基于自编码器的电力负荷聚类分析
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

TM933

基金项目:

国家自然科学基金(51767023)


Clustering Analysis of Power Load Curve Based on Auto-Encoder
Author:
Affiliation:

Fund Project:

Research on Strategy of virtual power plant dynamic peak planning to improve new energy consumption

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    电力负荷聚类分析研究是负荷特性模拟、需求侧管理等应用的基础。针对负荷数据日趋多样性、随机性,传统K-means算法无法有效处理高维数据,且存在人工给定聚类数目K值及随机选取初始聚类中心易收敛至局部最优的问题,本文提出一种基于自编码器(Auto-Encoder,AE)降维的电力负荷聚类方法。首先利用自编码器网络对采集的负荷数据提取特征,降低数据维度,然后通过密度权值Canopy算法对降维后的数据预聚类,得到初始聚类中心和最优聚类数目K值,将预聚类结果结合K-means算法进行聚类。算例结果表明,该方法能够有效对负荷数据进行特征提取,并减少聚类过程中的复杂度,提高了聚类结果准确度和聚类效率。

    Abstract:

    Power load clustering analysis is the basis of load characteristic simulation, demand side management and other applications. In view of the increasing diversity and randomness of load data, the traditional K-means algorithm cannot effectively process high-dimensional data, and there are problems such as manually giving the clustering number K value and randomly selecting the initial clustering center, which is easy to converge to the local optimum. In this paper, a power load clustering method based on Auto-Encoder (AE) is proposed to reduce the dimension. First, the auto encoder network was used to extract features from the collected load data and reduce the data dimensions. Then, the density weight Canopy algorithm was used to precluster the data after dimensionality reduction to obtain the initial cluster center and the optimal cluster number K value. The preclustering results were combined with the K-means algorithm for clustering.The example results show that this method can effectively extract features from load data, reduce the complexity of clustering, and improve the accuracy and efficiency of clustering results.

    参考文献
    相似文献
    引证文献
引用本文

赵忠啟,常喜强,樊艳芳,等. 基于自编码器的电力负荷聚类分析[J]. 科学技术与工程, 2021, 21(32): 13737-13743.
Zhao Zhongqi, Chang Xiqiang, Fan Yanfang, et al. Clustering Analysis of Power Load Curve Based on Auto-Encoder[J]. Science Technology and Engineering,2021,21(32):13737-13743.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2021-03-31
  • 最后修改日期:2021-09-16
  • 录用日期:2021-08-19
  • 在线发布日期: 2021-11-16
  • 出版日期:
×
律回春渐,新元肇启|《科学技术与工程》编辑部恭祝新岁!
亟待确认版面费归属稿件,敬请作者关注