基于时域同步平均法的行星齿轮振动信号分离技术

摘要

基于某个特定的行星齿轮转过最小圈数后回到起始点这一特性,提出一种基于时域同步平均法的振动信号分离技术.由于实验采样时间较短,为了更快获取数据,根据行星齿轮啮合信号形状和齿宽,以及能量泄露小的特点,提出选择合适的窗函数-Tukeywin函数对信号进行加窗截取.选取窗宽为行星齿信号的一个齿宽,利用时域同步平均法对加窗信号进行相加平均,可提取到完整的行星齿轮信号.实验结果与对比分析表明,该方法简单且行之有效.%The minimum number of turns was based on a particular planetary gear turned back to the starting point of this feature, a method of vibration signal separation technology was proposed, that was based on time-domain synchronous average. The method needed to get data faster because of short sampling time, so a right window function to extract signals was chosen. According to the characteristics of meshing signals, such as the shape of meshing signals, tooth width and its energy leak, the paper selected Tukeywin function and made sure the selected window width equal to the tooth width of the meshing signals. Time -domain synchronous average can extract complete signals of the planetary gear teeth through Tukeywin function after averaging signals. The experimental results and contrast show that this method is simple and effective.