Research On Transmission Weak Fault Diagnosis At Variable Speed

Rolling bearing and gear are the important components of automobile transmission.Therefore,in transmission fault diagnosis,gear and bearing faults are often taken as the key research objects.Due to the complex structure and many parts of the transmission,when the transmission fails,the fault feature information in many parts will be disturbed by strong noise,so it is difficult to extract and identify fault features effectively.Moreover,in practical engineering application,the automobile transmission usually operates under the unsteady variable speed condition,so it is of great significance to study the fault diagnosis method of the automobile transmission under the unsteady variable speed condition.In recent years,the research on fault diagnosis method of rotating machinery equipment under the condition of unsteady rotating speed has become one of the key targets of researchers.This paper takes the automobile transmission as the research object,analyzes the gear fault of the automobile transmission under the unsteady variable speed condition,studies the fault feature extraction of the automobile transmission under the unsteady variable speed condition.The main research contents of this paper are as follows:(1)By analyzing the fault type,vibration mechanism and vibration characteristics of the gear and bearing in the automobile transmission,the vibration signal characteristics in the time domain,frequency domain and fault characteristic order domain of the gear and rolling bearing under the condition of variable speed are respectively analyzed through simulation.Based on this,the vibration data of acceleration sensor is analyzed by using the test bench of automobile transmission fault,taking time-domain characteristic value and sample entropy analysis as the main part,so as to determine the best sensor measurement point.The fitting method of speed curve based on photoelectric speed sensor is analyzed to provide reliable experimental data for fault diagnosis method of variable speed.(2)The basic theory of multipoint optimal minimum entropy deconvolution(MOMEDA)and maximum second-order cyclostationary blind deconvolution(CYCBD)and the influence of filter length on the noise reduction effect of these two fault diagnosis methods were studied in depth.On this basis,fault feature extraction methods based on computed order tracking(COT)and MOMEDA,CYCBD and time-angle(A-T)spectrum were proposed respectively.The fault feature extraction effects of the two methods were studied through experiments and fault simulation signals.Aiming at the "spectrum ambiguity" problem in the frequency domain under the condition of variable speed,the COT method is used to transform the non-stationary time domain signal into stationary angular domain signal,and then the stationary diagonal domain signal is used to reduce noise,so as to extract the fault characteristics of automobile transmission.However,this method requires angular resampling of vibration signals,which will destroy the intrinsic time characteristics of original vibration signals and cannot effectively guarantee the extraction effect of fault characteristic order.The components of the variable speed vibration signal are relatively complex,and the background noise is large,so the direct application of A-T spectrum method to extract fault features will not achieve the ideal noise reduction effect.In order to clearly identify fault characteristics under the condition of variable speed,this paper USES the method combining CYCBD and A-T spectrum to realize the extraction of fault characteristics under variable speed.