Urbanization has led to more than half of the world’s population living in cities. The design of sustainable and resilient urban environments is becoming increasingly critical to optimize the well-being of their inhabitants and the planet. One of the major challenges in achieving this is the complex wind flow patterns in densely built-up areas, which require accurate prediction and analysis to effectively harness the potential benefits of wind flow, such as natural ventilation and wind power generation. However, the unique features of urban landscapes, such as high-rise buildings, narrow streets, and irregular building shapes, make the analysis of wind flow within urban canopies complicated.

In recent years, Computational Fluid Dynamics (CFD) has become a vital tool for studying wind flow in urban areas. Still, the complex geometries of buildings can lead to challenges, including recirculation, reattachment, intense turbulence, and dead zones. Moreover, vegetation plays a crucial role in controlling wind flow in dense areas by acting as a physical barrier, significantly reducing the wind speed and alter the wind flow direction. Additionally, generating geometry for computa-tional fluid dynamics (CFD) simulation in complex urban environments is a chal-lenging and time-consuming process.

Hence, for this thesis, the City4CFD software will be used to automatically recon-struct a 3D model of Stanford University at LoD1.2, which will significantly reduce the time and effort required to generate the complex geometry necessary for com-putational fluid dynamics (CFD) simulations in urban environments. The results will be compared to those obtained from an already manually reconstructed model at LoD2.1 and real-world measurements conducted within the area of interest. This will allow me to determine the differences introduced by different level of detail.

The research will address several sub-questions, such as the steps needed to auto-matically reconstruct a 3D city model, the potential improvements in simulation accuracy by increasing LoD, and the impact of complex geometries on wind flow. The results of the thesis indicated that a more complex geometry Level of Detail (LoD) can enhance the accuracy of simulations by providing a more precise de-piction of wind flow patterns. In other words, a higher LoD geometry, such as LoD2.1, can more accurately predict wind patterns in urban environments based on real-world measurements. The study observed that the LoD2.1 model, which in-corporates more complex features, generated simulation outcomes that were closer to the measurements compared to the less detailed LoD1.2 model.

Currently more than 4 billion people live in urban areas around the globe, a trend that is expected to be increased in the upcoming years. While urbanisation provides the space for innovation and new opportunities, in the meantime physical, technical and social challenges are rising and the cities’ vulnerability is increasing. A tool to tackle these issues are Computational Fluid Dynamics
(CFD) simulations, which can provide insight in various topics.

CFD simulations are valuable for modelling complex urban phenomena such as wind flow, microclimates and thermal comfort. A CFD requires as an input a 3D geometric dataset that represents objects in the urban environment which are most commonly buildings and then according to this input the air flow is simulated around it.

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 creating the mesh that is needed for the CFD. This
synthesis project focuses on addressing the aforementioned issue by removing the shared faces.

The ultimate goal of this project was to create an open-source product that can efficiently and in an automated way remove the adjacent faces between buildings. The benefits will be imminent during the meshing process, as we strive to reduce the time that consultancies spend fixing the input geometries before running a CFD simulation, along with an overall improved user experience.

This report is organised in four main sections. The first section is the general introduction of the issue that needs to resolved. The second section defines more in depth the problem and sets the research questions, in accordance to that, in the third section the research methodology is developed. In the fourth section the results of both methods are presented. The fifth sectionfocuses on a reflection of the project, while the sixth section presents the final conclusions. Finally, the seventh section contains the specifics of the project management itself.

The project was carried out in cooperation with Dassault Syst`emes and is developed in the context of the GEO1101 course in MSc Geomatics TU Delft. In addition to this report we have created a GitHub repository (https://github.com/Fabisser/facesBgone) that contains the source code of the two methods.

Planting trees is widely considered an effective way to create a good urban wind environment, improve air quality, mitigate heat island effects, improve pedestrian wind comfort and reduce building energy consumption. To assess tree effects and find suitable tree setups in urban areas, Computational Fluid Dynamic (CFD) simulations can be used.
To handle trees in CFD simulations, the implicit tree modeling approach, i.e, the porosity model, is widely used where finite volume cells that roughly account for trees are marked as porous zones. Some studies have also attempted to model trees as obstacles rather than porous zones, which can be referred to as an explicit tree modeling approach. The difference between these two approaches deserves further study. Also, for practical purposes and lack of information, the geometric features of trees are usually oversimplified or even ignored in CFD simulations.
This thesis investigates the difference between implicit and explicit tree modeling approaches and analyzes the impact of tree Level of Detail (LoD) and shapes on the flow structure. For comparative analysis, several numerical test cases with different urban complexities, tree modeling approaches, tree LoDs, tree shapes, Leaf Area Density (LAD) values, and wind directions were used for CFD simulations.
The results show: (a) the implicit models always allow some of the wind flow into the porous cells no matter how high the LAD values are, resulting in smaller wind acceleration on the lateral sides of implicit tree models; (b) for the idealized street canyon and realistic urban geometry test cases simulated in this thesis, the velocity magnitude differences between the LoD2 cases and the LoD3 cases are rather limited, with maximum differences in the order of 0.5 m/s; (c) differences in tree shapes, LAD values, and wind directions will change the effects of tree modeling approaches and tree LoDs on wind. For instance, the case using an isolated explicit LoD2 conifer tree model has a different wake flow pattern from other explicit cases. Also, with the inflow direction perpendicular to buildings, the higher the LAD values, the larger the velocity magnitude difference between cases using LoD2 tree models and those using LoD3 tree models.

In the Netherlands, the coastal dunes are essential to protect the country against flooding. However, the rising sea levels increase the risk of flooding along these sandy shores. Moreover, due to a combination of human and natural activities, dune erosion has increased and will continue to do so in the coming decades. Besides flood protection, the Dutch shoreline is important for preserving biodiversity, the generation of drinking water and recreation. In recent year, the number of recreational buildings on the beach, such as bars and holiday homes has increased. This is relevant because previous studies show that such beach buildings affect the wind flow and limit aeolian transport of sediment towards the dunes.

This research studies the impact of beach house configurations on dune-ward sediment transport to limit the adverse effects on dune development. We use CFD simulations to study the wind flow around a 3D model of a beach with holiday houses, based on a section of the Noordwijk beach in the Netherlands. We implement the CFD software OpenFOAM to solve the RANS equations for turbulent, steady-state flow. The sediment transport that occurs is calculated using the wind direction and speed near the ground surface of the solution.

The study consists of multiple 3D models in which the placement of the houses is varied systematically, to study the effects of beach house configurations. Variations are made by rotating houses, individually or within a row, and changing the distance between the houses and dunes. We determine the annual effect on sediment transport by applying varying wind conditions based on historical wind data from Noordwijk.
As we have many simulations to run, and all need different parameters and settings, we automated the process. First, a PostgreSQL database is used to store all requirements for the CFD simulations, the metadata and the results the simulations give. Then, a Python script links the information stored in the database to the correct settings in OpenFOAM. This way, many simulations are run in a row, and the results area to compare.

The results show that rotating the houses individually towards the prevailing wind direction appears to improve the amount of dune-ward sediment that takes place, compared to beach houses placed perpendicular to the shore. Rotating a row of houses as a whole has a limited effect on the amount of sediment transport. However, combining the rotation of the row of houses and the houses individually towards the prevailing wind direction shows the best improvement in sediment transport. Changing the distance between the houses and the dune foot so that a row of houses forms a funnel shape pointing towards the dunes also yields promising results.
Because we use a simplified model and do not take factors such as moisture levels or fetch distance into account, the results of this study overestimate the amount of sediment transport that takes place and might not quite resemble the reality. Further research using scale models or wind tunnels is necessary to confirm the suggestions made in this thesis.

In 2017, the ‘Kustpact’, a Dutch national agreement, was signed to develop the Dutch coast without damaging its ecosystem nor its aesthetics. Consequently, the formation and protection of coastal dunes have become a focal point for research and development. For this reason, based on the idea of Performance-Based Design (PBD), this Master thesis aims to investigate the impact of certain geometric design parameters of beach house configurations on wind turbulence for the purpose of widening the dunes.

A parametric design space was set up to easily automate Computational Fluid Dynamics (CFD) simulations for different purposes throughout this study. The sediment mobility estimation was the objective function for all simulations conducted. It is based on the difference of transport rate between two lines from the beginning of the plateau to the foot of the dunes. To simplify the parametric model, a Sensitivity Analysis (SA) was conducted to statistically identify the most influential parameters in the model. Afterward, the filtered parameters from the SA are used to build a surrogate model to detect the trends that result from combining the most influential parameters. Finally, a Surrogate-Based Optimisation (SBO) is conducted based on the former surrogate and using a Genetic Algorithm (GA). Three generations were run and used to increase the accuracy of the surrogate and predict an optimum solution for widening the dunes.

The study concludes on a list of design criteria to respect in order to widen the dunes mainly pertaining to the consistent position of the configuration relative to the dunes, the important overlap between houses and the larger wind-facing direction of the configuration.