Research On Weak Fault Diagnosis Method Based On Stochastic Resonance In Strong Noise Background

In this paper,we study the weak fault diagnosis method based on Stochastic resonance(SR).The scholars have proposed the method of using SR to solve the mechanical weak fault,but these methods are inadequate and can improve and perfect.According to the shortcomings of the current SR research,and put forward the corresponding solution.Finally,the proposed method is applied to the weak signal fault diagnosis of mechanical equipment.The main contents of this paper include the following aspects:(1)Aiming at the problem that the traditional algorithm is not enough to filter the noise and the weak signal is difficult to be extracted,the influence of the multi-component average SR on the weak signal enhancement effect is studied.The weak fault diagnosis method of multi-component average SR is proposed,the effective EEMD components in order to achieve the purpose of reducing noise,and then input to the SR system and the overall average of the effective components,so as to achieve the purpose of extracting weak fault characteristics of the signal.(2)The optimization of the potential parameters of the classical SR system is easy to fall into the local optimization problem.Based on the traversal and randomness of the chaotic motion,the chaos algorithm is used to solve the chaos optimization to achieve global optimization of parameter matching.(3)In order to study the influence of the SR model of the constrained potential on the detection of weak signals,a new method of adaptive constrained potential bistable SR is proposed to analyze the characteristics of potential wide and potential height adjustment of the classical bistable SR system.Constrained potential SR system model,and describes in detail the function and structure of its potential function,and applied to the early failure of the bearing and the experimental verification of planetary gearbox.(4)Aiming at the problem of parameter matching optimization of classical bistable SR potential model,the influence of piecewise nonlinear potential SR on the extraction of weak signal is studied.A new SR potential function with piecewise nonlinear single parameter optimization is proposed Model,compared with the traditional classical stochastic resonance,and applied to the early failure test of bearing.(5)In view of the saturation characteristics of classical bistable SR,the influence of SR of piecewise linear potential function on the extraction of weak signal is studied with the increase of the input signal,and a new piecewise linear unsaturated SR potential model,in order to improve the weak signal enhancement effect.(6)For the symmetry of the potential model of the traditional SR,the potential model can not achieve the structural optimal problem,and the influence of the asymmetric potential SR model on the weak signal extraction is proposed.An asymmetric potential stochastic resonance model is proposed to realize that the structure is optimized,and the model is compared with the traditional classical SR and applied to the bearing failure test.In summary,this paper studies the application of weak fault diagnosis method of SR under strong background noise,the proposed method is compared with the traditional method,the practicability and superiority of the method have been proved effective in the treatment of mechanical fault signal.