基于MHSI-MCKD的滚动轴承故障特征提取方法

Rolling bearing fault feature extraction method based on MHSI-MCKD

为了解决滚动轴承的故障信号因受噪声干扰而导致故障特征提取困难,以及最大相关峭度解卷积(MCKD)的解卷积周期T、移位数M和滤波器长度L需要依赖人为主观经验选取的问题,提出了一种基于改进的谐波显著性指标(MHSI)与MCKD相结合的滚动轴承故障特征提取方法(MHSI-MCKD)。首先,提出了一种改进的谐波显著性指标(MHSI)作为评价MCKD降噪效果的依据;其次,以MHSI为适应度函数,采用步长搜索法得到MCKD的最优参数组合;最后,利用参数优化的MCKD对滚动轴承的故障信号进行降噪处理,并通过Hilbert包络谱获取滚动轴承的故障特征。在仿真和实验分析中,利用MHSI-MCKD方法实现了对MCKD关键参数的自适应寻优,增强了信号中故障特征的显著程度,提取出了滚动轴承内外圈的故障特征。仿真及实验结果表明,所提MHSI-MCKD方法可用于提取滚动轴承的故障特征。

To solve the problem that fault feature extraction of rolling bearing is difficult due to noise interference and that the deconvolution period T,shift number M and filter length L of maximum correlation kurtosis deconvolution(MCKD)need to be selected by subjective experience,a rolling bearing fault feature extraction method(MHSI-MCKD)based on the combination of modified harmonic significance index(MHSI)and MCKD was proposed.Firstly,a modified harmonic significance index(MHSI)was proposed to evaluate the noise reduction effect of MCKD.Secondly,MHSI was used as fitness function to obtain the optimal parameter combination of MCKD by step search method.Finally,the fault signal of rolling bearing was denoised by MCKD with optimized parameters,and the fault feature of rolling bearing was obtained by Hilbert envelope spectrum.In the simulation and experimental analysis,the MHSI-MCKD method was used to realize the adaptive optimization of the key parameters of MCKD,enhance the significance of the fault feature in the signal,and extract the fault feature of the inner and outer ring of the rolling bearing.Simulation and experimental results show that the proposed MHSI-MCKD method can be used to extract fault feature of rolling bearing.