基于广域行波波头频率分量幅值比信息的输电网非同步故障定位

现有电网行波故障定位方法大多依赖行波到达时间和波速信息,需要精确同步。提出一种利用线模电压初始行波波头频率分量幅值比信息的输电网故障定位方法。该方法基于初始行波波头频率分量幅值的衰减特征和行波最短传播路径,采用虚拟故障点法,利用全网测量点的实际和理论幅值比差值构建故障线路识别判据;在保证训练普适性和少训练样本的前提下,利用径向基函数(RBF)神经网络拟合初始行波波头2个不同频率分量幅值比与故障距离的非线性关系实现故障精确定位。所提方法只需初始行波波头较低的频率分量幅值比信息,无需全网布置测量点和精确同步。在PSCAD/EMTDC中建立了IEEE 30节点系统的仿真模型,仿真结果表明所提方法定位精确、可靠性较高。

Asynchronous fault location of transmission system based on wide area amplitude ratio information of frequency components in traveling wave fronts

Most of the existing fault location methods of traveling wave in power network mainly rely on the arrival time and wave velocity, which require accurate synchronization. A fault location method for transmission system based on the amplitude ratio information of frequency components in the aerial-mode voltage initial traveling wave front is proposed. Based on the attenuation characteristics of frequency component amplitude in the initial traveling wave front and the shortest propagation paths of traveling wave, a fault line detection criteria based on the difference between the actual and theoretical amplitude ratio of frequency components is constructed by using the virtual fault point method. Under the premise of ensuring the training universality and fewer training samples, the nonlinear relationship between the amplitude ratio of two different frequency components in the traveling wave front and the actual fault location is fitted by using RBF(Radial Basis Function) neural network, which can realize accurate fault location. The proposed method only relies on the amplitude ratio information of lower frequency component and does not need to arrange measuring points in the whole system and synchronize accurately. The simulation model of IEEE 30-bus system is set up in PSCAD/EMTDC, and the simulative results verify the high accuracy and reliability of the proposed method.