转子系统电机电流信号特征提取方法  

Approach to Extraction of Rotor System Motor Current Signal

作　　者：张天赐[1] 庞新宇[1,2] 杨兆建 ZHANG Tian-ci;PANG Xin-yu;YANG Zhao-jian(College of Mechanical Engineering,Taiyuan University of Technology,Shanxi Taiyuan 030024,China;Shanxi Key Laboratory of Fully Mechanized Coal Mining Equipment,Shanxi Taiyuan 030024,China)

机构地区：[1]太原理工大学机械工程学院,山西太原030024 [2]煤矿综采设备山西省重点实验室,山西太原030024

出　　处：《机械设计与制造》2018年第5期8-11,共4页Machinery Design & Manufacture

基　　金：国家自然科学基金资助项目-基于振动与电机电流信息融合的转子系统载荷识别及故障诊断方法(51475318);山西省青年科技研究基金资助项目-扭矩激励下转子系统磨损机理与试验研究(2014021024-2)

摘　　要：电机定子电流信号可以反应电机拖动系统中的扭矩波动情况,然而负载波动的特征频率易被工频所湮没。针对这一问题,提出小波阈值去噪、EEMD和互相关分析相结合的方法,该方法首先对电流信号进行小波阈值去噪的预处理以提高信噪比,然后对信号进行EEMD分解,利用互相关系数确定重构信号的IMF层数,以抑制工频及其谐波并突出特征频率。将该方法用于转子系统正弦扭矩激励的特征频率提取上,结果表明该方法能够有效的抑制工频及其谐波的干扰并提取负载扭矩的特征频率。The motor stator current signal can respond to the torque ripple of the motor drive system, however, the characteristic frequency of the load fluctuation is easy to be lost by the power frequency. To solve this problem,the method of wavelet threshold to noise,ensemble empirical mode decompositiona（EEMD）and cross-correlation analysis is proposed.At first,wavelet threshold denoising for current signals to improve signal to noise ratio.Then the EEMD decomposition of the signals and the mutual relations determine the number of the reconstructed signal the intrinsic mode functions（IMF）layers to suppress the power frequency and its harmonics and prominent characteristic frequency. The method is applied to the characteristic frequency extraction of sinusoidal torque excitation of rotor system. The results show that the method can effectively restrain the interference of power frequency and harmonic and extract the characteristic frequency of the load torque.

关 键 词：EEMD 互相关分析 电机电流信号 特征提取 转子系统 

分 类 号：TH16[机械工程—机械制造及自动化] TP206[自动化与计算机技术—检测技术与自动化装置]