The Online Condition Monitoring For Great Rotating Machine

The method of analysis, examination of vibration signal, which is widely used in many fields, is the base of vibration monitoring and diagnosis. As the complexity of the mechanical structure, traditional signal processing methods are not suitable for detailed analysis of high frequency and non-stationary signal analysis. It is necessary to introduce new signal processing method, which can extract fault signal effectively.The paper has introduced the method of harmonic wavelet, analyzed the method of harmonic wavelet package, and provided the procedure of harmonic wavelet and harmonic wavelet package. On this basis, the theory of harmonic wavelet and harmonic wavelet package is used to analyze some simulated signal; simulation results show that this method presented can effectively extract the fault signal in the non-stationary signal and has high precision.Taking the characteristics of rolling bearing fault signal into account, the method of applying harmonic wavelet package to detect rolling bearing fault. At first, the paper analyzes the main failure modes and vibration mechanism of rolling bearings, and introduces some common method of the rolling bearing's fault diagnosis. Second, the method of harmonic wavelet package is applied to rolling bearing's fault diagnosis for inner raceway, outer raceway and the ball, the results show that this method can effectively detect rolling bearing's fault. Finally, the method of harmonic wavelet package is applied to monitor the condition of mill, experimental results show that this method can accurately monitor the operational status of mill and has a high practical value in the actual project application.