【摘要】 轴系回转误差运动的特征是一项复合信号,它含有周期分量和随机分量。在识别和提取人工周期项,并经FISHER检验后,可近似地认为它是平稳随机过程。若在主轴某一回转截面内正交安装两个传感器X和Y,而随机变量x_i和y_i都具有正态分布时,则z=（x²+y²）^1/2服从参数σ>0的瑞利分布（Rayleigh distribution）。本文提出了一项新的试验分析方案。目的是简化现有的试验方法,提高测试精度。并建议将提取的周期函数项作为诊断影响精度的因素,而以瑞利分布的均值和方差来评价轴系回转轴心漂移的平均距离。更多还原

【Abstract】 The characteristics of error motion of shafting is a complex signal which includes periodic component and random term. After distinguishing and extracting the man-made periodic terms and Fisher test, the residual component may be considered as a stationary stochastic process. Using two pick-ups with electric output located in two perpendicular direction X and Y, thus, z=（x²+y²）^1/2 accord with Rayleigh distribution of σ>0, provided the stochastic variants have Gaussian distribution. The purpose of this paper is to simplify the method of testing axial error motion and to raise the testing accuracy. It is suggested also that the extracting periodic terms are regarded as the influence factor in the error motion, and the mean—value and variance of Rayleigh distribution as the evaluating the rotation accuracy of shafting.更多还原