Journal of Computer Applications ›› 2017, Vol. 37 ›› Issue (8): 2405-2409.DOI: 10.11772/j.issn.1001-9081.2017.08.2405

Previous Articles     Next Articles

Residential electricity consumption analysis based on regularized matrix factorization

WANG Yang1,2, WU Fan1, YAO Zongqiang1, LIU Jie3, LI Dong3   

  1. 1. Tianjin Electric Power Company, State Grid Corporation of China, Tianjin 300010, China;
    2. Guanghua School of Management, Peking University, Beijing 100871, China;
    3. College of Computer and Control Engineering, Nankai University, Tianjin 300350, China
  • Received:2016-12-09 Revised:2017-03-10 Online:2017-08-10 Published:2017-08-12
  • Supported by:
    This work is partially supported by the National High Technology Research and Development Program (863 Program) of China (2011AA05A117),the Major Projects of Tianjin Science and Technology Support Program (17YFZCGX00610),the Science and Technology Project of Tianjin Electric Power Company (KJ15-1-35,KJ17-1-23).


王扬1,2, 吴凡1, 姚宗强1, 刘杰3, 李栋3   

  1. 1. 国家电网 天津市电力公司, 天津 300010;
    2. 北京大学 光华管理学院, 北京 100871;
    3. 南开大学 计算机与控制工程学院, 天津 300350
  • 作者简介:王扬(1983-),男,天津人,高级工程师,博士,主要研究方向:智能信息处理;吴凡(1973-),男,上海人,高级工程师,主要研究方向:电力信息化;姚宗强(1968-),男,天津人,高级工程师,主要研究方向:电力信息化;刘杰(1979-),男,河北唐山人,副教授,博士,CCF会员,主要研究方向:机器学习;李栋(1980-),男,天津人,博士,主要研究方向:数据挖掘。
  • 基金资助:

Abstract: Focusing on the electricity user group feature, a residential electricity consumption analysis method based on geographic regularized matrix factorization in smart grid was proposed to explore the characteristics of electricity users and provide decision support for personalized better power dispatching. In the proposed algorithm, customers were firstly mapped into a hidden feature space, which could represent the characteristics of users' electricity behavior, and then k-means clustering algorithm was employed to segment customers in the hidden feature space. In particular, geographic information was innovatively introduced as a regularization factor of matrix factorization, which made the hidden feature space not only meet the orthogonal characteristics of user groups, but also make the geographically close users mapping close in hidden feature space, consistent with the real physical space. In order to verify the effectiveness of the proposed algorithm, it was applied to the real residential data analysis and mining task of smart grid application in Sino-Singapore Tianjin Eco-City (SSTEC). The experimental results show that compared to the baseline algorithms including Vector Space Model (VSM) and Nonnegative Matrix Factorization (NMF) algorithm, the proposed algorithm can obtain better clustering results of user segmentation and dig out certain power modes of different user groups, and also help to improve the level of management and service of smart grid.

Key words: smart grid, Nonnegtive Matrix Factorization (NMF), energy information of electricity users, customer segmentation, Geographic Information System (GIS)

摘要: 针对细粒度、多类别的用户用电行为分析问题,提出了基于地理信息正则化矩阵分解的居民用户用电行为分析算法,探索用户用电的群体特点,为个性化的、更优的电力调度提供决策支持依据。该模型首先基于矩阵分解理论将用户映射到能表征其用电行为特点的潜在特征空间,然后采用k-means聚类算法在潜在特征空间上实现用电用户群的细分聚类。特别地引入了地理信息作为矩阵分解的正则化因子,使得学习到的潜在特征空间不仅满足用户群特征的正交,而且使得地理位置相近的用户在潜在特征空间的映射也相近,与真实物理空间保持一致。将所提方法应用于中新天津生态城智能电网采集到的真实居民用电数据分析挖掘任务中。实验结果表明,与基准的向量空间模型(VSM)和非负矩阵分解(NMF)算法相比,所提方法能够取得更好的用户细分聚类结果,挖掘出一定的用户群体用电模式,有助于辅助智能电网提升经营和服务水平。

关键词: 智能电网, 非负矩阵分解, 用户用电信息, 用户细分, 地理信息系统

CLC Number: