Research On Fault Diagnosis Method Of Rotating Machinery Based On HVD And Time-delayed Stochastic Resonance

Rotating machinery structure is widely used in various mechanical equipment.Gears and bearings are important parts of rotating mechanism,and they are also the most vulnerable parts to failure.Early detection and diagnosis of fault types play an important role in prolonging the life of equipment,improving work efficiency and improving economic efficiency.Therefore,it is particularly important to find an effective fault diagnosis method.In this paper,a mechanical fault diagnosis method based on Hilbert Vibration Decomposition(HVD)and asymmetric bistable stochastic resonance with time-delayed feedback is proposed and the effectiveness of this method in bearing fault diagnosis is verified by an example study.Firstly,the basic theory of stochastic resonance(SR)is studied.The system is improved by adding time-delay feedback to the stochastic resonance system.The stationary probability density function(PDF),average traversal time(MFPT)and signal-to-noise ratio(SNR)of asymmetric bistable system with time-delay driven by Gauss white noise are derived analytically.By adjusting the parameters of asymmetry,delay time,delay intensity and noise intensity,the influence degree of the parameters on the system is analyzed,and the best matching parameters are found to obtain the strongest signal-to-noise ratio,which provides a method to improve the useful signal intensity for the signal processing of rotating machinery system and makes the fault diagnosis of the system more accurate.In view of the limitation that stochastic resonance can only detect low-frequency signals,the variable-scale stochastic resonance(SSR)is studied.On this basis,a variable-scale stochastic resonance(TASSR)algorithm for asymmetric bistable systems with time-delay term is deduced and the high-frequency signals of DDS multi-stage gear test-bed are detected by this method.Then,the advantages of Hilbert vibration decomposition processing signal are studied.By removing the end-point effect,the Hilbert vibration decomposition is improved so that the vibration signal of rotating machinery system is decomposed moreeffectively and the decomposition results are more accurate,which lays a foundation for the later processing of rotating machinery system signal.Finally,aiming at bearing fault system,a fault diagnosis method based on Hilbert vibration decomposition and stochastic resonance is proposed.The actual working condition signal is decomposed into multiple components by Hilbert vibration decomposition method.The fault signal components are selected to envelope by kurtosis as reference.Then the envelope signal is strengthened by TSSSR algorithm,and the fault frequency in the decomposition is obtained to identify faults.