Adaptive monostable stochastic resonance based on PSO with application in impact signal detection

Based on the characteristics of the resonance in single potential well of impact signal and the reasonable selection of parameters lacking of effective theoretical basis and the deficiency of single parameter optimization in traditional adaptive stochastic resonance, a new adaptive monostable stochastic resonance based on particle swarm optimization (PSO), which can realize multi-parameter synchronous optimization, is proposed. The model of monostable stochastic resonance that avoids the resonance transition between potential wells in classical bistable system is applied and the weighted kurtosis index of the output of the monostable system is determined as the fitness function of PSO algorithm and multi-parameter in stochastic resonance are selected adaptively. As a result, the impact components in original signals are detected effectively. The algorithm procedure of the impact with adaptive stochastic resonance, combining PSO with the scale transformation stochastic resonance, is provided. The proposed method is applied to the simulation of impact signal and the detection of engineering impact signal, and the results show that the presented method can raise the signal-to-noise ratio effectively, which has advantages of simplicity and fast convergence speed, and possesses a good prospect of engineering application.