基于泄漏电流的人工污秽试验数据特征量的状态转换分析

State transition analysis of artificial contaminated test discharge based on leakage current

为了降低高压线路中绝缘子发生污闪的几率,提出一种基于泄漏电流特征量的污闪在线监测和沿面电弧放电机理的研究方法。由于沿面电弧放电过程复杂,仅通过研究泄漏电流的幅值很难识别沿面电弧的放电状态,故提取泄漏电流信号包络线的均值、均方差、峰度、偏度和三次谐波畸变因子5个统计指标;将其看作5个随机过程的样本,分别计算各随机过程的一步状态转移矩阵,并分析其状态转移过程。结果表明:沿面电弧放电过程表现出了一定的统计规律;泄漏电流每个特征量仅包含电弧放电过程的部分信息,结合多个指标的分析可以更好地实现电弧放电状态的识别;由一步状态转移矩阵可求解多步状态转移矩阵,从而可以实现对未来各放电状态出现概率的计算和污闪风险的预测。

In order to reduce the probability of insulator pollution flashover in high voltage transmission lines,a method for studying flashover on-line monitoring system and mechanism of surface arc discharge was proposed based on the characteristics of leakage current in the present work.Due to the complexity of surface arc discharge process,only through the study of amplitude of leakage current is difficult to identify the state of the surface arc discharge,so the extraction of five statistical indicators including the mean,variance,kurtosis and skewness and three times harmonic distortion factor of the leakage current signal envelope,as samples of five stochastic process,then the state transition matrix of stochastic process was calculated and analyzed.Results show that the surface arc discharge process expressed certain statistical regularity;leakage current characteristics,each contains only one part of the arc discharge process information,combined with multiple index analysis can better recognise the arc discharge state;By step state transition matrix,multi-step state transition matrix can be obtained,which can achieve the calculation of the probability of discharge states in the future and the forecast of pollution flashover.