摘要
针对配电网高阻接地故障因特征微弱故易于与电容投切、负荷投切混淆的问题,提出了一种基于改进变分模态分解和特征分布系数的辨识方法。首先,利用变分模态分解算法处理线路暂态零序电流,得若干模态分量。其中,利用麻雀搜索算法来优化确定变分模态分解的参数。然后,选取峰度最大的模态作为特征模态,绘制其包络分布直方图,计算峰度、标准差、轴区采样判别系数。最后,根据综合特征分布系数是否超过阈值,实现对不同情况下的高阻接地故障和扰动事件的辨识。在PSCAD/EMTDC进行的大量仿真验证了该算法的可靠性。
To solve the problem that high impedance grounding fault of distribution network is easy to be confused with capacitor switching and load switching due to weak characteristics,an identification method based on improved variational mode decomposition and characteristic distribution coefficient is proposed in this paper.Firstly,the variational mode decomposition algorithm is used to deal with the transient zero-sequence current of the line,and several mode components are obtained.Among them,sparrow search algorithm is used to optimize and determine the parameters of variational mode decomposition.Then,the mode with the largest kurtosis is selected as the characteristic mode,the envelope distribution histogram is drawn,and the kurtosis,standard deviation and axial section sampling discriminant coefficient are calculated.Finally,according to whether the comprehensive characteristic distribution coefficient exceeds the threshold value,the high impedance grounding fault and disturbance events can be distinguished under different conditions.The reliability of the algorithm is verified by a lot of simulations on PSCAD/EMTDC software.
作者
李浩
张禄亮
LI Hao;ZHANG Luliang(School of Electric Power Engineering,South China University of Technology,Guangzhou 510006,China)
出处
《南方电网技术》
CSCD
北大核心
2022年第12期109-117,共9页
Southern Power System Technology
基金
广东省自然科学基金面上项目(2022A1515011587)
国家自然科学基金面上项目(52077081)。
关键词
高阻接地故障
变分模态分解
麻雀搜索算法
特征模态
特征分布系数
high impedance grounding fault
variational mode decomposition
sparrow search algorithm
characteristic mode
characteristic distribution coefficient
