中文摘要：

针对多车整备时电气化铁路牵引供电系统电气量不稳定现象,通过挖掘多车接入牵引网时牵引变电站内部中压评估点的电压录波数据,提取多分量调幅调频信号的幅值和频率信息,并对其波动细节进行详细分析。首先设计高阶Butterworth滤波器对信号进行低通滤波。其次基于集成局域均值分解（ensemble local mean decomposition,ELMD）研究了牵引网电压波动包络线提取方法,分解出若干个单分量信号。为改善网压信号的端点效应,基于最小二乘法（least mean square,LMS）自适应滤波器对电压数据端点处进行波形延拓;随后使用对称差分Teager能量算子对调制频率信号的突变时刻检测,实现ELMD包络信号的分段截取;最后对各分段信号进行集成经验模态分解（ensemble empirical mode decomposition,EEMD）从而实现电压波动的详细分析和频率估计。实验结果表明,该方法能有效地提取牵引供电系统的电压波动特征,实现电压波动的调制信号的频率、幅值的有效估计。

英文摘要：

Aiming at unstable phenomena in traction power supply system of electrified railway for multi-vehicle service operation, recorded voltage data of traction substation was excavated in this paper. Amplitude and frequency of modulated FM signals are extracted, and then detailed fluctuation analysis is carried out. Firstly, high order Butterworth filter is designed to perform low pass filtering. Secondly, envelope extraction is studied and a number of single component signals are obtained based on ensemble local mean decomposition（ELMD）. In order to improve voltage endpoint effect, adaptive filter based on least mean square（LMS） is extended. Then symmetric difference Teager energy operator is used to detect mutation time of modulated frequency signal, realizing segmentation of ELMD envelope signal. Finally, ensemble empirical mode decomposition（EEMD） of the segmented signals is carried out to achieve detailed analysis of voltage fluctuations and frequency estimation. Experimental results show that the method can effectively extract traction power system voltage fluctuation characteristics and estimate voltage fluctuation frequency and amplitude.