Study On The Weak Fault Signal Detection Method Of Machinery Equipment Based On Stochastic Resonance Theory

The application scope of mechanical equipment as a core equipment in the field of industrial is more and more widely.The real-time detection and fault diagnosis of mechanical equipment is of great significance to ensure production safety,avoid the occurrence of production accidents and economic losses.And in the process of realization of mechanical fault diagnosis,the key technology is detection and extraction of mechanical fault characteristic signal.This paper studies the method of weak signal detection based on the theory of stochastic resonance and analyzes its application value in the field of mechanical equipment weak fault diagnosis.Aiming at the limitation of the theory of stochastic resonance,the cascaded multi-stable stochastic resonance system is established,and its application characteristics and advantages are studied in the paper.This paper focuses on the empirical mode decomposition under strong background noise,the theory of enhanced cascade stochastic resonance system and the detection method of multi-scale cascaded stochastic resonance.Based on the basic theory of stochastic resonance,a multi-stable stochastic resonance model is studied,and the origin of the Langevin equation is briefly introduced and the results are obtained by the Fokker-Planck equation.The multi-stable stochastic resonance method achieves detection of any frequency signal by the secondary sampling method,and a fault detection method of cascaded multi-stable stochastic resonance system is proposed which can realize the extraction of weak characteristic signal.Both the simulation and example analysis show that the cascaded multi-stable stochastic resonance method has higher application value and can efficiently extract the early fault signal of mechanical equipment.This paper presents a method of fault characteristic signal extraction based on reinforcement cascaded multi-stable stochastic resonance system,and studies the change of the stochastic resonance system induced by adding the second driven periodic signal.The condition that the second driven periodic signal is effectively controlled the stochastic resonance is analyzed.The example shows that the method can effectively extract theweak fault signal and has a good application prospect in the field of signal detection.An empirical mode decomposition method based on cascade multi-stable stochastic resonance system is proposed for the EMD problem in strong noise background.It can effectively remove high-frequency noise,reduce the number of EMD layers,and makes the physical meaning of EMD more specific.Finally,a fault diagnosis example shows that the method reduces the number of IMF,improves the efficiency of the operation,and can accurately detect the fault characteristic frequency.Aiming at the problem that it is difficult to detect with low signal-to-noise ratio,a method of multi-scale cascaded stochastic resonance is proposed.In this method,the noise signal is decomposed into multiple signals of different scale frequencies by wavelet transform,and the reconstructed signals consisting of an approximate signal and each detail noise are used as as the input signals of the cascade stochastic resonance systems.Finally,the average value of the output signal from each cascaded multi-stable system is the output result of the whole system.Then the method is applied to the case of mechanical fault diagnosis.The results show that this method can effectively detect the weak signal and improve the output SNR,and has obvious superiority compared with other methods.