基于自适应随机共振的轴承故障诊断研究

Research on Fault Diagnosis of Bearings with Adaptive Random Resonance

针对集总经验模式分解法很难作出非稳态条件的轴承故障诊断,提出一种自适应随机共振的轴承故障诊断方法.研究以非对称阱宽势函数取代对称双稳态势函数,借助左势阱宽变化以实现布朗粒子跃迁的有效控制;以非对称阱宽诱导随机共振系统输出信号的信噪比作为量子遗传算法的适应度函数,获得最优的非对称阱宽随机共振系统,实现轴承早期故障特征信号的增益和提取.仿真及轴承故障试验显示,研究可以增益和提取强背景噪声下的微弱故障特征频率信号,实现电机驱动端轴承的故障诊断,增益性能优于集总经验模式分解方法.

Aimed at the phenomenon that Ensemble Empirical Mode Decomposition(EEMD)is difficult to diagnose bearing faults under unsteady state conditions,an adaptive stochastic resonance(SR)method is proposed to diagnose the faults of bearings.In the proposed method,the symmetric bistable potential in classical SR methods were modified as an asymmetric well-width potential.Then,the parameters of SR with asymmetric well-width were optimized by using quantum genetic algorithms to obtain an optimal SR system,thereby extracting mechanical fault characteristics.Simulation and experimental data show that the proposed method can not only extract the weak impulsive characteristics overwhelmed by heavy background noise,but also extract the incipient fault characteristics of rolling element bearings.In addition,the comparison results suggest that the proposed SR method is superior to EEMD.