改进经验小波变换与最小熵解卷积在铁路轴承故障诊断中的应用

经验小波变换是一种小波框架下的自适应信号分解方法,对旋转机械的非线性、非平稳振动信号有很好的分解作用。针对传统经验小波变换过程中频谱划分过多的问题,提出根据互信息值对频谱进行重新划分与合并的方法,能有效减少频带数量;选择峭度值最大的分量进行信号重构,再使用最小熵解卷积对重构信号进行降噪;对降噪后的信号进行包络分析,能够有效地诊断出滚动轴承的微弱故障。通过仿真信号与铁路货车轮对轴承实验信号验证了该方法的有效性,为下一步工程应用奠定了基础。

Application of improved wavelet transform and minimum entropy deconvolution in railway bearing fault diagnosis

Empirical wavelet transform is an adaptive signal decomposition method in wavelet framework,which can decompose nonlinear and non-stationary vibration signals of the rotating machinery.Aiming at the problem of too much spectrum division in the process of traditional empirical wavelet transform,a new method of spectrum division and combination based on mutual information value was proposed,which can effectively reduce the number of frequency bands.Then,the component with the maximum kurtosis value was selected for signal reconstruction,and the minimum entropy deconvolution was used to reduce the noise of the reconstructed signal.The signal envelope analysis after noise reduction can effectively diagnose the weak fault of rolling bearings.Simulation signals and experimental signals were used to verify the method in the paper,which lays a foundation for further engineering applications.