基于全息希尔伯特频谱分析的交流串联电弧故障电流特征研究

Research on Characteristics of AC Series Fault Arcs Based on Holo-Hilbert Spectral Analysis

针对交流串联电弧故障信号特征提取和分析的难题,提出采用循环掩码经验模态分解(itEMD)进行信号分解以获取多层内涵模态分量(IMF),结合全息希尔伯特变换(HHT)展现高维幅度调制频率调制(AM-FM)谱图的分析方法,从而获取高区分度的电弧故障信号特征谱图。采用基于GB/T 31143—2014的电弧故障实验平台,采集7种屏蔽负载模式下电弧故障电流信号,分析研究不同环境下电弧故障特征。所提方法对于电弧故障检测技术的开发具有一定的理论指导意义。

To solve the problem of extraction and analysis of alternative-current series arc fault characteristics,an analysis methodology that applies iterated masking empirical mode decomposition(itEMD)for current signal decomposition to obtain multi-layer intrinsic mode functions(IMFs)is proposed,and Holo-Hilbert transform(HHT)is utilized to represent high-dimensional amplitude-modulation-and-frequency-modulation(AM-FM)spectra,so that the high discriminative characteristic spectra of series arc faults can be achieved.The arc fault current signals are acquired under seven masking load modes in accordance with GB/T 31143—2014,and arc fault characteristics under different conditions are investigated by the proposed methodology.The presented study is of significant theoretical importance to the development of arc fault detection techniques.