Research On The Mechanism And Methodology Of Stochastic Resonance Signal Recovery

Stochastic resonance (SR) enhances the nonlinear system behavior with the assistance of noise, including the sensitivity and selectivity of the response to the exterior stimulus. The energy-transfer mechanism makes the weak information revealed in the output spectrum, while the time-waveform is distorted. The investigation on the signal recovery in bistable SR was carried out. It helps the SR technique develop and spread in the engineering applications.Noise, bistable system and input signal are the three essential factors in SR. The noise-induced SR method, the parameter-tuning SR method, and the twice sampling SR method change the characteristics of the noise, the bistable system and the input signal, respectively. With the new cooperation, they can all produce the SR phenomena when the system exceeds the small-parameter area. If treating the strong noise, the parameter-tuning SR method is better than the noise-induced SR method. If the input frequency is large, the twice sampling SR method is the optimal choice. Moreover, the actions of the bistable system parameters can build the system behavior in an orderly way, together with the twice sampling frequency.The distortion analysis was made both from the particle's dynamics and the signal processing. The reasons causing deviation in the output were presented. The function of the recovery system was given. By the discussion on the signal recovery courses of the sinusoidal signals with different amplitude, the pulse distortion appearing the recovery waveform caused by the particle's transitions at the bistable potential'inflexions was explained. A method to process the subthreshold SR output via cascaded-bistable system and choosing the recovery system's parameters was put forward.In terms of the spectrum feature, the signal is classified as the sinusoid-alike signal or the pulse-alike signal. The cascaded-bistable SR or mono-stable SR was introduced to process different types of signals. The final recovery signal is just the suggested recovery system's response to the SR output. Meanwhile, the recovery system is optional, as parameter-tuned or parameter-fixed. We define the series of methods mentioned above as the SR recovery methodology based on the signal classification. With the association of the parameter-tuning SR and the twice sampling SR methods, it can make the large-parameter SR recovery come true. Since the methodology requires no average processing, it is applicable to a single sample with the limited length. The numerical simulations present that it can recover the waveform containing weak information in time domain effectively. Because the optimal SR state is not required strongly in the pulse-alike signal recovery, the system parameters can be tuned in a wider range than the traditional SR methods.According to the processing targets and the engineering conditions, the bistable SR technique and the recovery methodology were introduced to analyze the vibration of the flowmeter system and the metal cutting course. The applications have presented the practicability and effectiveness of the SR techniques, especially in the feature detection and fault diagnosis.