摘要
目前,建筑能耗监测平台已在商业建筑中普遍使用。作为一种建筑能耗性能评估的手段,建筑能耗监测数据能够对监控建筑运行情况、分析建筑能耗水平起到重要作用。但在实际能耗监测平台运行过程中,由于设备故障及人员操作失误等原因,导致能耗监测数据出现异常和缺失,严重影响了建筑能耗水平分析和节能工作的进行。因此,分别以我国3个不同气候区的商场建筑为例对实测数据进行分析,针对不同分项能耗数据的特点,研究利用数据可视化与箱形图检测异常值,以及利用邻近点均值、线性回归和序列均值等方法填补缺失值的修复方法,并对过渡季中空调的启停控制进行判断。提出的方法能够有效地处理不同分项能耗数据中的异常值与缺失值,进而提高建筑能耗监测平台中数据的可利用率。
At present,building energy monitoring platform has been widely used in commercial buildings.As a promising measure of evaluating building energy performance,building energy consumption monitoring data plays an important role in building operation status monitoring and building energy consumption analysis.However,during the operation of the practical energy consumption monitoring platform,abnormal and missing energy consumption monitoring data often occurs due to equipment failure or personnel operation error,which seriously affected the analysis of building energy consumption and conservation.Actual measurement data of three shopping malls located within three different climate zones were analyzed.Data visualization and box plot method were applied to analyze the characteristics of energy consumption from different sections.A method of detecting errors such as outliers,missing data slots as well as repairing them via linear regression and sequence mean was developed.Finally,a method of detecting the on/off status of the HVAC systems during transition seasons was also presented.The method proposed in this paper can be an effective tool for data processing in energy consumption data of different items,so as to improve the availability of data in the monitoring platform of building energy consumption.
作者
李梅香
彭惠旺
陈毅兴
李定鹏
覃雪婷
黄宇
LI Mei-xiang;PENG Hui-wang;CHEN Yi-xing;LI Ding-peng;QIN Xue-ting;HUANG Yu(School of Civil Engineering,Guangzhou University,Guangzhou 510006,China;School of Architecture,South China University of Technology,Guangzhou 510641,China;School of Civil Engineering,Hunan University,Changsha 410082,China)
出处
《建筑节能（中英文）》
2021年第5期37-45,共9页
Building Energy Efficiency
基金
“十三五”重点研发计划资助项目“建筑全性能仿真平台内核开发”(2017YFC0702204)。
关键词
商业建筑
能耗监测
异常值
缺失值
数据处理
commercial building
energy consumption monitoring
outliers
missing value
data processing
