论文部分内容阅读
基于固有时间尺度分解(intrinsic time-scale decomposition,ITD)方法的线性变换和三次样条插值,提出一种改进的固有时间尺度分解方法(improved intrinsic time-scale decomposition,IITD),将IITD方法和谱峭度(spectrum kurtosis,SK)相结合,实现轴承故障的智能诊断.首先采用改进ITD方法对采集的轴承振动信号进行分解,得到若干个固有旋转分量(proper rotation component,PRC),然后利用谱峭度法对相关性最大的PRC进行滤波处理,最后对滤波后的PRC进行Hilbert包络解调来提取故障特征频率,从而识别轴承故障类型.仿真和实验分析结果表明:该文所提IITD-SK法可成功提取出故障特征频率,实现轴承故障的有效诊断,与传统的傅里叶变换、包络谱分析以及EMD方法的结果相比,该方法诊断效果更佳.“,”The fault feature of bearing is difficult to extract.In order to solve this problem,a new method of fault diagnosis based on improved intrinsic time-scale decomposition (IITD) and spectrum kurtosis (SK) was proposed.Firstly,the IITD algorithm is utilized to decompose the bearing vibration signal into several proper rotation components (PRCs).Then spectral kurtosis is used to on filter the PRC of the largest correlation coefficient.Finally the Hilbert envelope demodulation is applied to extract fault characteristic frequency and identify the fault types of bearings.Simulation and experimental analysis results showed that the IITD-SK method can successfully extract the fault feature frequency and effectively diagnose bearing fault.Compared with traditional Fourier transform,envelope spectral analysis and EMD,the proposed method is better.