Multi-objective optimization of a residential zone by proposing appropriate comfort factors using none dominated sorting genetic algorithm

Abstract

To identify optimal solutions within a plethora of alternatives with respect to architectural design, multi-objective optimization has been widely acknowledged as a dependable approach. Nevertheless, such optimizations have frequently been conducted as single-objective analyses or have failed to consider both the quantity and uniformity of illumination concurrently with energy consumption as primary objectives. This study posits a proposal consisting of two metrics, one pertaining to the quantity of light and the other to its uniformity. These metrics are derived from the useful daylight illuminance (UDI) and are deemed appropriate for employment in an optimization process. In conjunction with energy consumption, these metrics serve as objectives in a multi-objective optimization that is carried out with a NSGA-II algorithm specifically developed and calibrated for this particular problem. The result of the simulation yields a solution set in three dimensions. The interdependence between the variants and their relationship with each variable were thoroughly examined to facilitate the selection of the optimal solution from the pareto front.