Research On Fault Feature Enhancement And Extraction Of Support Bearing In Helicopter Tail Drive System

Helicopters have a wide range of applications in both military and civil areas,playing an important role.The transmission system is one of the three key moving parts of the helicopter.The tail transmission system is responsible for transmitting power to the tail reducer.As a key supporting component in the tail transmission system,the health status of the rolling bearing directly affects the overall reliability of the helicopter.Therefore,it is necessary to carry out fault diagnosis research on the support bearing of the helicopter tail transmission system to ensure the safe and stable flight of the helicopter and reduce the occurrence of accidents.This paper takes the support bearing in the helicopter tail transmission system as the research object,through the research on the key problems such as the fault generation mechanism and the modeling of vibration signal of the support bearing in the helicopter tail transmission system,the adaptive decomposition and fault component extraction of the fault vibration signal of the support bearing in the helicopter tail transmission system,the fault mode extraction of the support bearing in the helicopter tail transmission system based on the initial center frequency guidance,fault period detection and weak feature enhancement of support bearing in helicopter tail drive system,establishment of equivalent experimental system for helicopter tail drive system and experimental verification of diagnosis method,the fault diagnosis technology of the support bearing of the helicopter tail drive system is formed,which can provide reliable technical support for the safe operation of the helicopter.The main contents of this paper are as follows:(1)According to the structure and load-bearing characteristics of the support bearing of the helicopter tail drive system,the fault vibration generation mechanism and typical fault forms of the support bearing are studied.The fault vibration signal model of the rolling bearing considering the influence of sliding,alternating load and other factors is established,and the typical early impact fault characteristics are analyzed,which provides a theoretical basis for the targeted fault feature enhancement and extraction method of the support bearing.(2)In order to solve the problem that the early fault vibration signal of the support bearing of the helicopter tail transmission system has a small impact amplitude and is easy to be covered by noise,an adaptive signal decomposition and fault component extraction method is adopted to reduce noise interference and improve the signal-to-noise ratio.In order to solve the problems that the key parameters are difficult to determine and the target extraction is easy to be disturbed when the variational mode decomposition method decomposes the vibration signals,a variational mode decomposition method based on the optimization of cross-correntropy change rate is proposed.At the same time,the periodicity-assisted log-cycligram index is used to select the target mode.The experimental results show that,compared with the existing methods,the variational mode decomposition method based on cross-correntropy change rate optimization is simple in determining the number of decomposed modes and requires less calculation.The periodicity-assisted log-cycligram index is used to select the target fault mode,which can avoid the influence of random pulse interference and cyclostationary interference,and further enhance the reliability of the proposed method.(3)Considering the coupling and interaction between the vibration signal caused by the early local fault of the support bearing of the helicopter tail transmission system and other interference components,after analyzing the convergence characteristics of the initial center frequency of the variational mode extraction method,the fault mode extraction method based on the initial center frequency guidance is used to extract the fault signal components directly.The fault mode extraction method based on the initial center frequency guidance is divided into the variational mode decomposition method based on envelope curve fitting and the variational mode extraction method based on window fusion according to the way of extracting the initial center frequency.In the variational mode extraction method based on envelope curve fitting,the spectrum is divided by the constructed sliding window and envelope characteristics,and the initial center frequency of the mode is extracted.The relative amplitude ratio is used to optimize the balance parameters in order to obtain the optimal fault mode.In the variational mode extraction method based on window fusion,the log-cycligram index and correlation coefficient are used to construct the fusion window,which can accurately determine the initial center frequency.In addition,the balance parameter are directly determined according to the window bandwidth without special optimization process.Guided by the initial center frequency,the variational mode extraction method can directly extract the target mode and enhance the signal-to-noise ratio of the fault characteristic signal.The experimental results show that the variational mode extraction method based on the initial center frequency guidance does not need to consider the influence of other interference components while extracting the fault mode.(4)In order to solve the problem that the signal components of the faulty bearing in the tail drive system are affected by the transmission path and noise,a blind deconvolution method based on the maximum correntropy criterion is proposed to recover the periodic fault pulses.The maximum correntropy criterion blind deconvolution method uses the maximum correntropy criterion as the objective function to iteratively update the inverse filter coefficient vector,and uses the auto-correntropy spectrum to determine the fault pulse period to construct the objective vector,so as to improve the reliability of the deconvolution method.The experimental results show that the blind deconvolution method based on the maximum correntropy criterion has good robustness to noise,can effectively enhance the weak fault characteristics and achieve reliable fault diagnosis.(5)Based on the principle of equivalence,a test-bed for the support bearing of the helicopter tail transmission system is designed,and the proposed method is used to process and analyze the fault vibration signal.The results show that the proposed methods have good results in fault component enhancement and fault feature extraction,and can effectively diagnose the early rolling bearing faults on the helicopter tail drive system simulation test-bed.