摘要
本文用奇异值分解方法研究数控工作台动态特征。测试数控工作台三个方向的加速度信号;用奇异值分解得到信号的奇异谱;找出构成信号的主分量;通过提取主分量动态特征来研究信号的动态特征及其影响因素;通过计算信号能量来对信号及主分量进行量化分析。结果表明:构成信号的前两个主分量的信号特征明显,依此可以提取出混沌信号的动态特征;工作台铅垂方向的信号与水平面上运动方向的信号有很大的互相关性,说明信号来自于同一激振源,而水平面上与运动方向相互垂直的两个方向的信号互相关性小;运动方向的信号总能量最大,水平非运动方向的最小,铅垂非运动方向的居中;丝杠在铅垂方向的弯曲振动远比水平方向的大。
The SVD （singular value decomposition） method is used to study dynamic characteristics of NC （numerically controlled） table in this paper. First, acceleration signals of the NC table in three directions are tested and the singular spectrum of the signals with the SVD to find out the principal components of the signals is obtqined ; second, the dynamic characteristics of the signals and its contributing factors through drawing dynamic characteristics of principal components are studied; third, quantitative analysis on signals and principal components through calculating signal energy are made. The result shows that signal characteristics of the previous two princi- pal components are apparent. Based on the dynamic characteristics of chaotic signal, the signal in perpendicular direction of table is greatly correlated with signal in the horizontal moving direction, indicating that they are from the same vibration source. But signals perpendicular to each other in terms of moving direction at the horizontal level is rarely correlated. The total signal energy is maximum in moving direction, minimum in horizontal non-moving direction and middle in perpendicular non - moving direction. Meanwhile, bending vibration Of lead screw in perpendicular direction is far more violent than that in horizontal direction.
出处
《南阳理工学院学报》
2009年第1期8-11,共4页
Journal of Nanyang Institute of Technology
基金
国家973计划(2005CB724101)资助项目:数字化制造基础研究
