual tasks and summarize it in a Gantt chart which will serve as a timeline to our project. We will split up the workload between the two of us and set milestones in order to achieve the goals set here in this project definition phase.First, we will present the hydrodynamical problem and define the limitations of our study. To succeed in this task, we have to calculate several dimensionless numbers that characterize our flow, e.g. the Froude number and define the range investigated in our simulation. Afterwards, we decide on which assumptions can be met to simplify the problem without losing accuracy in our solution, e.g. steady flow assumption. Moreover, we must know which physical values we want to compute during this simulation (velocity field, pressure field) and which ones to deduce from the results (wave height, drag coefficient) that can later be compared to previous simulations and experiments.Meanwhile, we want to accomplish a literature research to be able to compare our results to the state of the art numerical of flow simulations in this specific problem. This data depends on the dimensionless numbers chosen as well as the numerical approach. After making a short summary of each report, we will focus on these different numerical approaches to find the one that is best suited for our study. Moreover, a part of our research will be dedicated to the Volume of Fluids (VOF) method in order to understand the numerical principles and come to grips with how it is best used in our simulation.Afterwards, we shall define clearly the different steps within our study (pre-processing, computation, post-processing and verification/improvement) and identify the tasks that have to be performed. For the pre-processing step, we will just have to import the geometry of the hull and then define the domain by enclosing the hull to create the fluid volume. After that, we will use the  symmetry of the problem in order to simulate only half of the structure and therefore accelerate the simulation.Then, we will generate the mesh: at the beginning, we will choose a coarse mesh with a small number of cells and improve this mesh step by step by refining it in the areas of high gradients. We favour hexahedral cells to get a structured mesh, but we might have to add a thetrahedral mesh in areas where the geometry does not allow a structured mesh.For the computation to work, we will choose a physical model and select which equations we want to solve and how we discretise them: the VOF method seems to be adapted for “Open Channel Flow” which we are likely to choose. After that, we will have to define our boundary conditions (velocity inlet; pressure outlet). Furthermore, we will choose a numerical turbulence model for our simulation. We will use the RANS approach that appears to be the best economical method in terms of resources and select a turbulence model adapted to our problem. In our case, we will compare two of them: the k-epsilon model and the Sparlart-Allmaras model. Finally, we will select an adapted wall treatment (wall function approach or near-wall model approach) and a beach treatment. All these choices are dependent on our previous researches concerning this simulation problem and could be changed or improved.During the computation, we need to be attentive in checking the convergence of our solution, e.g. continuity, residuals.In the post-processing phase, we will visualize our results for the 2 cases (corresponding to the 2 different Froude numbers) and compare it to the data initially found.We will calculate the drag coefficient and compare it to the experimental data.Moreover, a step of improvement will be necessary to make our simulation more accurate, robust and faster by modifying the mesh and/or changing the methods (numerical approach, order schemas, Courant number). Lastly, the verification phase allows us to check the relevance of our final simulation.In this fourth and final part, we will write a scientific report: in the initial one, we will describe the physical problem with the governing equations and the different steps and choices made for our simulation with justifications.We will write the final scientific report at the end of the project and we will draw the conclusions from this project which will include discussing our results and the differences with regards to the numerical and experimental data found previously in our literature research. We will talk about further improvements and future investigations that can be carried out and conclude with the management of this group project by creating a timetable and comparing it to our initial Gantt chart.