Building performance optimization for university dormitory through integration of digital gene map into multi-objective genetic algorithm

• Published in: Applied energy Vol. 307; p. 118211
• Main Authors: Chi, Fang'ai, Xu, Ying
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.02.2022
• Subjects: algorithms; Digital gene map; electricity; energy; Energy consumption; genes; Multi-objective genetic algorithm; statistics; Thermal comfort; Visual comfort; Multi-objective genetic algorithm; Visual comfort; Energy consumption; Thermal comfort; Digital gene map
• ISSN: 0306-2619; 1872-9118
• DOI: 10.1016/j.apenergy.2021.118211

•A multi-objective approach was proposed to optimize the dormitory building design.•We proposed the digital gene map for the university dormitory building.•Most important building performance evaluation criteria were employed in the MOGA.•Optimized solutions can provide some suitable trade-offs between the multiple objectives.•The proposed methodology can be applicable to different types of university dormitories. College students spent most of their time in dormitory buildings, which account for a large proportion of the overall electricity use in university. Building performance is determined by its genes (i.e., building component characters), which can be optimized in the design stage. However, a building is comprised of many components, of which some are contradictory to each other in the building performance optimization process. In doing so, a digital gene map was proposed for the university dormitory, characterized by the binary code strings. Based on the digital gene map, the building elements can be parameterized to create some dynamic variables, to facilitate the multi-objective genetic algorithm. Via the multi-objective genetic algorithm and the data statistics tool of Design Explorer in this work, the “Pareto front” solutions can be obtained to optimize the decision-makings in the dormitory building design. For evaluations of the building performance improvement potentials for various types of study rooms with optimized solutions, we conducted the comparison studies in this work. Through comparison studies, we found that the optimized solutions from the multi-objective genetic algorithm, for the nine types of study rooms, have better compromised building performances. The methodology proposed in this work can be applicable for different types of university dormitories under various climate conditions, due to the dormitory buildings have a similar gene pool and chromosomal structure.