“点”“域”结合法在低频振荡分析中的应用

Application of "Individual" and "Ensemble" Integration Method in Low Frequency Oscillation Analysis

随着互联电网规模的不断扩大,大区互联电网区域间的低频振荡已成为威胁电力系统安全的重要因素之一。基于传统的希尔伯特-黄变换(HHT)和复奇异值分解(C-SVD),本文提出了一种"点""域"结合的低频振荡分析方法。对于测量"点",也就是单一的测量信号,首先对滤波去噪以后的信号进行经验模态分解(EMD),得到固有模态函数(IMFs),并建立虚拟空间场来分析IMFs之间的能量分布关系,然后利用Hilbert变换来计算主导振荡模态的时变振荡参数;对于测量"域",利用C-SVD来提取主导振荡模态的动态时变特性和能量空间分布。通过比较"点"和"域"的计算结果来确定主导振荡模态的时变振荡参数和空间分布。仿真算例和实测数据的计算结果证明算法的可行性和有效性。

With the rapid development of inter-connected power system, the low frequency Oscillation has become one of serious factors threatening the power system stability. Based on the traditional Hilbert-Huang Transform（HHT） and complex singular value decomposition（C-SVD）, a novel method which combines ＂single measured signal＂ and ＂ensemble measurement matrix＂ is proposed to analyze low frequency oscillation problems in this paper. As for the individual measured signal, the energy relationships among intrinsic mode function（IMF） are explored by setting virtual ensemble matrix based on the decomposition results of empirical mode decomposition（EMD）. And then the instantaneous parameters of dominant IMF are calculated by Hilbert Transform. As for the ensemble measurement matrix, C-SVD is utilized to extract the dominant proper orthogonal mode（POM）. Next, the temporal characteristics and spatial distribution of dominant POM are analyzed. The oscillatory parameters and mode shape of dominant oscillation mode are determined by comparing the calculated results based on two mentioned different methods. The effectiveness of the proposed method is demonstrated by the simulation data and actual data in Europe transmission system.