基于高频暂态分量进行相关分析及模糊推理的选相新方法

Novel Technique for Fault Phase Selection Based on Correlation Analysis and Fuzzy Set Using the Transient Signals in EHV Power System

提出一种利用故障分量中高频分量实现故障选相的新原理。该原理直接利用电流互感器饱和前传变的暂态高频电流,使用小波提取相应频段暂态信号特征,对提取的信号特征进行相关分析,使用模糊集合对相关系数的隶属度综合评判进行选相。用该原理实现的选相装置可以配合基于暂态量的单端保护使用,具有超高速的特点,且不受过渡电阻、故障初始角及系统振荡的影响。对于三相换位不完全造成模量不平衡的情况也能进行正确选相。对某典型500 kV线路进行各种故障类型仿真,取得了满意的效果。

This paper presents a novel technique for fault phase selection in EHV power systems based on transient protection theory. This technique uses the transient current transferred from current transformer before its saturation. Wavelet transform is proposed to extract the high frequency signals and correlation analysis is applied to the results. When different types of fault occur, the feature of high frequency signals shows different similar relationships. By analyzing the characteristics of faults, a series of new principles are derived base on fuzzy set. This scheme is insensitive to fault resistance, fault inceptive angle and power swing. It can also perform correctly in non-balanced transmission lines. The EMTP simulation verifies all results presented in the paper.