基于EMD的时频分析方法的电力故障信号检测

提出电力系统故障信号的经验模态分解(Empirical Mode Decomposition,EMD)时频分析方法.通过对检测点获得的故障电流信号进行EMD分解,得到一系列的本征模态分量(Intrinsic Mode Function,IMF),利用Hilbert变换得到Hilbert谱及边际谱,分析相应瞬时频率及其振幅,对故障信号进行时间-频率-幅值的联合分析.分析发现,通过瞬时频率突变能准确定位故障时刻,Hilbert谱的峰值变化也能反映故障时刻及故障特征信息;通过边际谱分析可以获得故障信号所含的真实频率,为进一步故障检测提供了依据.仿真试验证明了结论的正确性,表明EMD时频分析方法能准确地检测故障时刻,提取故障信息.

Power fault detection using empirical mode decomposition

The method of time-frequency analysis of empirical mode decomposition(EMD) is proposed to analyze the fault signal in the power system.The current fault signal is successfully decomposed into a series of smooth intrinsic mode functions(IMFs) by EMD.Hilbert spectrum and Hilbert marginal spectrum are obtained from Hilbert spectra analysis.Its respective instantaneous frequency and amplitude are analyzed to obtain time-frequency-amplitude analysis.It is found fault time can be detected by instantaneous frequency and can be also detected by Hilbert spectrum;and the real frequency of fault signal is analyzed by Hilbert marginal spectrum,which has further provided the basis for the fault detection.The simulation indicated that,the time-frequency analysis of EMD method can detect accurately fault time and extract the fault information.