LMD的LabVIEW实现及其在齿轮故障诊断中的应用

将局部均值分解(local mean decomposition,LMD)算法在LabVIEW平台上加以实现,开发出LabVIEW的LMD模块。为减小误差,采用三次样条插值法代替滑动平均法来获得局部均值函数和包络估计函数,用形态学滤波算法得到瞬时频率和瞬时幅值的平滑曲线,并通过仿真信号验证LMD算法对于多分量信号的分解能力。最后,利用开发的模块对实测齿轮磨损、断齿故障信号进行分析,成功提取出故障特征频率信息,结果表明开发的LMD模块可以有效应用于齿轮故障的诊断。

LMD realization by LabVIEW and its application in gear fault diagnosis

According to the local mean decomposition （LMD） algorithm, the LabVIEW LMD module was developed. It used cubic spline interpolation instead of sliding average method to obtain local mean function and envelope estimation function in order to reduce the error and used morphological filtering algorithm to smooth the instantaneous frequency and amplitude curve. The simulation signal was used to verify the decomposition ability of LMD for multi-component signals. Finally, the developed module was used to measure wear fault signal and broken teeth fault signal and successfully extract the fault characteristic frequency information. The results show that the developed LMD module could diagnosis the gear fault effectively.