基于二次相移数据重组的直行运动误差在线检测分离技术

通常的直线误差分离方法，一般是先行分离出工件的形状误差，再据此求出运动误差，由于存在着误差残留，不利于运动误差的在线检测．本在三点法测量数据的基础上，提出了两种基于二次相移的直行运动误差先行分离方法：单信道二次相移数据重组法和多信道二次相移数据重组法，即通过对三测头数据按照二次相移原则进行单信道或多信道的数据重组，便能够在EST的首次操作时消除形状误差的影响，在反滤波基础上先行分离出直行运动误差的平移分量和转动分量．权函数的对比分析表明：先行分离出运动误差的单信道或多信道二次相移数据重组法与先行分离出形状误差的普通频域三点法具有相同的权函数，它们在本质上具有同源性和统一性．实测结果证明了这两种方法的正确有效性．

On-line Measuring and Separating Techniqueof Linear Motion Error Based on Quadratic Phase-shift Data Re-organization

In common straightness error separation techniques (EST), straightness error is separated and obtained firstly, and then linear motion error can be obtained secondly. Owing to inevitable error residuals, it will not be beneficial to the on machine measurement of linear motion error. To solve the problem, two new methodssingle channel quadratic phase shift data re organization method and multi channel quadratic phase shift data re organization method are proposed in this paper, which are based on the principle of quadratic phase shift and use the same measurement data coming from ordinary three point method. Through the re organization of data from one channel or multi channels , the influence of straightness error can be eliminated, and the linear motion error can be acquired directly on the basis of inverse filtering. Meanwhile, comparative analysis indicates that the same weighting function will be produced between the proposed methods for motion error measurement and the ordinary frequency domain three point method for straightness error measurement, and the homology and unitarily in nature will be confirmed. At last, satisfactory separating results in some experiments prove the correctness and effectiveness these proposed methods.