基于数据驱动和物理模型的居住建筑节能策略研究

Investigation on Residential Building Energy-Saving Strategies Based on Data-Driven Model and Physics-based Model

基于2018年度北京市1511栋居住建筑的基本信息,以及电力、燃气、集中热力等完整分项能耗数据,分析得出北京市典型居住建筑特点,建立了符合居住建筑能耗特征的数据驱动模型。同时,在软件EnergyPlus中建立了北京市典型居住建筑的物理模型,并利用数据驱动模型得出能耗指标对该物理模型进行校准。基于校准后的模型,对常见的居住建筑节能策略进行了定量研究和比较。采用数据驱动模型和物理模型相结合的方式,探索了能耗统计大数据应用的方法,为今后北京市居住建筑节能政策的制定和大规模推广提供了有效的数据支撑。

Based on the basic information of 1511 residential buildings in Beijing and the complete energy consumption data such as electricity,gas,and central heating of the year 2018,a reference residential building in Beijing was obtained,and a data-driven model of residential building energy consumption characteristics was established.Meanwhile,a physical model of the reference residential building in Beijing was established in the software EnergyPlus,and the physical model was calibrated against the energy consumption index derived from the data-driven model.Based on the calibrated model,quantitative research of common residential building energy-saving strategies was carried out.The study uses a combination of data-driven models and physical models to explore new methodology for the application of big data in building energy consumption,to provide effective support for the promotion of energy efficiency policies for residential buildings in Beijing.