基于大数据的电力用户群体识别与分析方法研究

Research on Method of Power User Group Identification and Analysis Based on Large Data

随着能源互联网及智能用电技术的发展,深入了解电力用户群体特征,并提供精准电力服务,成为"互联网+"智慧能源的重要研究内容.通过对售电体积累的用户社会属性、用电行为等大数据进行分析,构建用户间相似度权重模型,提出基于"用户-标签"二元网络的相似群体识别方法,并通过群体分析获取群属性和典型负荷特征,预测新入网用户的用电行为.该方法通过对标签数据的分析,便于发现群体中的重要用户,为互联网售电体实施个性化用电服务和增值服务推荐提供支撑,进而提升电力服务质量.

With the rapid development of Energy Internet and smart grid, it is an important part of “In-ternet plus” smart energy to have a good understanding of the power users group behaviors and providing accurate power service. Based on large data analysis of the power users, the user similarity weight model was built and the “user-label” two-unit network was put forward to identify similar groups. In addition, the group attributes and typical electricity consumption characteristics were obtained. This method could help to identify the important user groups, who might need customized electricity services and value-add-ed services, for power suppliers. So the quality of power service could be greatly enhanced.