Optimization of the urban building efficiency potential for mid-latitude climates using a genetic algorithm approach

Urban planning has a considerable impact on the future energy efficiency of buildings and planners lack useful tools to support their decisions. This paper is intended to contribute to fill this gap by presenting a new method based on a genetic algorithm which is able to search for optimum urban forms in mid-latitude climates (35−50°). Here, more energy efficient urban forms are defined as those which have high building absorptance in winter and low summer building absorptance. These forms can be designed by choosing among regular tri-dimensional building geometries with fixed floor space index, which can be parameterized by adjusting the following variables: number of floors, building length ratio, grid azimuth, and aspect ratio on both directions. The results obtained show that adequate urban planning, based on the consideration of the local radiation conditions as a function of latitude, may result in significantly better building thermal performance. In particular, it is concluded that the highest latitudes are more restrictive in terms of optimal solutions: pavilions (cross-sectional square blocks) are best solutions for latitude of 50° and terraces (blocks infinite in length) are preferred for 45°. For lower latitudes, all urban forms are possible. In terms of grid angle with the cardinal direction, it is concluded that the angle should stay between −15° and +15°, except for the latitude of 50° where it can ranges from −45° to +45°. For slabs and terraces urban forms, the spacing between blocks in the north–south direction should be maximized, quantified by a building-height-to-street-width (aspect) ratio which decreases with the increase of latitude, ranging from 0.6 for a latitude of 35°, to 0.4 for a latitude of 45°. For pavilions, the north–south aspect ratio is independent of latitude and should stay close to 0.7. The pavilion is the urban form which allows for a larger number of floors.