Big data-based optimized model of building design in the context of rural revitalization

Abstract Rural revitalization policy is an important strategic measure of the country, and optimizing the design of public buildings is an important step in achieving rural revitalization. The research innovatively focuses on energy consumption in public buildings and introduces a novel multi-objective particle swarm optimization algorithm (BBMOPSO-A [backbone multi-objective particle swarm optimization algorithm]), which is designed to address the characteristics of building energy consumption (BEC, the total amount of energy consumed in the operation of a building). The new method enhanced the search capability of the algorithm by introducing new local and individual extreme points to address the multi-objective nature of the BEC problem. In addition, to reduce the computational cost, the study also innovatively employed an agent-assisted model to effectively replace the original model, ensuring that performance was not sacrificed while cost was reduced. These experiments confirmed that this new method model could effectively reduce algorithm costs. The accuracy of this method was 5.02% higher than multi-objective genetic algorithm, 3.72% higher than the multi-objective artificial ant colony algorithm, and 2.67% higher than the multi-objective particle swarm algorithm. Therefore, the backbone multi-objective particle swarm optimization algorithm has better performance and is more suitable for optimizing building energy consumption.