University of Cambridge > > Sustainability in the Built Environment (GreenBRIDGE) > Computational Building Performance Analysis

Computational Building Performance Analysis

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Aaron Gillich.

Research from the UCL Bartlett School of Graduate Studies and Energy Institute.

1) Tia Kansara (UCL Energy Institute) The Energy Code in Abu Dhabi – ‘In 2011 the Department of Municipal Affairs, Abu Dhabi, developed the Energy Code in line with the ICC - International Code Council. This year it will be launched and there is a considerable debate around the topics that the Code discusses. Energy is a growing area of concern as the Emirates continue to increase the built environment attracting more people to live in an energy-intensive environment. What will happen? What are the forecasts for energy supply and demand? This talk will outline the discussion around the Energy Sector in Abu Dhabi.’

2) Greig Paterson (UCL Bartlett School of Graduate Studies, Complex Built Environment Systems Group) Utilising Monitored Data to Gain Environmental Feedback in Real-time as Early Design and Briefing Decisions are Made Building simulation is often rejected by architects at the early design stages. Furthermore, studies have shown that the predicted energy performance of buildings is often lower than the actual performance once built. In view of this, this research explores machine learning techniques, utilising monitored and simulated data, to form a platform for performance prediction. Using school design as a test case, the aim is to develop a design tool which allows environmental performance indicators to be communicated to the user in real-time as early design parameters are altered interactively – allowing the architect to sketch performance as well as form.

3) Samuel Wilkinson (UCL Bartlett School of Graduate Studies, Complex Built Environment Systems Group) Towards Machine Learning for Environmental Performance Prediction in Generative Design: Estimating Tall Building Surface Pressure. The trend towards creating ever taller buildings continues, with dramatic technological improvements in materials, structures, and modelling. However, as height increases so too do gravitational, wind and seismic forces. Wind forces on tall buildings must be mitigated to avoid occupant discomfort from swaying, reduce risk of facade damage, and improve structure efficiency. Although these problems can generally be avoided through early decision form-finding, current fluid simulations are costly and do not provide rapid performance feedback necessary for such exploratory or optimization studies. Here an approach is described that uses machine learning to make predictions of the wind loads by recognizing shape features, through the use of procedural geometry generation and regression analysis. Results will be shown whereby wind loads can be immediately predicted and visualized on models of arbitrary complexity.

4) Kinda Al-Sayed (UCL Bartlett School of Graduate Studies, Space Group): Natural Growth Processes in Urban Form – “Cities appear to exhibit autonomous growth behaviour; they appear to imitate natural growth in response to large-scale human interventions. In Barcelona, the uniform grid subdivides in response to the increase in global accessibility of street network. In Manhattan, localized processes lead to the splitting and multiplication of elongated patches reinforcing regular patterns similar to those in reaction-diffusion models. The generic mechanisms that are extracted from historical transformations indicate that urban form has some inherent natural processes that build its complexity from the micro scale to the macro scale. Understanding the dynamics of this self-organised complexity will have significant implications on urban design.”

This talk is part of the Sustainability in the Built Environment (GreenBRIDGE) series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2024, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity