首页|基于负荷时间序列聚类的有序用电管理策略

基于负荷时间序列聚类的有序用电管理策略

扫码查看
针对目前有序用电方案制定未充分考虑用户负荷特性差异的问题,提出了一种基于负荷时间序列聚类的有序用电管理策略.首先提取单个用户的典型负荷曲线形态;之后采用Canopy和k-shape时间序列聚类算法,解决用户对象较多时传统聚类算法准确性不足的问题,对不同用户的负荷时序数据进行聚类,从而将多个用户划分成不同用电类型;最后根据不同类型用户的负荷特性,制定针对性的有序用电管理策略.以某区域电网实际负荷数据进行算例分析,结果表明,所提方法相比传统方法能更加准确地对多个不同负荷特性的用户进行分类,从而更有效地指导有序用电管理策略的合理制定.
Orderly power utilization management strategy based on load time series clustering
Aiming at the problem that in the current orderly power utilization planning the difference in user load characteristics is not fully considered,an orderly power utilization management strategy based on load time series clustering is proposed.First,the typical load curve shape of a single user is extracted.After that,the Canopy and k-shape time series clustering algorithms are used to solve the problem of insufficient accuracy of traditional clustering algorithms when there are many user objects,and the load time series data of different users are clustered,thereby classifying multiple users into different power consumption types.Finally,according to the load characteristics of different types of users,the targeted orderly power utilization management strategies are formulated.The actual load data of a certain regional power grid is analyzed as an example.The results show that the proposed method can more accurately classify multiple users with different load characteristics than traditional methods,so as to more effectively guide the rational formulation of orderly power utilization management strategies.

time seriesload clusteringorderly power utilizationload characteristicssimilarity measurement

方晓明、孙彬、郑峰、童光波、肖伟、丁同、陈红坤

展开 >

国网湖北省电力有限公司黄冈供电公司,湖北黄冈 438000

武汉大学电气与自动化学院,湖北武汉 430072

时间序列 负荷聚类 有序用电 负荷特性 相似性度量

国网湖北省电力有限公司黄冈供电公司科技项目

SGHBHG00FZWT2001457

2024

武汉大学学报(工学版)
武汉大学

武汉大学学报(工学版)

CSTPCD北大核心
影响因子:0.621
ISSN:1671-8844
年,卷(期):2024.57(1)
  • 27