解调同步压缩变换及其应用于变转速滚动轴承故障诊断

Demodulation Synchrosqueezing Transform and Its Application to Variable Speed Rolling Bearing Fault Diagnosis

为实现对变转速下滚动轴承的故障诊断,提升设备服役的安全性,在传统基于短时傅里叶变换(STFT)的同步压缩变换(FSST)的理论框架上,针对STFT存在的调制成分造成其瞬时频率(IF)估计不准确、时频模糊的问题,提出一种解调短时傅里叶变换(DSTFT),即在STFT中引入解调算子消除调制对信号的影响,获得更加准确的IF估计值。在此基础上,推导基于DSTFT的解调同步压缩变换(DSST),可以通过时频平面变换系数的重排获得能量高度集中的时频表达。所提出的DSST方法具有IF估计准确、时频脊线能量集中性好等优点。通过数值模拟分析和试验台振动信号分析,验证所提方法在变转速滚动轴承故障诊断中的有效性。

In order to diagnose the fault types of rolling bearings under variable speed conditions and improve the safety of equipment,by using the theoretical framework of synchrosqueezing transform(FSST)based on short time Fourier transform(STFT),a demodulated short time Fourier transform(DSTFT)was proposed.It was designed to solve the problems of inaccurate estimation of instantaneous frequency(IF)and time-frequency ambiguity caused by modulation component of STFT.A demodulation operator was introduced into STFT to eliminate the influence of modulation on the signal,and the more accurate IF estimation was obtained.On this basis,demodulation synchrosqueezing transform(DSST)based on DSTFT was deduced,by which highly concentrated time-frequency representations can be obtained through the rearrangement of time-frequency transformation coefficients.The proposed DSST method has the advantages of accurate IF estimation and good energy concentration of time-frequency ridge.The effectiveness of the proposed method in fault diagnosis of variable speed rolling bearings is verified by using numerical simulation analysis and vibration signal analysis of experimental bench.