Application Research On Fault Diagnosis Of Rolling Bearing With Variable Speed Based On Order Analysis And STFT Method

As an important rotating part in mechanical equipment,rolling bearing is also one of the parts of mechanical equipment prone to failure,so the research on bearing fault diagnosis has been a hot academic research,the fault diagnosis of special variable-speed and variableload is a difficult problem in this field.The traditional time-domain analysis and frequencydomain analysis methods are only used to analyze the signal under stationary condition,but not suitable for the signal under non-stationary condition.So in this paper,we study nonlinear non-stationary of order ratio analysis and short time Fourier transform(STFT)analysis method,using STFT method to reconstruct the signal of variable-speed signal,and then make through Angle resampling signals from the Angle of the nonlinear time domain into linear domain signals,according to the characteristic of order ratio spectrum distribution fault diagnosis research.Specific studies were carried out as follows:In this paper,the development background and significance of rolling bearing fault detection are described in detail,and the common rolling bearing faults and their diagnosis and analysis methods are studied.Aiming at the problem of bearing fault diagnosis with synchronous speed signal acquisition,order analysis method was adopted to analyze the fault signals.Firstly,according to the speed pulse,the vibration signal of the bearing within each rotating week is obtained.Then to Angle such as uniform sampling of the signal,because the signal is variable speed,make every turn in the signal length is differ,every Angle in order to ensure that all data have been collected,the need for each turn after the signal is linear fitting in the perspective of sampling,make time domain variable speed vibration signal can be converted into uniform Angle domain signal.Finally,FFT frequency analysis method is adopted to analyze the order ratio of the angular domain signals,and the bearing faults are diagnosed according to the characteristics of the order spectrum.Using the variable speed signal of the gearbox bearing of a locomotive to diagnose the fault of the outer ring of the rolling bearing accurately.Aiming at the problem of bearing fault diagnosis without speed signal,an order analysis technique based on speed tracking is proposed.Firstly,the time-frequency diagram is obtained by using the short-time Fourier transform(STFT),and the rotating frequency or the gear meshing frequency is calculated according to the parameters of the gear and the bearing and the speed range.Then the rotational speed matching tracking method is used to track the rotational frequency or meshing frequency according to the time-frequency diagram,and then the linear fitting and resampling are carried out to obtain the approximate rotational speed signal.Finally,equal Angle sampling and FFT spectrum analysis were carried out to get the order spectrum,and the bearing faults were diagnosed according to the characteristics of the order spectrum.The experiment verifies the fault diagnosis by the variable speed signal of the gearbox bearing of a locomotive and the results accurately diagnose the outer ring fault of the rolling bearing.The research content of this paper has reference significance for on-the-spot diagnosis of rolling bearing fault with and without speed signal.