离散频谱校正技术在感应电机转子故障诊断中的应用  

Application of discrete spectrum correction technology to induction machine rotor fault diagnosis

作　　者：李彧[1] 侯新国[1] 胡文彪[1] 卜乐平[1] 

机构地区：[1]海军工程大学电气工程学院,武汉430033

出　　处：《海军工程大学学报》2013年第4期41-46,共6页Journal of Naval University of Engineering

基　　金：国家自然科学基金资助项目(50677069)

摘　　要：基于坐标变换的感应电机故障特征提取方法可将基波分量转化为直流量,抑制基波分量泄漏的影响。但若不考虑故障特征频率分量自身的泄漏,经频谱分析得到的故障特征频率分量幅值是不准确的。为此,提出了一种基于频谱校正技术的感应电机转子故障诊断方法。该方法对传统的相位差校正法加以改进,使其在幅值校正时不再依赖窗函数的谱函数,然后选择合适的窗函数对电机定子电流信号加窗处理,减小其各频率成分之间的干涉。该方法不仅能够检测出电流信号中是否存在故障特征频率分量,还能够获取其准确的频率和幅值,可对故障程度的量化提供算法支持。仿真和实验结果证明:采用该方法计算得到的故障特征频率分量的频率和幅值具有较高的精度。The method of extracting fault characteristics of the induction machine according to coordi- nate transformation can restrain the fundamental component leakage by transforming fundamental component into DC component. The amplitude of fault characteristic frequency component cannot be inaccurate by use of frequency analysis without its own leakage taken no account. For this reason, the method is proposed for the fault diagnosis of induction machine rotor based on discrete spectrum cor- rection technology. This method is an improvement on the traditional phase difference correction method, making its amplitude not to depend on spectral function of the window function. The selec- tion of an appropriate window function for stator current signal analysis can reduce the interference between individual frequencies. The proposed method can detect the existence of fault characteristic frequency component and its frequency and amplitude. The simulation and experiment results show that this method can be used to meet the higher precision of frequency and amplitude.

关 键 词：感应电机 故障诊断 转子故障 频谱校正 

分 类 号：TN991.6[电子电信—信号与信息处理] TM711[电子电信—信息与通信工程]