Aristotle University Of Thessaloniki (AUTH)
Study and modelling of flow phenomena in urban street canyons
Abstract
dc:descriptionAir pollution is a major health hazard for the people that live in cities. Street-scale Air Quality Models (AQMs) are a cheap and efficient way to study it and provide solutions. AQMs employ empirical parameterisations to include all the complex phenomena of wind flow between buildings. The modelling of the recirculating flows inside street canyons comprises the dominant parameter for the accuracy of the results. This work aims to propose an explicit or implicit description of the recirculation zone. This could improve the ability of street scale AQMs, to predict the concentration of airborne pollutants inside street canyons. To achieve this, a computational fluid dynamics (CFD) model is developed and the results obtained are analyzed with Machine Learning techniques. In the first Chapter the scale of the air pollution problem is presented and the main parameters of the wind flow in street canyons are reviewed. Furthermore, there is a brief discussion on the core assumptions of the most well-known AQMs regarding the recirculation zone and the thesis objectives are outlined. In the second Chapter, the most important flow parameters in street canyons are reviewed and the current approaches of street scale AQMs are discussed. It is highlighted that AQMs divide the street canyon in regions with different flow characteristics. The assumptions behind such divisions originate mainly from experimental studies. After the demonstration of the problem’s complexity, a relatively simple case of perpendicular to the street wind flow is selected for further study. In Chapter 3, the details of the employed numerical methods are presented. These include the Large Eddy Simulation methodology, the respective CFD model in OpenFOAM open source code and the theory behind k-means clustering. In Chapter 4 the extended validation procedure is described. The current model’s results are compared with experimental and CFD data from the literature, presenting satisfactory agreement. In Chapter 5, the predicted wind velocity and pollutant concentration results for the selected geometries are analyzed using the previously described in Chapter 3 list of exploratory techniques. The flow pattern differences for the studied building height to street width ratios are shown, while the importance of source location is demonstrated for the wide canyons. The four expressions which can describe four regions with common flow characteristics inside the canyon, derive from the results obtained from the k-means clustering method. These expressions are strongly dependent on the street canyon dimensions and are also independent of the wind velocity. This constitutes an implicit definition of the recirculation zone, through the parametric description of the velocity field and can be expanded to cover more parameters, such as oblique wind direction. In Chapter 6, the main conclusions from the present thesis are outlined and the physical meaning of the proposed four regions is discussed. Moreover, the future directions and the limitations of the proposed numerical methodology are described. Finally, in the Appendices, there is information for the adopted modifications of OpenFOAM solver and the developed Python scripts for the preprocessing of the LES results for the k-means clustering algorithm.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Chatzimichailidis, Arsenios
- Χατζημιχαηλίδης, Αρσένιος
Subjects
dc:subject × 16- Αστική ποιότητα αέρα
- Υπολογιστική ρευστοδυναμική
- Αστική χαράδρα
- Αλγόριθμοι ομαδοποίησης
- Urban air quality
- Computational fluid dynamics
- Street canyon
- Clustering algorithms
- Φυσικές Επιστήμες
- Γεωεπιστήμες και Επιστήμες Περιβάλλοντος
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Χημικού Μηχανικού
- Natural Sciences
- Earth and Related Environmental Sciences
- Engineering and Technology
- Chemical Engineering
Rights
dc:rights- Statement dc:rights
-
- BY
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/47368