基于PMU误差校正的输电线路参数在线辨识方法  

Transmission line parameter identification method based on PMU error correction

作　　者：张江南 雷江龙 贺勇 吴坡 余娟[2] 向明旭 任鹏凌 ZHANG Jiangnan;LEI Jianglong;HE Yong;WU Po;YU Juan;XIANG Mingxu;REN Pengling(State Grid Henan Electric Power Company Research Institute,Zhengzhou 450052,China;State Key Laboratory of Power Transmission Equipment&System Security and New Technology(Chongqing University),Chongqing 400044,China)

机构地区：[1]国网河南省电力公司电力科学研究院,河南郑州450052 [2]输配电装备及系统安全与新技术国家重点实验室(重庆大学),重庆400044

出　　处：《电力系统保护与控制》2022年第19期130-137,共8页Power System Protection and Control

基　　金：国家电网公司科技项目资助(524609210084)。

摘　　要：线路参数辨识对准确反映电网的实际运行状态具有重要作用,但其对相量量测单元(Phasor Measurement Unit,PMU)数据精度要求极高,实测数据难以直接满足应用需求。对此,提出了一种基于PMU误差校正的输电线路参数辨识方法。首先,根据误差类型及测量数据特点,分类型对PMU数据进行预处理。然后,针对缺乏显著特征的偏置误差,基于输电线路物理等效模型提出了考虑数据规律及误差灵敏度特性的PMU测量误差的校正方法。最后,在上述校正方法的基础上,提出了全厂站PMU数据质量的治理方案。通过模拟和实际数据进行仿真,结果表明所提方法相较于现有方法可改善PMU量测数据质量,进而实现输电线路参数的有效辨识。Line parameter identification plays an important role in accurately reflecting the actual running state of a power grid,but it requires high precision of phasor measurement unit(PMU)data,so the measured data cannot directly meet the application requirements.Therefore,a transmission line parameter identification method based on PMU error correction is proposed in this paper.First,the PMU data is pre-processed based on categories using different error types and data characteristics.Then,given the bias error lacking significant features,a PMU error correction method based on physical equivalent model of transmission line is proposed.This considers the data law and error sensitivity characteristics.Finally,on the basis of the above correction method,a data quality governance program of the whole PMU station is proposed.Through simulation and measured data,the results show that the method proposed can improve the quality of PMU measurement data compared with the existing methods,and then realize the effective identification of transmission line parameters.

关 键 词：输电线路 参数辨识 PMU误差校正 数据治理 

分 类 号：TM75[电气工程—电力系统及自动化]