自适应随机共振与ELMD在轴承故障诊断中的应用

The Application of Self-adaptive Stochastic Resonance and ELMD in Bearing Fault Diagnosis

针对随机共振(Stochastic resonance,SR)在处理轴承故障信号时需要满足小参数(信号频率、幅值、噪声强度远小于1)这一条件以及轴承故障特征难以提取的问题,提出基于自适应变尺度随机共振与总体局部均值分解(Ensemble local mean decomposition,ELMD)的轴承故障诊断方法。首先,对实测的信号按照一定的频率进行压缩,使其满足随机共振小参数的要求,然后,通过遗传算法(Genetic algorithm,GA)对变尺度随机共振双稳系统中的结构参数a,b进行优化,最后将随机共振输出信号进行ELMD分解,通过各PF分量的频谱图寻找轴承故障特征频率。对实测轴承故障信号的实验分析,结果表明本文提出的方法可有效地应用于轴承的故障诊断中。

Aiming at the problems that the stochastic resonance（ SR） processing the signal in the bearing fault diagnosis needs to meet the condition of small parameters（ frequency,amplitude and noise intensity are far less than 1） and the fault features of bearing were difficult to extract,a method of the self-adaptive re-scaling stochastic resonance based on genetic algorithm（ GA） and the ensemble local mean decomposition（ ELMD） was proposed.Firstly,the measured signal is compressed according to certain frequency,making it meet the requirements of small parameter of stochastic resonance. Secondly, the structure-parameter of the scale-transformation stochastic resonance system is optimized by genetic algorithm. Finally,the output signal of stochastic resonance is decomposed by ELMD,through each frequency spectrum of component of the PF,the character frequency of rolling bearing are looked for. Through experimental analysis of the measured bearing fault signal,the results show that the proposed method can be effectively used in bearing fault diagnosis.