倒频谱法在齿轮箱故障诊断中的应用

基于LABVIEW的虚拟仪器平台建立数据采集系统，实现了对齿轮进行全生命周期实验过程中采集到大量振动信号数据，提出使用时域分析法（时域同步平均）与频域分析法（功率谱）相结合的信号处理方法对振动信号数据按齿轮可能出现的各种运行状态加以分类，从而有重点的应用倒频谱法对已经出现故障的齿轮箱振动信号进行分析，倒频谱的边频带频谱识别能力有助于研究啮合频率及边频特征，进而准确诊断出齿轮故障性质，并定位出故障齿轮。实验结果得到的齿轮各阶段运行状态数据在工程中很实用，具体涉及齿轮状态判断、故障阈值设定、故障识别及定位等用途。

Application of Cepstrum in Gearbox Fault Diagnosis

A data acquisition system was designed on the basis of LABVIEW to collect vibration signals during the whole gear life cycle. By adopting the method of combining time synchronous averaging analysis in time domain and power spectrum analysis in frequency domain, the vibration signal data were classified according to the running states of gears. Then the vibration signals of the faulted gearbox were analyzed by cepstrum analysis method. Cepstrum is helpful to recognize the periodic signature in sideband spectrum caused by a faulted gear and makes contribution to fault diagnosis and fault location. The experimental results on gear life cycle are applicable in gear state detection, fault threshold setting, fault detection, fault location and so on.