基于自适应最优核时频分布理论的间谐波分析新方法

A Novel Approach to Interharmonics Analysis Based on Adaptive Optimal Kernel Time-frequency Distribution

间谐波是非整数倍基波频率的谐波信号,对电网的危害日益严重,因此,准确地分析间谐波的特征对电力系统具有十分重要的意义。提出一种基于自适应最优核时频分布理论的间谐波分析新方法。自适应最优核时频分布理论是一种现代信号处理方法,它是采用非线性变换处理非平稳信号的时频分析方法。文章首先对信号求取模糊函数,并在模糊域采用自适应最优高斯核函数来抑制交叉项,然后求模糊函数的二维傅里叶变换,从而得到仅有自项分量的时频分布。该法用于多分量、时变的含有间谐波的信号分析,具有很好的时频聚集性,强自适应性和抗噪声性能,且不受交叉项的影响,时间–带宽积几乎可以达到Heisen-berg不确定性原理给出的下界。仿真结果表明了该方法的有效性。

Interharmonics is a kind of harmonics signal which is not an integer of the fundamental frequency component. It is harmful to power system, so it is important to analyze interharmonics accurately. A novel approach to interharmonics analysis based on adaptive optimal kernel time-frequency representation （AOK TFR） is presented. AOK TFR is a time-frequency analysis theory that can analyze unstabilized signal by nonlinear transformation on modern signal processing. Adaptive optimal Gauss kernel is applied to restrain cross-components in ambiguity domain and then two dimensions FFF is done. Thus auto-components＇ timefrequency distribution is obtained. A kind of signal is analyzed with this method, which is multi-component, time-varied and interharmonics contained. Better time-frequency resolution, adaptiveness and anti-noise abilities are represented and this method is independent on multi-components. Its time- bandwidth product almost can meet lower limit of Heisen-berg theory. Its validity has been shown by simulation results.