基于多元大数据平台的用电行为分析构架研究

随着智能电网的发展,越来越多的测量装置向底层延伸。高级测量体系和配电网的发展不可避免地使用户用电数据量呈几何倍数增长,另一方面,电网也在积极寻求方法让需求侧可以充分地参与电网调控,增强电网可控性和经济性。在上述背景下,运用配用电数据分析用户用电行为建立相关驱动方法,可充分利用现有资源,为政府政策制定、电力公司业务拓展和用电行为引导提供新的解决思路。在配用电数据采集、聚合、处理和应用等方面提出了以大数据平台为基础的整体构架,设计了基于流处理和批处理的数据驱动方法,提出了适用于多维大数据用电行为分析的随机矩阵相关性算法,最后讨论了用电行为分析面向不同对象的应用场景。

Architecture based on multivariate big data platform for analyzing electricity consumption behavior

With the development of smart grid, more and more measuring devices extend to bottom layer. The development of advanced measurement system and distribution network inevitably leads to the geometric increase of user data. On the other hand, the power grid is also actively seeking ways to allow the demand side to fully participate in the grid regulation for enhancing the grid controllability and economy. In the above background, the electricity consumption behavior of users is analyzed based on the data of electricity distribution and consumption to set the related driving methods for the full use of existing resources to provide new solutions to the formulation of government policy, the business development of electricity utilities and the guidance of electricity consumption behavior. An overall architecture based on the large data platform is given for the acquisition, aggregation, processing and application of electricity distribution and consumption data, a data driven method based on the stream processing and batch processing is designed, a stochastic matrix correlation algorithm suitable for the electricity consumption behavior analysis based on the multidimensional big data is proposed, and the application scenarios of electricity consumption behavior analysis are discussed for different objects.