Generating hourly electricity demand data for large-scale single-family buildings by a decomposition-recombination method

作     者：Mengjie Han Fatemeh Johari Pei Huang Xingxing Zhang 

作者机构：School of Information and EngineeringDalarna UniversityFalun 79188Sweden Built Environment Energy Systems GroupDepartment of Civil and Industrial EngineeringUppsala UniversityUppsala 75121Sweden 

出 版 物：《Energy and Built Environment》 (能源与人工环境(英文))

年 卷 期：2023年第4卷第4期

页      面：418-431页

核心收录：

学科分类：0711[理学-系统科学] 081302[工学-建筑设计及其理论] 08[工学] 0714[理学-统计学（可授理学、经济学学位）] 0813[工学-建筑学] 0701[理学-数学] 

基　　金：The authors are thankful for the financial support from the UBMEM project from the Swedish Energy Agency(Grant No.46068) 

主　　题：Data generation Time series decomposition Hourly electricity demand Large-scale buildings 

摘      要：Household electricity demand has substantial impacts on local grid operation,energy storage and the energy per-formance of *** demand data at district or urban level helps stakeholders understand the demand patterns from a granular time scale and provides robust evidence in energy ***,such type of data is often expensive and time-consuming to collect,process and *** built upon smart meter data have to deal with challenges of privacy and security in the whole *** data due to confiden-tiality concerns or system failure can further increase the difficulty of modeling and *** addition,methods using historical data to make predictions can largely vary depending on data quality,local building envi-ronment,and dynamic *** these challenges,this paper proposes a statistical method to generate hourly electricity demand data for large-scale single-family buildings by decomposing time series data and recom-bining them into *** proposed method used public data to capture seasonality and the distribution of residuals that fulfill statistical characteristics.A reference building was used to provide empirical parameter settings and validations for the studied *** illustrative case in a city of Sweden using only annual total demand was presented for deploying the proposed *** results showed that the proposed method can mimic reality well and represent a high level of similarity to the real *** average monthly error for the best month reached 15.9%and the best one was below 10%among 11 tested *** than 0.6%improper synthetic values were found in the studied region.