A Novel Type of Stochastic Resonance Potential Well Model and Its Application

Shangbin Jiao,Wei Jiang,Qing Zhang,Weichao Huang,Shuang Lei

doi:10.1109/access.2019.2950692

Shangbin Jiao, Wei Jiang + Show 3 more

Open Access

PDF Available

https://doi.org/10.1109/access.2019.2950692

Copy DOI

Export

Save

Cite

Journal: IEEE Access	Publication Date: Jan 1, 2019
Citations: 17	License type: CC BY 4.0

Affiliation: Xi'an University of Technology

Abstract
Highlights/Summary
Full-Text PDF
Similar Papers

Abstract

Listen

For a stochastic resonance system, the characteristics of the nonlinear model have an important influence on the output. To further improve the enhanced detection of stochastic resonance, a novel potential well stochastic resonance model is constructed to simultaneously solve the problem of the output saturation and high barrier in the classical bistable model. Analytical expressions of the Kramers rate and output signalto-noise ratio are presented, and weak signal detection is theoretically analyzed. The performance of the system based on the novel potential well model is simulated and analyzed. Finally, the proposed model is used for the detection of multiple high-frequency weak signals in an α-stable noise environment and a practical bearing fault signal. The simulation and experimental results demonstrate that the output of the stochastic resonance system proposed in this paper exhibits a large output signal-to-noise ratio and a high spectral peak at the characteristic frequency.

Highlights

Stochastic resonance (SR) was proposed by Benzi in 1981 to study the problem of Earth’s palaeometeorological glaciers, which well explained the periodic alternations of ice ages and warm climate periods [1], [2]
The results indicate that the novel potential well model performs better than the classical bistable model in detecting the inner race fault signal, and the results verify the effectiveness of the novel potential well SR (NPWSR) model in bearing fault diagnosis
The results indicate that the novel potential well model performs better in detecting the outer race fault signal, and the results verify the effectiveness of the proposed model in bearing fault diagnosis

Summary

INTRODUCTION

Stochastic resonance (SR) was proposed by Benzi in 1981 to study the problem of Earth’s palaeometeorological glaciers, which well explained the periodic alternations of ice ages and warm climate periods [1], [2]. The novel potential well model consists of a piecewise bistable potential model and an applied signal SR model, which can solve the problems of the output saturation and high barrier in the classical bistable system.

OUTPUT MEASUREMENT INDEX AND DETECTION SCHEME

OUTPUT SNR GAIN

Findings

CONCLUSION