【摘要】 齿轮是重要的机械传动部件,齿轮的故障预测是设备故障诊断的重要研究内容之一。从故障诊断的角度可以将齿轮故障分为分布式故障和局部故障,分布式故障能量分布与啮合频率及其倍频有关,局部故障能量分布与齿轮轴旋转频率及其倍频有关。针对二者特征频率的差别,可以通过构建时序分析中的ARMA预测模型对定轴齿轮振动信号的频谱进行预测,然后结合预测值和当前值对齿轮故障进行预测。构建的预测模型具有较好的预测精度,预测结果可用于齿轮运行状态分析和故障预测分析,具有重要的应用价值。更多还原

【Abstract】 Gear is the important mechanical transmission parts,the gear fault prediction is one of the important contents in the research of equipment fault diagnosis. Gear faults can be divided into distributed and local fault analyzed based on fault diagnosis,the energy distribution of the distributed fault is related to the meshing frequency and its octave,the energy distribution of the local fault is related to the rotation frequency of the gear shaft and its octave. According to the difference of their characteristic frequency,the frequency spectrum of the vibration signal of the fixed axis gear can be predicted by constructing the ARMA prediction model in the time series analysis and then the gear faults can be predicted by combining the predictive value and the current value. The prediction model constructed in this paper has good prediction accuracy,which can be used in the analysis of gear operation state and fault prediction,so it has important application value.更多还原