ISODATA方法在配网状态估计不良数据辨识中的应用

Application of ISODATA to Bad-Data Identification in Distribution Networks

采用ISODATA技术 (迭代自组织数据分析技术A) ,利用标准残差Rn 和相邻采样时刻量测值之差ΔZ作为特征值 ,对量测数据进行模糊聚类分析 ,并根据隶属度的大小来辨识其是否属于不良数据 .数字仿真表明 ,该计算方法简单、快速、可靠 .

In the process of distribution system state estimation, bad data identification can make estimation more reliable. An ISODATA method for bad data identification is presented in this paper. In this method, the normalized residual and the difference between data measured at successive sampling times are taken as characteristic values, and by use of the fuzzy clustering analysis method, the bad data are identified based on the membership degree of measured data belonging to good data set. Digital test shows that the method is simple, fast, and reliable.