Cloud-edge collaboration method for abnormal power consumption pattern recognition considering dynamic expression of information

被引：0
|
作者：
Liu H. [1 ]
Wang Y. [1 ]
Hu W. [1 ]
Xiao X. [1 ]
机构：
[1] College of Electrical Engineering, Sichuan University, Chengdu
来源：
Dianli Zidonghua Shebei/Electric Power Automation Equipment | 2022年 / 42卷 / 07期
基金：
中国国家自然科学基金;
关键词：
abnormal power consumption recognition; cloud-edge collaboration; data compression; dynamic expression of information; similarity measurement;
D O I：
10.16081/j.epae.202204051
中图分类号：
学科分类号：
摘要：
Abnormal power consumption recognition is an important part of the power consumption check and the operation status identification of metering devices，and is of great significance to maintain the safe operation of power grid and protect the rights and interests of normal users. In order to recognize multiple power consumption patterns of users，existing methods tend to cause complicated calculation on the basis of ensuring recognition accuracy. While，simple calculation methods considering efficiency are difficult to accurately measure the similarity of different power consumption patterns，so it is difficult to give consideration to calculation efficiency and accuracy. In addition，uploading power consumption data to the cloud for centralized computing consumes a large amount of network bandwidth and computing resources，which further limits the application of anomaly recognition. Therefore，a cloud-edge collaboration method for abnormal power consumption pattern recognition considering the dynamic expression of information is proposed. According to the computing resources of the edge terminal and the cloud，the cooperative tasks are reasonably allocated to realize the cloud-edge collaboration recognition of abnormal power consumption. Aiming at the problem of limited computing power of edge servers，the dynamic compression and re-expression of power consumption data are carried out to reduce the data amount and ensure the accuracy of data information. After receiving the compressed data，the cloud takes the segmented weighted dynamic time warping distance as the basis for the similarity measurement of compressed data，and identifies abnormal power consumption based on the density clustering algorithm with adaptive parameter selection. The effectiveness of the proposed method is verified based on the actual data set. © 2022 Electric Power Automation Equipment Press. All rights reserved.
引用
收藏
页码：59 / 67
页数：8
相关论文
共 21 条
  • [11] SONG Junying, CUI Yiwei, LI Xinran, Et al., Load curve clustering method combining improved piecewise linear representation and dynamic time warping［J］, Automation of Electric Power Systems, 45, 2, pp. 89-96, (2021)
  • [12] SHI Liangyuan, ZHOU Renjun, LI Juan, Et al., New energy-load characteristic index based on time series similarity measurement ［J］, Electric Power Automation Equipment, 39, 5, pp. 75-81, (2019)
  • [13] WANG He, LI Shiqiang, YU Huanan, Et al., Compression acquisition method for power quality data of distribution network based on distributed compressed sensing and edge computing［J］, Transactions of China Electrotechnical Society, 35, 21, pp. 4553-4564, (2020)
  • [14] SUN Wenwen, HE Guoqing, LIU Chun, Et al., Research and application of key technologies for edge computing equipment used in photovoltaic power generation under background of IoT［J］, Electric Power Automation Equipment, 41, 7, pp. 38-43, (2021)
  • [15] PIGLIONE F., Data size reduction with symbolic aggregate approximation for electrical load pattern grouping［J］, IET Generation，Transmission & Distribution, 7, 2, pp. 108-117, (2013)
  • [16] MOU Tingting, Wei LU, WANG Lanjun, Et al., Stability analysis of consumption mode based on principal component analysis［J］, Automation of Electric Power Systems, 41, 19, pp. 59-65, (2017)
  • [17] DOU Jian, LIU Xuan, LU Jizhe, Et al., Research on electricity anti-stealing method based on power consumption information acquisition and big data［J］, Electrical Measurement & Instrumentation, 55, 21, pp. 43-49, (2018)
  • [18] XU Yanjun, WU Meng, BAI Jialing, Et al., Identification method of abnormal events of gateway metering device based on multi-feature extraction and deep learning［J］, China Measurement & Testing Technology, 47, 5, pp. 104-111, (2021)
  • [19] Nengwen ZHAO, Jing ZHU, Yao WANG, Et al., Automatic and generic periodicity adaptation for KPI anomaly detection［J］, IEEE Transactions on Network and Service Management, 16, 3, pp. 1170-1183, (2019)
  • [20] SANDER J，, Et al., A density-based algorithm for discovering clusters in large spatial databases with noise［J］, Knowledge Discovery & Data Mining, 96, 34, pp. 226-231, (1996)
123