采用粒子群算法的自适应变步长随机共振研究

Self-adaptive step-changed stochastic resonance using particle swarm optimization

针对传统的自适应随机共振只能实现单参数优化和变步长随机共振计算步长选取困难的缺陷,提出一种基于粒子群优化算法(Particle Swarm Optimization,PSO)的自适应变步长随机共振方法,实现了变步长随机共振最优输出的自适应求解。该方法以双稳系统的输出信噪比作为粒子群算法的适应度函数,通过变步长随机共振系统的结构参数和计算步长的自适应同步选取,能够最优地检测出大参数条件下的微弱信号。仿真数据和工程实际数据的分析表明,该方法简单易行,适用范围广,收敛速度快,能有效的检测出强噪声背景下的高频微弱信号,具有良好的工程应用前景。

The traditional adaptive stochastic resonance (SR) can only realize one-parameter optimization, and it is very difficult to select the computing step of step-changed stochastic resonance (SCSR). A new adaptive SCSR based on particle swarm optimization (PSO) was proposed here to realize the adaptive solving of optimal output of SCSR. The output signal to noise ratio of a bi-stable system was taken as a fitness function of PSO algorithm, the structural parameters and calculation step of SCSR were selected adaptively, as a result, the weak signal under conditions of large parameters was detected optimally. With the proposed method, simulation data and vibration signals measured on defective bearings with inner race fault were analyzed. The results showed that the proposed method has advantages of simplicity, fast convergence speed and widely applicable range, and possesses a good prospect of engineering application.