An adaptive 3D model for solar optimization at the urban scale

There are numerous criteria in urban planning and a same construction program can result in various geometric configurations, depending on where the emphasis is given. Real cities are even more complex, since built on different periods, with different trends or programs. Besides, it is common to see new elements flourish on buildings roofs or facades where they were not originally planned, such as elevator shafts or air conditioning blocks. The scope of this paper is to propose a way of analyzing existing urban environments, composed of thousands of elements, in order to optimize their forms on physics criteria - here the maximization of the solar potential on windows and roofs, which determine respectively the passive solar heat gains and the active solar energy through the utilization of solar panels such as building integrated photovoltaic (BIPV). Assistance in urban planning and optimization processes both requires quick calculations and a fast reproduction of the existing configuration. Parametric models appear to be an efficient way of satisfying those conditions. Besides, they permit to easily play on the parameters influencing the studied physics criteria keeping a coherent and realistic urban configuration. In this paper we present a parametric typology of housings with an adaptive level of detail which satisfies the needs of optimization and permits a precise analysis of the solar radiation at the urban scale.