University of Twente Student Theses


D-HYDRO flood simulations for waterboard Noorderzijlvest

Hop, F.J. (2021) D-HYDRO flood simulations for waterboard Noorderzijlvest.

[img] PDF
Abstract:Within this research, 2D flood simulations in D-HYDRO have been performed and analysed. This is done to give an advice to the waterboard on how to use 2D flood simulations in D-HYDRO. It is important that these simulations can be set up quickly, since flood simulations are needed urgently when the water level reaches alarming levels. D-HYDRO is a new software package being developed by Deltares. The D-FLOW FM module of the software package allows for the 1D, 2D or 3D simulations of water flow. D-HYDRO simulations are compared to each other within a sensitivity analysis, to analyse how different modifications to the model affect model output and computation time. Also, D-HYDRO and TYGRON simulations have been compared, in order to see how different D-HYDRO models perform compared to TYGRON. The sensitivity analysis resulted in an advice for how different model components should be used when making a flood simulation. The tested model components are grid size, flexible mesh, culverts, roughness values, precipitation, wind, dams, model size, initial water level, and breach inflow. Within this report, for each of these components, an advice is given on how to use them within flood models. This advice is based on both the effect on model output, computation time and model set up time. The comparison between D-HYDRO and TYGRON has shown similar inundation patterns for both software packages. However, there are local differences in flood propagation between the simulations that are performed. These differences can mainly be related back to the known differences between the simulations. The most important difference being that TYGRON uses a fine grid with grid cell sizes of 1*1 meter, while D-HYDRO used 10*10 meter grid cells. Despite this big difference, the results were still similar for the cases that have been tested. Even when performing simulations with even bigger grid cells (up to 100*100 meter), D-HYDROhas shown promising results. With the large grid cells, the computation shortens significantly. A simulation that takes 71 hours at 10*10 meters, takes only 2 minutes at 100*100 meters. This comes with a loss in accuracy, but the overall inundated area is similar for both simulations. From this research it can be concluded that 2D D-HYDRO flood simulations are promising for the waterboard. When using large grid cells, results can be gathered quickly. Flexible mesh allows important areas to have a more detailed grid, and thus a more detailed simulations, while less important areas have less details. This has proven to be a good tool to remain accurate while saving on computation time. Furthermore, models can be set up with a small amount of data, and additional data can be used for additional accuracy. This can all be done within an intuitive user interface, that is relatively easy to learn. With some training of employees and modifications to the database, the waterboard can use D-HYDRO for 2D flood simulations.
Item Type:Essay (Bachelor)
Faculty:ET: Engineering Technology
Programme:Civil Engineering BSc (56952)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page