Thesis Defence: Saeran Vasanthakumar

Monday, July 6, 2015 10:00 am - 10:00 am EDT (GMT -04:00)
Of the thesis entitled: A Computational Design System for Environmentally Responsive Urban Design

Abstract:

This thesis introduces a computational design tool that attempts to coordinate urban energy transfers and needs by iteratively organizing, prototyping, and then evaluating the performance of different typology solutions. Developing low-energy, high-density urban typology is a critical goal for cities given current energy consumption and urban growth trajectories. This target is contradicted in part by the increase of operational building energy due to the microclimatic conditions and increased structural and mechanical inputs required by dense urban typologies. Studies have shown that the energy impact of urban typology design is significant and justifies auditing and coordinating building energy requirements in urban neighborhoods.

Despite this, current urban energy modelling tools do not account for the consequences of different typology choices, and urban modeling tools do not integrate state-of-the-art environmental and energy simulation methods. Recent advances in computational tools can be used to efficiently generate a solution space of potential typologies to fill this gap in current urban design and analysis software. As such, the broader goal of this research area is to develop a design system that derives high density urban fabric according to a nuanced simulation of urban energy demand.

Of the multiple energy reduction strategies available, daylighting offers significant opportunity for architectural optimization because it varies greatly, even at relatively high densities, due to the effects of ambient light, surface reflectance and building geometry. In conjunction with the decreasing contribution of heating demand in the overall building energy budget, this indicates that gains in urban energy efficiency today can be made by focusing on reducing lighting energy demand. Therefore the current goal is to develop a proof-of-concept that encodes and traverses an urban design solution space to increase the daylighting potential of built typology while achieving target density goals.

The proof-of-concept will consist of a parametric grammar-based form generator that is extended with existing software or algorithmic models to achieve the current goal. Specifically, the tool consists of three parts: a model of complex urban dynamics to derive density targets, a generative rule set to encode building typology, and a performance simulator to derive solar zoning envelopes and interior illuminance metrics. Daylighting metrics and material simulation is achieved with the RADIANCE/DAYSIM modeller. Existing urban modelling algorithms will be translated within the shape grammar-based system to map the dynamics of non-uniform urban densities.

The thesis design system integrates research from two domains through computational methods: urban modelling and building performance simulation. The synthesis of this existing research and work thus puts forward a model of integrated city design via generative design systems. The contribution of the synthesis lies in the development of the urban energy-centric form generator, which extends procedural type generation to simulated environmental and material data. The proof-of-concept is licensed under the open-source GNU General Public License, and packaged as an Python-based plug-in for Grasshopper3D, the visual scripting interface for the Rhinoceros3D CAD modeller. 

The examining committee is as follows:

Supervisor:

Committee Members:

Philip Beesley, University of Waterloo

Maya Przybylski,University of Waterloo
Kevin Stelzer, B+H Architects

External Reader:

José Duarte



The committee has been approved as authorized by the Graduate Studies Committee.


The Defence Examination will take place:  

Monday July 6, 2015
10:00AM

Lawrence Cummings Lecture Theatre - ARC 1001

A copy of the thesis is available for perusal in ARC 2106A.