Heuristic optimization for designing centralized air distribution systems in non-residential buildings

Abstract

Well-performed and efficient designs of air distribution systems are essential for maintaining a healthy and comfortable indoor environment in many buildings. Yet, designing these systems is challenging, and conventional methods still rely on rules of thumb, which may result in cost-inefficient and possibly underperforming designs. Until today, research efforts have focused on addressing a subset of the design problem; therefore, they are hardly used in practice. In this paper, we propose a simulation-based design method driven by metaheuristic techniques to optimize the overall system design. On top of that, the developed method resolves conventional methods' limitations, considers the variable occupant behavior, and assesses different performance aspects. Local search heuristic and constructive heuristic algorithms are utilized to accomplish the design optimization problem. The design optimization objective of the developed method is to minimize the life cycle costs. The developed method was tested on a floor plan of a multi-story university building. It generated an optimized design configuration with a 13.62% reduction in LCC in comparison with the existing one. The optimized design configuration was also more balanced. This output indicates the capability of the proposed optimization method to produce cost-efficient and well-performed designs, which can be useful for HVAC engineers, building owners, and occupants.