基于改进动态主元分析的配电网高阻故障检测

High Resistance Fault Detection of Distribution Network Based on Improved Dynamic Principal Component Analysis

传统配电网高阻故障检测方法无法对配电网数据缺失值完成插补,导致零序电流和零序电压的检测结果存在偏差。现提出基于改进动态主元分析的配电网高阻故障检测方法。利用GMDH算法填充配电网的数据缺失值,寻找配电网数据的最优复杂度。改进动态主元分析法,扩充配电网数据,建立配电网电量观测数据矩阵。求解该矩阵,获取SPE及T;的统计量,完成配电网的高阻故障检测。实验结果表明,利用所提方法检测配电网高阻故障,可有效检测出零序电流和零序电压,验证了所提方法具有一定应用优势。

Recently,due to the lack of data missing value interpolation in the distribution network,the detection results of zero-sequence current and zero-sequence voltage in the distribution network high resistance fault detection method have large errors.In this paper,a high resistance fault detection method for distribution networks based on improved dynamic principal component analysis is designed.GMDH algorithm was introduced to fill in the missing data of the distribution network for obtaining the optimal complexity of distribution network data.The dynamic principal component analysis method was improved to expand the distribution network data,founding and solving the distribution network electricity observation data matrix.The statistics of SPE and T;were obtained to detect the high resistance fault of the distribution network.The experimental results show that the zero-sequence current and zero-sequence voltage of the distribution network high resistance fault detected by the algorithm have minor errors.