Research On Weak Signal Detection Based On Coupled,Combined And Asymmetric Bistable Stochastic Resonance System

With the in-depth study of weak signal detection technology,weak signal detection technology based on stochastic resonance is widely applied to various fields of life.Stochastic resonance is an innovative weak signal detection method.Most of the traditional weak signal detection techniques focus on how to reduce the noise intensity effectively and then the output signal-to-noise ratio of the system can be improved.However,the stochastic resonance method can enhance the energy of weak signals by utilizing the partial energy of noises transformed to weak signals after being processed by the nonlinear system,and finally improve the output signal-to-noise ratio of the systemFirstly,the theoretical knowledge involved in the existing research results is described in this thesis.Secondly,the stochastic resonance phenomena of different nonlinear systems are studied on the basis of the theoretical knowledge.Finally,the application of different nonlinear stochastic resonance systems in weak signal detection is also studied.The main work contents of the thesis are as follows:(1)The weak signal detection technique for nonlinear coupled bistable stochastic resonance systems under ?-noise is studied.Firstly,the method of generating ?-noise is introduced and the influence of noise parameters on the probability density function of noise is analyzed.Secondly,by setting the SNR improvement as the index,the influence of system parameters and noise intensity on signal-to-noise ratio improvement is studied,respectively.Finally,the single-frequency and multi-frequency weak signals are detected respectively and the corresponding experimental results verify that the system can effectively detect the fault signal.(2)The new asymmetric bistable system is analyzed and the weak signal detection technology under asymmetric bistable stochastic resonance system is studied.Firstly,under the condition of adiabatic approximation,the expression of the output signal-to-noise ratio of the asymmetric bistable stochastic resonance system under the action of additive and multiplicative Gaussian white noise is derived in the thesis.Secondly,the effects of system parameters,periodic signal amplitude,additive noise intensity and multiplicative noise intensity on signal-to-noise ratio are analyzed respectively.Then,in practical engineering applications,the stochastic resonance based on asymmetric bistable system is applied to the bearing fault diagnosis using the adaptive algorithm.Finally,the corresponding experimental results verify that the system can effectively detect the fault signal.(3)Combined with the harmonic potential well and the Gaussian potential well,the weak signal detection method under the second-order underdamped joint bistable stochastic resonance system is studied.Firstly,the harmonic potential well,the Gaussian potential well,and the joint bistable potential well are analyzed respectively.Secondly,under the condition of adiabatic approximatio,the expression of SNR of second-order underdamped joint bistable stochastic resonance is derived and the influence of system parameters on the output SNR is studied respectively.Finally,the system is applied to the simulated fault signal and the practical bearing fault signal detection.The corresponding experimental results verify that the system can effectively detect the fault signal.