Parameterization of built environment for wind resource assessment


Thesis Type: Postgraduate

Institution Of The Thesis: Middle East Technical University, Faculty of Architecture, Department of Architecture, Turkey

Approval Date: 2018

Thesis Language: English

Student: ŞAFAK SAKÇAK

Principal Supervisor (For Co-Supervisor Theses): Ali Murat Tanyer

Co-Supervisor: Mehmet Koray Pekeriçli

Abstract:

Use of wind power for energy generation in urban areas is gaining importance considering the energy consumption and environmental impact behaviors of built environments. Therefore, assessment of urban wind resource is crucial. The study of air flows in cities is also related to other concerns as natural ventilation, air contamination and pedestrian comfort. Two of primary research topics proposed in feasibility studies for urban wind energy are improvement on the understanding of urban wind flows with developed methods; also, an enhanced knowledge on the interaction of urban geometry with air movements. Related literature points out that there is a need for better communication between disciplines concerned with research and design. As an interface which help both sides to configure their studies, algorithm aided design tools may offer opportunities. The aim of this study is to propose a framework for parameterization of built forms for wind related studies with the use of parametric model algorithms by visual programming interfaces. Thus, a faster and diversified production of sample models can be provided by changing values of geometric variables. An example model within this framework was developed. To achieve this, a literature survey was conducted with a focus on extracting and aggregating possible shape attributes, architectural form and urban formation parameters. UML diagrams were used before translating parameters into the interface of Dynamo software. Simple flow simulations made on samples derived from parametric models. By comparing results with parameter sensitivity analysis, influences of the parameters and relevancy of the example model were observed.