Anomaly detection algorithm based on improved <i>K</i>-means for electric power data

WU Rui; ZHANG Anqin; TIAN Xiuxia; ZHANG Ting

doi:10.3969/j.issn.1000-5641.201921012

Issue 4

Jul. 2020

Turn off MathJax

Article Contents

Article Navigation > Journal of East China Normal University (Natural Sciences) > 2020 > (4): 79-87

WU Rui, ZHANG Anqin, TIAN Xiuxia, ZHANG Ting. Anomaly detection algorithm based on improved K-means for electric power data[J]. Journal of East China Normal University (Natural Sciences), 2020, (4): 79-87. doi: 10.3969/j.issn.1000-5641.201921012

Citation:

WU Rui, ZHANG Anqin, TIAN Xiuxia, ZHANG Ting. Anomaly detection algorithm based on improved K-means for electric power data[J]. Journal of East China Normal University (Natural Sciences), 2020, (4): 79-87. doi: 10.3969/j.issn.1000-5641.201921012

Citation:

PDF( 836 KB)

Anomaly detection algorithm based on improved K-means for electric power data

doi: 10.3969/j.issn.1000-5641.201921012

College of Computer Science and Technology, Shanghai University of Electric Power, Shanghai　200090, China

Received Date: 2019-08-25
Available Online: 2020-07-20
Publish Date: 2020-07-20

Abstract

Abstract

Anomaly detection methods are widely used for applications in the field of electric power, such as equipment fault detection and abnormal electricity consumption detection. The proposed algorithm combines densities of data objects with the maximum neighborhood radius to select data points that are closer to actual cluster centers for the initial selection; this, in turn, improves random selection of the initial cluster centers. In addition, a new anomaly detection method based on an improved K-means algorithm for electric power data is proposed. Experiments show that the algorithm is more suitable in both clustering performance and anomaly detection. When this algorithm is applied to the field of electric power, abnormal data can be effectively detected.
- initial cluster centers,
- density,
- anomaly detection

FullText(HTML)

References(15)

References

[1]	邱承武, 宓群超. 利用电力数据四级网实现小电厂数据的采集 [J]. 电力系统自动化, 2006, 30(3): 105-106. DOI: 10.3321/j.issn:1000-1026.2006.03.020.
[2]	秦璇. 电力统计数据的质量评估及其异常检测方法研究 [D]. 长沙: 长沙理工大学, 2013.
[3]	刘冬兰, 马雷, 刘新, 等. 基于深度学习的电力大数据融合与异常检测方法 [J]. 计算机应用与软件, 2018, 35(4): 61-64. DOI: 10.3969/j.issn.1000-386x.2018.04.011.
[4]	蒋盛益, 姜灵敏. 一种高效异常检测方法 [J]. 计算机工程, 2007, 33(7): 166-168. DOI: 10.3969/j.issn.1000-3428.2007.07.060.
[5]	MA J, CHEN Y S, HUANG Z Y, et al. Detect abnormal SCADA data using state estimation residuals [C]// IEEE Transmission & Distribution Conference & Exposition. IEEE, 2010.
[6]	王桂兰, 周国亮, 赵洪山, 等. 大规模用电数据流的快速聚类和异常检测技术 [J]. 电力系统自动化, 2016, 40(24): 27-33. DOI: 10.7500/AEPS20160123002.
[7]	庄池杰, 张斌, 胡军, 等. 基于无监督学习的电力用户异常用电模式检测 [J]. 中国电机工程学报, 2016, 36(2): 379-387.
[8]	MACQUEEN J. Some methods for classification and analysis of multivariate observe [C]// Proceedings of the 5th Berkeley Symposium on Mathematical Statistics and Probability. 1967: 281-297.
[9]	行小帅, 潘进, 焦李成. 基于免疫规划的K-means聚类算法 [J]. 计算机学报, 2003, 26(5): 605-610. DOI: 10.3321/j.issn:0254-4164.2003.05.012.
[10]	段桂芹. 基于均值与最大距离乘积的初始聚类中心优化K-means算法 [J]. 计算机与数字工程, 2015, 43(3): 379-382. DOI: 10.3969/j.issn1672-9722.2015.03.008.
[11]	李炎, 李皓, 钱肖鲁, 等. 异常检测算法分析 [J]. 计算机工程, 2002, 28(6): 5-6. DOI: 10.3969/j.issn.1000-3428.2002.06.003.
[12]	ISENKUL M E, SAKAR B E, KURSUN O, et al. Improved spiral test using digitized graphics tablet for monitoring Parkinson’s disease [C]// The 2nd International Conference on e-Health and Telemedicine. 2014: 171-175.
[13]	SAKAR B E, ISENKUL M E, SAKAR C O, et al. Collection and analysis of a parkinson speech dataset with multiple types of sound recordings [J]. IEEE Journal of Biomedical and Health Informatics, 2013, 17(4): 828-834. DOI: 10.1109/JBHI.2013.2245674.
[14]	左进, 陈泽茂. 基于改进K均值聚类的异常检测算法 [J]. 计算机科学, 2016, 43(8): 258-261. DOI: 10.11896/j.issn.1002-137X.2016.08.052.
[15]	韩最蛟. 基于数据密集性的自适应K均值初始化方法 [J]. 计算机应用与软件, 2014, 31(2): 182-187. DOI: 10.3969/j.issn.1000-386x.2014.02.049.