基于同步挤压小波变换的抗混叠低频振荡模态参数识别

Identification of anti-aliasing modal parameters for low frequency oscillations based on synchrosqueezing wavelet transform

针对经验模态分解法对低频振荡信号模态提取时,存在相邻频率分量混叠而导致分析结果不正确的问题,本文提出基于同步挤压小波变换的抗混叠低频振荡模态参数识别新方法,首先利用同步挤压小波变换将低频振荡信号分解为一组无频率混叠的固有模态分量,实现各固有模态的精确提取;其次对各固有模态分量进行希尔伯特(Hilbert)变换、计算其相对应瞬时幅值、瞬时频率及相位;最后运用瞬时频率、瞬时幅值计算其阻尼比,从而实现对低频振荡模态参数的有效识别,数值仿真及实例分析均表明该方法的可行性和有效性。同时该方法有助于评价阻尼控制器对系统不同振荡模态阻尼特性的影响,为阻尼控制器的设计研究及改进提供理论支撑,具有较好的实用价值。

In order to solve the problem of incorrect analysis results caused by the aliasing of adjacent frequency components when the empirical mode decomposition(EMD) is used to extract the modes of low frequency oscillation signal(LFO), a novel method based on synchrosqueezing wavelet transform(SST) is introduced to extract the anti-aliasing model parameters of LFO. It applies SST to extract a set of intrinsic mode components without frequency aliasing in LFO signal and calculate s the instantaneous parameters including magnitude, frequency and phase of every model function with Hilbert transformation(HT), and then calculates the damping ratio of every model function based on the parameters above and the formulas derived. Thus it realizes effective extraction of all parameters of LFO signal. This method can be adopted to analyze strong non-linear oscillation models in power systems and design damping controllers. The simulation and example analysis results indicate the feasibility and validity of the new method proposed in this paper.