Research on fault detection of asymmetric piecewise well-posed stochastic resonance system

Abstract

Stochastic resonance of an asymmetric piecewise well-posed system driven by a periodic forcing and Gaussian white noise is investigated. Aiming at the problem that the output saturation of the classical stochastic resonance (CSR) system needs to be further improved, the dimensionality of the quartic function is reduced to a quadratic function, and the well position of the function becomes asymmetric. First, the potential function and mean first passage time are analyzed, and then the signal to noise ratio formula of the system is derived through adiabatic approximation theory. Second, the system is simulated and tested. Theoretical analysis and numerical simulation show that the system in a well-posed symmetric case has better performance than the CSR system, but is better in a well-posed asymmetric case. Finally, the bearing fault detection is processed by using the proposed system. The results show that the fault frequency can be more accurately identified by the well-posed asymmetry, and the energy of the characteristic signal can be improved further. The theoretical basis and reference value of the system are provided for further application in practical engineering testing.

High-Performance Adaptive Weak Fault Diagnosis Based on the Global Parameter Optimization Model of a Cascaded Stochastic Resonance System

“…Thus, the low-frequency weak signal can be significantly enhanced while filtering out the high-frequency noise, and the output signal-to-noise ratio (SNR) is significantly improved [10,11]. Due to the advantages of SR in weak signal detection, SR has attracted continuous attention in the field of weak fault diagnosis in the past 40 years [12][13][14][15][16][17].…”

Research and application of a novel piecewise unsaturated asymmetric tristable stochastic resonance system

“…Cheng et al [23] studied an adaptive smooth unsaturated bistable SR system for bearing fault detection, which effectively suppressed the interference of low-frequency and high-frequency components. In addition, Zhang et al [24] found in the study of asymmetric system that after adding asymmetric factor, the system performance shows better advantages than symmetric system with the change of symmetry of potential well structure. Liu et al [25] studied the SR in an asymmetric bistable kinetic model driven by both additive non-Gaussian colored noise and a harmonic excitation, and concluded that the maximum SNR in asymmetry can exceed the symmetric one.…”