Louver configuration comparison in three Canadian cities utilizing NSGA-II

Abstract

Shading devices, particularly louvers, if chosen and designed correctly, reduce excessive uncontrolled solar gains and daylight in interior space, consequently improving building energy performance and occupants' visual comfort. However, various questions related to their adaptation to different climates, their designs, and optimized methods are unanswered. This paper aims to show that even in geographical areas with similar climate conditions, different louver designs are required.To this end, louver designs in three Canadian cities with almost similar climates are investigated. Objective functions are defined as minimizing energy use intensity and maximizing daylight use. Through a sensitivity analysis, louver parameters have been examined to determine which ones have a significant or insignificant effect on energy performance and visual comfort. Based on the results obtained from the sensitivity analysis, the proper range of each parameter was determined, and a metaheuristic algorithm was applied to reduce time and cost of optimization. Two algorithms (SPEA-2 and NSGA-II) are compared to understand which performs better in this problem. The results show that NSGA-II performs slightly better, and, using louvers in all case study cities positively affects visual comfort and energy performance. Second, louver depth and count are the most critical parameters in all three cities, and other parameters have a different effect based on the city latitude. The research contribution is to propose a workflow to integrate parametric design, sensitivity analysis, and NSGA-II to optimize louver design with various parameters and ranges. Also, the research proposes a guideline for designing louver in the three investigated cities.