基于改进快速峭度图的轴承故障诊断方法研究

Bearing fault diagnosis method based on improved fast kurtosis graph

针对传统快速峭度图在轴承故障诊断中,通过计算滤波信号的包络峭度确定频带范围容易受到随机脉冲干扰的问题,提出一种基于改进快速峭度图的轴承故障诊断方法。首先运用自相关算法、改进谱幅值调制(improved spectral amplitude modulation, ISAM)算法对包络信号进行处理,获取修正信号并以其频谱峭度衡量冲击响应的强度;然后以修正信号谱峭度代替传统快速峭度图中的包络峭度,得到可降低各频带随机干扰的改进峭度图;最后通过对3组信噪比不同的仿真信号、2组轴承试验信号的平方包络谱故障诊断情况进行对比分析,验证该方法的有效性和稳定性。通过与原始快速峭度图、功率谱峭度图、ISAM法3种方法对比,验证该方法的优越性。对比结果表明,所提方法可有效降低随机脉冲的影响,实现轴承故障特征频率的准确提取。

Here,aiming at the problem of traditional fast kurtosis graph being prone to random pulse interference in determining frequency band range by calculating envelope kurtosis of filtered signal during bearing fault diagnosis,a bearing fault diagnosis method based on improved fast kurtosis graph was proposed.Firstly,the autocorrelation algorithm and improved spectral amplitude modulation(ISAM)algorithm were used to process an envelope signal,obtain a modified signal,and measure the intensity of impact response based on its spectral kurtosis.Then,the modified signal spectral kurtosis instead of envelope kurtosis in traditional fast kurtosis graph was used to obtain an improved kurtosis graph for reducing random interference in per frequency band.Finally,the effectiveness and stability of the proposed method were verified by contrastively analyzing square envelope spectrum fault diagnosis case using 3 sets of simulated signals with different signal-to-noise ratios and 2 sets of bearing test signals.Compared with the original fast kurtosis graph,power spectrum kurtosis graph and ISAM method,superiority of the proposed method was verified.The comparison results showed that the proposed method can effectively reduce effects of random pulses and realize accurate extraction of bearing fault feature frequencies.