[Back]


Talks and Poster Presentations (with Proceedings-Entry):

D. Bogar, G. Rapone, A. Mahdavi, O. Saro:
"GA-Optimisation of a curtain wall facade for different orientations and climates";
Talk: BSA2013 - Building Simulation Applications 1st IBPSA Italy Conference, Bozen, Italein; 2013-01-30 - 2013-02-01; in: "BSA2013 - Building Simulation Applications 1st IBPSA Italy Conference", IBPSA Italy (ed.); Eigenverlag / wissenschaftliches Kommittee der IBPSA Italy, 1/1/1 (2013), 3821 - 3828.



English abstract:
The importance of considering the environmental
performance of buildings since the conceptual stages
of the design process is growing as a consequence of
the restrictive requirements of building regulations
and energy certification. The building envelope plays
a central role in controlling the thermal, the acoustic
and the light flows exchanged between the indoor
and the external environment. Therefore designing a
good façade system and installing it properly is
fundamental to increase the overall performance of
the building.
This study focused on a modular curtain wall façade
that can integrate photovoltaic panels on its opaque
areas. The aim was to identify the layouts that
minimise total cost, calculated as the sum of
investment and operation costs, and at the same time
maximise daylighting in the rooms. The methodology
consisted in a simulation based multi-objective
optimisation solved by means of Genetic Algorithms,
whose behaviour is inspired by Darwin´s Theory of
Evolution.
An office room was modelled and its façade divided
into twelve modules of equal size, two of which were
fixed glazed elements and the remaining ten having
three possible construction types: glazing, spandrel or
photovoltaic elements. The optimisation process
assigned a specific construction to each of the ten
modules in order to obtain a set of façade layouts that
fulfill both objective functions. Different climatic
scenarios and orientations were considered.
The dynamic thermal simulations were carried out
with EnergyPlus while the optimisation employed a
modified version of the widely used NSGA-II
Genetic Algorithm. A custom-made software was
written in Matlab in order to interface the simulation
and optimisation processes, to automate their
interaction and to make the input of the problem´s
parameters easier through a graphical user interface.

German abstract:
see english version