Decomposition of Aggregate Electricity Demand into the Seasonal-Thermal Components for Demand-Side Management Applications in “Smart Grids”

详细信息    查看全文

• 关键词：Load decomposition and profiling ; Smart grids ; Demand ; side management ; Temperature ; demand dependencies ; Correlation and regression analysis ; Sliding ; window data analysis ; Power ; factor analysis
• 刊名：Lecture Notes in Computer Science
• 出版年：2017
• 出版时间：2017
• 年：2017
• 卷：10097
• 期：1
• 页码：116-135
• 丛书名：Data Analytics for Renewable Energy Integration
• ISBN：978-3-319-50947-1
• 卷排序：10097

文摘

Aggregate active and reactive power demands measured at 84 Scottish medium-voltage (MV) buses are used in this paper for the correlation and regression analysis, aimed at demand profiling and load decomposition. Demand profiles are presented with respect to the long-term seasonal variations, medium-term weekly and short-term diurnal cycles, allowing for the characterisation and presentation of load behaviour at different time-scales. The linear relationships between active and reactive power demands, temperature and power factor variations are quantified using regression analysis, on a per-hour of the day basis, as well as using a sliding-window regression approach for estimating relative coefficients within a seasonal moving window. The paper presents three different approaches for the decomposition of aggregate network demand into the temperature-dependent loads (i.e. thermal heating and cooling loads) and temperature-independent loads, providing important basic information for the application of the “smart grid” functionalities, such as demand-side management, or balancing of variable energy flows from renewable generation.