基于振动信号残差法的齿轮故障诊断分析与仿真

摘要

At first, in the paper vibration signals have been collected and dealt based on time synchronous averaging ( TSA) in order to prevent gear fault. Then fast Fourier transform method has been used in order to make signals averaged into two parts, so normal signal and fault signal of gear has been obtained. After calculating valid residual value, quantitative presentation of fault variety can be gotten. By comparing, the results show that there is no difference in time domain waveform, while vibration amplitude is smaller in residual error diagram than in time domain waveform. The fault signal is in the place of obvious peak value where the prominent position of peak value is the place of the most serious fault.%为了防止齿轮故障,文章基于振动信号残差法,将采集的振动信号进行同步时域平均后,再将其进行快速傅里叶变换,将平均后的信号独立分离成两部分,从而得到齿轮的常规信号和故障信号.然后,再通过计算残差信号有效值,对故障变化程度给出定量描述.通过比较,可以明显看出时域图中信号波形无明显差异,而经处理后的残差图中常规振动信号幅度较小,波形中出现明显的峰值变化处,即故障信号,同时峰值越突出的地方即故障越严重的位置.