基于时间序列分析的行星齿轮箱故障诊断

齿轮箱运转的环境会导致信号采集时产生强大的噪声干扰，提出时间序列的方法对齿轮箱进行故障诊断。采用时间序列分析对行星齿轮箱进行谱分析，通过与传统频谱分析比较可以看出，该方法具有较好的识别能力。通过Labview虚拟仪器设计的数据采集系统进行数据的采集提取特征信号，确定模型阶次为5阶，最后利用AR模型参数算法来确定其正常状态下齿轮参数容差范围，在阶次不变的情况下分析出故障信号的模型参数，与正常信号参数容差范围进行对比，从而对齿轮箱故障进行诊断，该方法在齿轮诊断方面效果较为显著。

Fault Diagnosis of Planetary Gearbox Based on Time Series Analysis

The environmental conditions of the gearbox operation will cause strong noise during signal acquisition, so a time-series method is proposed to perform fault diagnosis on the gearbox. Compared to the traditional spectrum analysis, we can see that the time series analysis has a better recognition ability for gearbox fault diagnosis by using time series analysis to investigate the spectrum of the planetary gearbox. Through the data acquisition system designed by Labview virtual instrument, the data is collected to extract feature signals. Then the AR model is used to determine the order of the model. Finally, the AR model parameter algorithm is used to determine the tolerance range of the gear parameters under normal conditions. When the order is unchanged, the model parameters of the fault signal are analyzed, and then compared with the tolerance range of the normal parameters to diagnose the fault of the gearbox. It indicates that the proposed method has a significant effect on gear diagnosis.