基于DBSCAN算法的电力自动化计量异常数据点剔除OA
Abnormal Data Point Elimination of Power Automation Measurement Based on DBSCAN Algorithm
电力自动化计量数据易受噪声干扰,导致异常数据识别非常困难,因此,提出基于密度的带噪声空间聚类(DBSCAN)算法的电力自动化计量异常数据点剔除方法.提取电力自动化计量数据特征,将时间序列重新排序.计算电力自动化计量数据点的局部密度与最小距离,判断并剔除异常计量数据点.实验结果表明,所提出的方法对电力自动化计量异常数据点检测准确,且剔除效果好.
The power automation measurement data is easy to be disturbed by noise,which makes it difficult to identify abnor-mal data.Therefore,this paper presents an abnormal data point elimination method of power automation measurement based on density based spatial clustering of application with noise(DBSCAN)algorithm.The characteristics of power automation meas-urement data are extracted and the time series is reordered.The local density and minimum distance of power automation meas-urement data point are calculated,and abnormal measurement data points are judged and eliminated.Experimental results show that the proposed method can accurately detect abnormal data point of power automation measurement and has good effect in elimination.
董新微;王超;江蒗;胡志亮;黄心心
安徽继远软件有限公司,安徽,合肥 230088安徽继远软件有限公司,安徽,合肥 230088安徽继远软件有限公司,安徽,合肥 230088安徽继远软件有限公司,安徽,合肥 230088安徽继远软件有限公司,安徽,合肥 230088
信息技术与安全科学
DBSCAN算法电力自动化计量数据局部离群因子算法数据点剔除异常检测
DBSCAN algorithmpower automation measurement datalocal outlier factor algorithmdata point eliminationabnormal detection
《微型电脑应用》 2026 (3)
61-65,5
国家电网公司资助项目(2022B0505020006)
评论