Engineering application study of twice sampling stochastic resonance

作者：Leng Yonggang; Wang Taiyong; Guo Yan; Xu Junyan

来源：China Mechanical Engineering, 2004, 15(20): 1847-1852.

The Kramers escape rate and the characteristics of noise frequency spectrum distributing as Lorentzian form were investigated under the theory of small parameter stochastic resonance (SR), and the viewpoint of creating stochastic resonance only in the low frequency region where the noise energy has been concentrated was also obtained. On the basis of the investigation and through the methodology of twice sampling frequency transformation, the small parameter stochastic resonance is successfully expanded into the applications of the large parameter stochastic resonance, and hence a weak signal submerged in strong noise is detected. The further quantitative analysis of the power spectral characteristics of the large parameter stochastic resonance and its signal-to-noise ratio indicate that the increase of sampling frequency can move the weak signal into the low frequency region concentrated by noise energy, which is benefit to produce a distinguishable SR spectral peak. The real example of monitoring and diagnosis of electromotor faults proves the validity of the technique in practical applications.

• 出版日期2004