摘要
针对低压台区线损异常情况的判断问题,以电力公司用电信息采集系统采集的日线损率数据为基础,提出了一种基于k-medoids聚类算法的低压台区线损异常识别方法,并以某地区819个台区为例进行算法可靠性的验证.首先应用局部异常因子LOF算法对低压台区异常日线损率数据进行判断、筛选和剔除;其次应用k-medoids聚类算法对日线损率数据进行聚类分析,得到低压台区日线损率数据的聚类中心点和欧氏距离,从而实现低压台区线损异常情况的判断;最后通过819个低压台区的实际数据验证算法的合理性.结果表明,算法能够对低压台区线损的异常情况做出准确的判断.
Aiming at the problem of judging the line loss anomalies in low-voltage stations,based on the daily line loss rate data collected by the power company's electricity information collection system,a line loss anomaly in low-voltage stations based on k-medoids clustering algorithm is proposed Identify the method,and take 819 stations in an area as an example to verify the reliability of the algorithm.First,the local anomaly factor LOF algorithm is used to judge,filter and remove the abnormal daily line loss rate data in low-voltage stations;secondly,the k-medoids clustering algorithm is used to perform cluster analysis on the daily line loss rate data to obtain the daily line loss in low-voltage stations The center point of the data and the Euclidean distance are used to determine the abnormality of line loss in the low-voltage station area;finally,the rationality of the algorithm is verified by the actual data of 819 low-voltage station areas.The results show that the algorithm can make an accurate judgment on the abnormal condition of line loss in the low-voltage station area.
作者
薛明志
陈商玥
高强
XUE Ming-zhi;CHEN Shang-yue;GAO Qiang(School of Electrical and Electronic Engineering,Tianjin University of Technology,Tianjin 300384,China)
出处
《天津理工大学学报》
2021年第1期26-31,共6页
Journal of Tianjin University of Technology
