Utilizing shaft-speed information to analyse vibration signals is an important method for fault diagnosis and condition monitoring of rotating machineries,especially for those running at variable speeds.However,in man...Utilizing shaft-speed information to analyse vibration signals is an important method for fault diagnosis and condition monitoring of rotating machineries,especially for those running at variable speeds.However,in many cases,shaft-speed information is not always available,for a variety of reasons.Fortunately,in most of the measurements,the shaft-speed information is embedded in the vibration response in many different forms,such as in the format of the fundamental shaft-rotation-frequency response and its harmonics,and the gear-meshing-frequency response and its harmonics,etc.Proper signal processing can be used to extract the shaft instantaneous speed from the measured vibration responses.In existing instantaneous shaft-speed-identification methods,a narrow-bandpass filtering technique is used explicitly or implicitly.In a complex gearbox system,such as that used in a wind turbine,the gear-meshing-response component could be modulated by many other shaft speeds,due to the configuration of the gearbox or due to the existence of component damage.As a result,it is very difficult to isolate a single vibration-response component for instantaneous shaft-speed detection.In this paper,an innovative approach is presented.The instantaneous shaft speed is extracted based on maxima tracking from the vibration-response spectrogram.A numerical integration scheme is employed to obtain the shaft instantaneous phase.Digital-domain synchronous resampling is then applied to the vibration data by using the instantaneous phase information.Due to the nature of noise suppression in the numerical integration,the accuracy of synchronous sampling is greatly improved.This proposed approach demonstrates the feasibility and engineering applicability through a controlled laboratory test case and two field wind-turbine cases.More detailed results and conclusions of this research are presented at the end of this paper.展开更多
文摘Utilizing shaft-speed information to analyse vibration signals is an important method for fault diagnosis and condition monitoring of rotating machineries,especially for those running at variable speeds.However,in many cases,shaft-speed information is not always available,for a variety of reasons.Fortunately,in most of the measurements,the shaft-speed information is embedded in the vibration response in many different forms,such as in the format of the fundamental shaft-rotation-frequency response and its harmonics,and the gear-meshing-frequency response and its harmonics,etc.Proper signal processing can be used to extract the shaft instantaneous speed from the measured vibration responses.In existing instantaneous shaft-speed-identification methods,a narrow-bandpass filtering technique is used explicitly or implicitly.In a complex gearbox system,such as that used in a wind turbine,the gear-meshing-response component could be modulated by many other shaft speeds,due to the configuration of the gearbox or due to the existence of component damage.As a result,it is very difficult to isolate a single vibration-response component for instantaneous shaft-speed detection.In this paper,an innovative approach is presented.The instantaneous shaft speed is extracted based on maxima tracking from the vibration-response spectrogram.A numerical integration scheme is employed to obtain the shaft instantaneous phase.Digital-domain synchronous resampling is then applied to the vibration data by using the instantaneous phase information.Due to the nature of noise suppression in the numerical integration,the accuracy of synchronous sampling is greatly improved.This proposed approach demonstrates the feasibility and engineering applicability through a controlled laboratory test case and two field wind-turbine cases.More detailed results and conclusions of this research are presented at the end of this paper.
