Geomatics Day - Student project 3

Finalising Geometry preparation for numerical simulations

Computational Fluid Dynamics (CFD) simulations are valuable for modelling complex urban phenomena such as wind flow and microclimates. A CFD requires a 3D geometric dataset that represents objects in the urban environment which are most commonly buildings.

When creating geometries automatically for CFD simulations, several clean up tasks must be completed for them to be usable without any issues. One of the problems arising is related to the redundant faces shared between adjacent buildings, which have no purpose for outdoor flow simulations and cause complications when running the CFD. This synthesis project focuses on addressing the aforementioned issues by removing the shared faces and, therefore, improving CFD results.

Our approach will be to search and evaluate the available literature and identify ideas and algorithms that can help us tackle the problem. If relevant code is available, we will try to test and adapt it according to our needs. Moreover, we aim to develop our own method, based on which an open-source implementation will be carried out. Different approaches will be investigated to achieve removing these faces, while conducting a comparative study on the advantages and disadvantages between them.

Our dataset will be taken from the 3D BAG, which is an up-to-date data set containing 3D building models of the Netherlands, often containing semantic characteristics. We will select diverse subsets of the 3D BAG with varying building configurations since we aspire the algorithm to be robust. We will then run validity tests on these datasets before the implementation to guarantee the quality of the input, in terms of geometrical and topological correctness. After running the code, we will run the validity tests again to ensure that we have not compromised the validity. As a final step, we will evaluate the results to determine the percentage of faces that were successfully removed.

The current project is carried out in cooperation with Dassault Systèmes. The ultimate goal is to create an open-source product that can solve these problems and be usable in many different software for developing CFDs. The benefits will be imminent, as we strive for higher quality simulations, improved computation time, the use of less resources, and an overall improved user experience to better replicate the urban environment.

This project is commenced and supervised by: Dassault Systemes