Clustering Method for Residential Electricity Consumption Behavior Based on Feature Optimization Strategy

被引：0
|
作者：
Zhang J. [1 ]
Xia F. [1 ]
Yuan B. [2 ]
Liu W. [3 ]
机构：
[1] College of Automation Engineering, Shanghai University of Electric Power, Shanghai
[2] NARI Group Corporation (State Grid Electric Power Research Institute), Nanjing
[3] Haining Yada Plastic Pipeline System Corporation, Haining
来源：
Dianli Xitong Zidonghua/Automation of Electric Power Systems | 2022年 / 46卷 / 06期
基金：
中国国家自然科学基金;
关键词：
Cluster analysis; Correlation coefficient; Feature optimization strategy; Improved density peak method; Information criterion;
D O I：
10.7500/AEPS20200414003
中图分类号：
学科分类号：
摘要：
Aiming at the validity and complementarity problems of the clustering feature selection for the electricity consumption load data, a feature optimization strategy based on information criteria is proposed and the clustering method for the residential electricity consumption behavior based on the strategy is studied. First, the Bayesian information criterion (BIC) and correlation coefficient are used as the criteria of the validity and relevance of features to optimize clustering features. Then, the truncation distance of the density peak method is optimized by using the cuckoo algorithm. Meanwhile, the idea of outlier detection is used to realize the automatic selection of cluster centers. Finally, according to the optimized feature set and the improved clustering by fast search and find of density peaks (CFSFDP) algorithm, the cluster analysis on the actual residential electricity consumption data is carried out to verify the clustering effect of the proposed method. Experimental results of different clustering methods show that the proposed improved CFSFDP algorithm has the best clustering effect. © 2022 Automation of Electric Power Systems Press.
引用
收藏
页码：153 / 159
页数：6
相关论文
共 26 条
  • [1] QUILUMBA F L, LEE W, HUANG H, Et al., Using smart meter data to improve the accuracy of intraday load forecasting considering customer behavior similarities, IEEE Transactions on Smart Grid, 6, 1, pp. 911-918, (2015)
  • [2] GU Jie, MENG Lu, ZHENG Ruicheng, Et al., Load-stratified probability forecasting for massive users considering cluster identification, Automation of Electric Power Systems, 45, 5, pp. 71-78, (2021)
  • [3] PANG Chuanjun, YU Jianming, FENG Changyou, Et al., Clustering modeling and characteristic analysis of power load based on long-short-term memory auto-encoder, Automation of Electric Power Systems, 44, 23, pp. 57-63, (2020)
  • [4] LI Zhenkun, QIAN Jin, FU Yang, Et al., Multiple congestion management in distribution network based on optimization classification for load aggregators, Automation of Electric Power Systems, 45, 19, pp. 109-116, (2021)
  • [5] SUN Weiqing, LIU Xiaonan, XIANG Wei, Et al., Master-slave game based optimal pricing strategy for load aggregator in day-ahead electricity market, Automation of Electric Power Systems, 45, 1, pp. 159-167, (2021)
  • [6] GONG Gangjun, CHEN Zhimin, LU Jun, Et al., Clustering optimization strategy for intelligent power user behavior analysis, Automation of Electric Power Systems, 42, 2, pp. 58-63, (2018)
  • [7] ZHANG Yantao, SONG Bingbing, ZHOU Yi, Et al., Evaluation and analysis method of receiving end power grid acceptance capacity based on multi-level collaboration and its application, Electric Power Automation Equipment, 41, 6, pp. 121-126, (2021)
  • [8] JIANG Wenqian, LI Xinran, QIAN Jun, Application of improved FCM algorithm in outlier processing of power load, Proceedings of the CSU-EPSA, 23, 5, pp. 1-5, (2011)
  • [9] SONG Yiyang, LI Cunbin, QI Zhiqiang, Extraction of power load patterns based on cloud model and fuzzy clustering, Power System Technology, 38, 12, pp. 3378-3383, (2014)
  • [10] WANG Shuai, DU Xinhui, YAO Hongmin, Et al., Research on load curve clustering with multiple user types, Power System Technology, 42, 10, pp. 3401-3412, (2018)
123