• CSCD核心库收录期刊
  • 中文核心期刊
  • 中国科技核心期刊

电力建设 ›› 2020, Vol. 41 ›› Issue (5): 28-36.doi: 10.12204/j.issn.1000-7229.2020.05.004

• 虚拟电厂 ·栏目主持 胡泽春副教授、刘敦楠副教授、王宣元高级工程师· • 上一篇    下一篇

基于改进k-means算法的VPP负荷曲线聚类方法及应用

艾欣1,杨子豪1,胡寰宇1,王智冬2,彭冬2,赵朗2AI   

  1. 1.华北电力大学电气与电子工程学院,北京市 102206;2. 国网经济技术研究院有限公司,北京市 102209
  • 出版日期:2020-05-01
  • 作者简介:艾欣(1964),男,教授,博士生导师,主要研究方向为新能源电力系统与微网; 杨子豪(1995),男,硕士研究生,主要研究方向为电力系统分析与控制; 胡寰宇(1995),男,博士研究生,主要从事电力系统分析与控制方面的研究工作; 王智冬(1981),男,高级工程师,主要研究方向为大电网规划与量化分析; 彭冬(1977),女,教授级高级工程师,主要研究方向为大电网规划与量化分析; 赵朗(1990),男,工程师,主要研究方向为大电网规划与量化分析。
  • 基金资助:
    国家电网公司科技项目(基于数据驱动的电网发展智能诊断分析与综合决策技术研究)(SGJSJY00JJJS1900018)

A Load Curve Clustering Method Based on Improved K-means Algorithm for Virtual Power Plant and Its Application

Xin1, YANG Zihao1, HU Huanyu1, WANG Zhidong2, PENG Dong2 , ZHAO Lang2   

  1. 1. School of Electrical and Electronic Engineering, North China Electric Power University, Beijing 102206, China; 2. State Grid Economic and Technological Research Institute Co., Ltd., Beijing 102209, China
  • Online:2020-05-01
  • Supported by:
    This work is supported by Science and Technology Project of State Grid Corporation of China(No. SGJSJY00JJJS1900018).

摘要: 能源互联网的建设,将物联网、人工智能、云计算等技术融入电网。虚拟电厂作为能源互联网的基本单元,其聚合、运行方式也将迎来改变。针对虚拟电厂如何有效参与电网运行,提出一种基于主成分分析降维和凝聚层次聚类与k-means聚类相结合的虚拟电厂负荷曲线聚类方法,并对聚类结果的应用进行了研究。首先,结合信息物理网络所获数据,采用主成分分析方法对参与虚拟电厂聚合的不同负荷的特征进行分析,对数据进行标准化处理并降低维度;然后, 利用凝聚层次聚类和k-means聚类相结合的算法,对所有参与聚合的负荷出力曲线进行聚类,得到同类别的负荷曲线簇并找出聚类中心;最后,分析聚类结果,建立与之匹配的评价体系,通过综合评价选取合适的负荷组合参与虚拟电厂聚合。

关键词: 虚拟电厂, 负荷曲线, 主成分分析, k-means算法, 层次聚类, 综合评价

Abstract: The construction of energy internet integrates the Internet of Things, artificial intelligence, cloud computing and other technologies into the power grid. As the basic unit of energy internet, virtual power plant (VPP) will change its aggregation and operation mode. In view of how virtual power plants can effectively participate in power grid operation, this paper proposes a VPP load curve clustering method based on principal-component dimension-reduced analysis, aggregation level clustering and k-means clustering, and studies the application of the clustering results. Firstly, combined with the data obtained from the information physical network, the principal-component analysis method is adopted to analyze the characteristics of different loads participating in the VPP aggregation, so as to standardize the data and reduce the dimension. Then, the algorithm combining aggregation hierarchical clustering and k-means clustering is used to cluster all load output curves participating in the aggregation, to obtain load curve clusters of the same class and find out the clustering center. Finally, the clustering results are analyzed, and the corresponding evaluation system is established. Through comprehensive evaluation, appropriate load combinations are selected to participate in the VPP aggregation.

Key words: virtual power plant, load curve , principal component analysis, k-means algorithm, hierarchical clustering, comprehensive assessment

中图分类号: