基于信号共振稀疏分解与包络谱的齿轮故障诊断  

Fault Diagnosis of Gears Based on Resonance-based Sparse Signal Decomposition and Envelope Spectrum

作　　者：张文义[1] 于德介[1] 陈向民[1] 

机构地区：[1]湖南大学汽车车身先进设计制造国家重点实验室,长沙410082

出　　处：《中国机械工程》2013年第24期3349-3354,共6页China Mechanical Engineering

基　　金：国家自然科学基金资助项目(51275161)

摘　　要：当齿轮出现断齿、裂纹等局部故障时,其振动信号会出现周期性冲击脉冲。在齿轮故障早期,由于冲击脉冲微弱,常淹没在齿轮的啮合频率、转频等谐波成分以及噪声中,因此,对于齿轮早期故障,直接对齿轮振动信号做包络谱分析以诊断齿轮局部故障通常效果不佳。针对这一问题,将信号共振稀疏分解方法与包络谱分析相结合,提出了基于信号共振稀疏分解与包络谱的齿轮故障诊断方法。该方法采用信号共振稀疏分解将冲击脉冲从齿轮振动信号中分离出来,然后对冲击脉冲做Hilbert包络分析,获取冲击脉冲出现的周期,进而对齿轮状态和故障进行识别。仿真算例和应用实例证明了该方法的有效性。When gear＇s local faults such as tooth crack or broken occured,the vibration signals of gears always had periodic impulse component. In the early stage of gear＇s faults, those periodic im- pulse components usually submerged in nosises and the harmonic components such as gear meshing frequency components and rotating frequency components. Therefore,it is difficult to detect gear＇s lo- cal faults by using envelope analysis of the vibration signals of the gear effectively. Aiming at that problem,a method for fault diagnosis of gears based on resonance--based sparse signal decomposition and envelope spectrum was proposed. In this method, the impulses were separated from the vibration signals of gears by using resonance--based sparse signal decomposition. Then the impulses were ana- lyzed by Hilbert envelope method, the cycle of the periodic impulse component can be acquired and the faults of the gear can be diagnosed. Simulation and application examples prove effectiveness of the method.

关 键 词：品质因子 稀疏分解 包络分析 齿轮 故障诊断 

分 类 号：TH113.1[机械工程—机械设计及理论] TH165.3