基于典型波形特征与改进DBSCAN的电压暂降同源识别方法  

Identification Method for Same-source Voltage Sags Based on Typical Waveform Characteristics and Improved Density-based Spatial Clustering of Applications with Noise

作　　者：汪颖[1] 谢佳妮 邓凌峰 肖先勇[1] WANG Ying;XIE Jiani;DENG Lingfeng;XIAO Xianyong(College of Electrical Engineering,Sichuan University,Chengdu 610065,China)

机构地区：[1]四川大学电气工程学院,四川省成都市610065

出　　处：《电力系统自动化》2021年第11期126-135,共10页Automation of Electric Power Systems

基　　金：国家自然科学基金资助项目(51807126)。

摘　　要：针对电网内单次短路故障可能触发多条母线上的电能质量监测装置记录电压暂降波形,造成电压暂降录波数据冗余、电网电压暂降水平过估计的问题,提出一种基于典型波形特征与改进基于密度的有噪空间聚类(DBSCAN)的电压暂降同源识别方法。首先,提出综合倾斜因子、波形相似度和持续时间三维特征,综合刻画电压暂降同源性;同时,研究电压暂降经多级变压器的传播特性,推导出13种变压器相电压等效传递矩阵,并提出基于Bhattacharyya距离的波形相似度计算方法。其次,利用共享近邻点(SNN)改进DBSCAN,对由三维特征构成的矩阵进行同源聚类,提出基于电网节点总数和监测点总数的聚类参数计算方法,实现电压暂降同源识别。应用实测数据和IEEE 30节点系统仿真数据进行验证,结果证明了所提方法的正确性和有效性。A single short-circuit fault in the power grid may trigger the power quality monitoring devices on multiple bus bars to record the voltage sag waveforms,which results in the redundant recording data and the overestimation of voltage sag.Aiming at these problems,this paper proposes an identification method for same-source voltage sags based on the typical waveform characteristics and the improved density-based spatial clustering of applications with noise(DBSCAN).Firstly,the threedimensional characteristics of the comprehensive tilt factor,the waveform similarity and the duration are proposed to comprehensively describe the same-source property of voltage sags.At the same time,the propagation properties of voltage sags through multiple levels of transformers are studied,and 13 kinds of equivalent transfer matrices of phase voltage of transformers are derived,while the waveform similarity calculation method based on Bhattacharyya distance is proposed.Secondly,the shared nearest neighbor(SNN)is used to improve DBSCAN,and the same-source clustering of the matrices composed of threedimensional characteristics is performed.Secondly,a clustering parameter calculation method based on the total number of power grid nodes and the total number of monitoring points is proposed to realize the same-source identification for voltage sags.The measured data and the simulation data of the IEEE 30-bus system are used for verification,and the results verify the correctness and effectiveness of the proposed method.

关 键 词：电压暂降 数据冗余 过估计 同源识别 三维特征 BHATTACHARYYA距离 基于密度的有噪空间聚类(DBSCAN) 

分 类 号：TM714.2[电气工程—电力系统及自动化]