基于HHT方法的电气化铁道谐波检测与分析

Electric Railway Harmonic Detection Based on HHT Method

将希尔伯特-黄变换(HHT)方法用于电气化铁道谐波检测中。由于电气化铁道谐波电压、电流信号中基波能量很大,其它次数的谐波能量相对较小,使得经验模态分解(EMD)方法在应用中出现模态混叠现象,不能准确地提取任意频率的谐波信号。为改善经验模态分解过程中产生的模态混叠现象,本文采用Yang提出的基于Fourier变换的EMD方法对电气化铁道谐波信号进行筛分。通过该方法可以有效地提取出任意频率的谐波分量,进而计算其Hilbert谱。通过对电气化铁道牵引变电所实测谐波电压、电流数据进行分析,结果表明:利用改进的HHT方法可以得到电气化铁道各次谐波准确的时频分布。最后通过HHT方法计算出各次谐波电压、电流含有率及总谐波电压、电流畸变率,并对计算结果进行分析。HHT方法为电气化铁道谐波检测与分析提供了一种新的途径。

Hilbert-Huang Transform （HHT） is applied in detection of harmonics in traction power systems. Mode-mixing appears during sifting by Empirical Mode Decomposition （EMD）, because the energy of fundamentals of electric railways are much larger than the other harmonics＇. Harmonics cannot be accurately ab- stracted at any frequencies with EMD. In order to sift the mode-mixing harmonic signals in good mode, an improved method named EMD based on Fourier Transform proposed by Yang is used in this paper. Harmonic components can be abstracted at any frequencies and then the Hilbert Spectrum is obtained by using Hilbert Transform. Actual data of traction substations are analyzed. The results show that the improved HHT method is effective to abstract harmonics at any frequencies and accurate time-frequency distribution of harmonic components can be obtained with this method. Total harmonic distortion （THD） and harmonic ratio （HR） are given in the last. HHT is a new method for harmonic detection and analysis in traction power systems.