Urban Lung -
clay printing and CFD
Exploring the potential of passive facade
based vertical convection facilitation systems
in high-density urban area
UPenn Design Studio 2023.1-2023.5
Site: New York City, NY, United States
Instructor: Robert Stuart-Smith
Co-Instructors: Sophia O'Neill
Collaborators: Yidan Wang, Tinghui Mo, Xinlin
Lu
Contribution: Conceptual design, developing,
drawings, CFD simulation and design
optimization, robotics scripting and fabrication
As urban density increases, sustainable building
design is required to reduce energy consumption
and mitigate climate impacts. Urbanization has
led to the development of densely built-up areas
characterized by close spacing of buildings
and structures, resulting in a significant
reduction in air velocity. Building facade design
is essential to sustainable building design as
it can significantly influence the microclimate
around a building. This research proposes a
new method of ceramic facade design, which
realizes the passive ventilation based on the building facade. The
proposed method involves wind field vectors
generated by computational fluid dynamics
software and using them as influencing factors
for multi-proxy field generation of building
facades. The system helps alleviate the problem
of weakened airflow in the city. In addition to
the energy-saving benefits, the proposed method
offers opportunities for sustainable building
development. By combining sustainable design
principles with aesthetic design, the proposed
method provides a promising solution for
improving urban streets' comfort and energy
saving in a humid subtropical climate.
In the last two decades, designers in the computational design world have increasingly employed agent based models (ABMs) in architectural design and construction. The researchers introduce this concept to delve into more intricate and adaptable art forms and shapes, evolving into the refined term ‘multi-agent systems’. Drawing inspiration from Craig Reynolds Boids algorithm, the authors developed a multi agent design method. This method encodes a series of agents with tasks to design a porous facade screen for manipulating airflow. Our established principles indicate that the wind vector derived from CFD analysis plays a crucial role as a controlling factor in generating the agent field path.
Facade Generation Logic
Prototype Robotic Production Process