基于“电-碳”关系的公共建筑运行阶段碳排放测算方法

“Electricity-carbon Relationship”Based Carbon Emission Estimation Method in the Operation Phase of Public Buildings

公共建筑运行阶段碳排放统计监测是做好建筑领域碳达峰、碳中和工作的重要基础。目前对单体公共建筑运行阶段碳排放数据主要以审计、核查等形式进行统计,存在能源数据获取难度大、统计周期长等问题。电力运行数据在公共建筑中具有广泛的监测基础,为了提高公共建筑运行阶段碳排放数据的全面统计效率,本研究提出了1种基于“电-碳”关系的单体公共建筑运行阶段碳排放测算方法,采用机器学习算法构建建筑电气化率与非电力能源碳排放综合因子的双变量预测模型,可利用建筑基本信息及电力监测数据实现整体碳排放测算。以夏热冬冷地区某市的公共建筑能源审计数据为基础进行案例分析,应用该方法对目标建筑的年度、月度碳排放测算平均绝对百分比误差分别为6.29%与8.57%,具有较高的测算精度。

The statistical monitoring of carbon emission during the operation phase of public buildings is an important foundation for the work of carbon peak and carbon neutralization in the construction field.At present,the carbon emission data in the operation phase of public buildings are mainly counted in the form of audit and verification,and there are problems such as the difficulty of obtaining energy data and the long statistical cycle.Electricity operation data in public buildings has a wide range of monitoring basis,in order to improve the comprehensive statistics and monitoring efficiency of the carbon emission level in the operation phase of public buildings,this study proposes a carbon emission estimation method based on the"electricity-carbon"relationship in the operation phase of public buildings,using machine learning algorithms to construct a bivariate prediction of the electrification rate of the building and the comprehensive factor of carbon emission from non-electric energy sources.The machine learning algorithm is used to construct a bivariate prediction model of building electrification rate and non-electrical energy comprehensive factor,and the basic information of the building and the electricity monitoring data are used to realize the overall carbon emission estimation.A case study is made based on the audit data of public building energy in a city in hot summer and cold winter.The average relative errors of annual and monthly carbon emission estimation are 6.29%and 8.57%,respectively,with high accuracy.