원문정보
초록
영어
The rolling bearing vibration is a complex, non-stationary and dynamic process. The vibration signal can not be described through a fixed time or frequency function. This increases the difficulty of fault diagnosis. Therefore, this paper proposed a rolling bearing fault diagnosis method based on local mean decomposition (LMD) and time-frequency analysis. Firstly, we need to draw time-domain waveform and power spectrum of vibration signal in this method. Secondly, we can gain the energy density distribution of signal using directly time-frequency analysis of signal, the time-frequency analysis of intrinsic mode function (IMF) and the time-frequency analysis of production function (PF). We gain the IMF by decomposing the original signal by the ensemble empirical mode decomposition (EEMD).and after the original signal is decomposed by the local mean decomposition,we gain the PF .Finally, according to the energy density distribution of signal, diagnose the fault of the rolling bearing. The simulation results show that we can gain the clearer the energy density distribution using the time-frequency of PF of signal to diagnose fault and improve the reliability of the fault diagnosis
목차
1. Introduction
2. The Decomposition Principle of EEMD and LMD
2.1. The decomposition principle of EEMD
2.2. The decomposition principle of LMD
3. Time-frequency Analysis
3.1. Short-time Fourier Transform
3.2. Cohen time-frequency distribution
4. Experimental Simulation Demonstration
4.1. The vibration signal analysis of the normal rolling bearing
4.2. The vibration signal analysis of the inner ring fault
4.3. The vibration signal analysis of the rolling element fault
4.4. The vibration signal analysis of the outer ring fault
5. Conclusions
Acknowledgements
References