Surrogate Modeling for Sustainable Building Design

Onur Dursun, Berk Ekici

Sustainable design of buildings requires a considerable amount of intensive computations, in order to derive accurate figures on building energy consumption and comfort. Research on the use of approximation models, or surrogate models as they came to be known, focuses on the development of suitable machine-learning methods, to approximate to a high degree the output of costly simulation models, at a fraction of the computation time.

Through this benefit, the extensive application of computational decision support tools, such as optimization is facilitated.


Chatzikonstantinou, I. Sariyildiz, S,, "Approximation of Visual Comfort Indicators in Office Spaces: A Comparison Study in Machine Learning", Architectural Science Review, , 2015
Dursun, O., Aydin, E., Chatzikonstantinou, I., Ekici, B., "Optimisation of Energy Consumption and Daylighting using Building Performance Surrogate Model", Architectural Science Association 2015 Conference, Melbourne, Australia, 2015
Note: You are required to cite the above reference when using any of the information on this page; please make sure you inform the cited author of your publication.