Detection of weak signals in second-order composite multi-stable matching stochastic resonance

First-order systems are susceptible to low-frequency noise interference, often necessitating the use of external filters. Second-order multi-stable stochastic resonance (SR) leverages the secondary utilization of noise energy to effectively transfer high-frequency noise energy to the lower frequency range, significantly improving the output signal-to-noise ratio (SNR). Therefore, this paper proposes a new second-order composite multi-stable matching stochastic resonance (SCMMSR) system. By investigating the SR characteristics of the SCMMSR system, this study analyzes the influence of its parameter changes on the signal-to-noise ratio improvement. This paper presents the adaptive detection of periodic and non-periodic signals in Gaussian noise. The SCMMSR system demonstrated superior detection performance compared to the first-order composite multi-stable stochastic resonance system. The SR system based on the new model was tested on actual bearing fault data, demonstrating its ability to detect faults in the inner race, outer race, rolling elements, and cage. The proposed method exhibits superior detection performance compared to the traditional second-order tristable stochastic resonance. This work extends the application of SR in engineering practice and verifies the effectiveness and practicality of the proposed SCMMSR method.