Geometric optimization of fenestration

Fenestration and its design have a significant impact on the energy use associated with the artificial lighting, heating and cooling of a building. To date, research into window optimization has been for windows that are constrained to have a regular geometric shape and position. This paper describes an approach in which a building façade is divided into a number of cells, each cell having one of two possible states, a solid wall construction, or a window. A Genetic Algorithm search method has been used to optimize the state of each cell, with the number or aspect ratio of the windows being constrained were desirable. It can be concluded that the approach results in design solutions that have interesting and innovative architectural forms and which minimize building energy use.