中文摘要：

工件表面综合形貌误差由表面粗糙度、表面波纹度和形状误差等成分组成，各成分对工件的使用性能有着不同程度的影响。各种误差成分的无失真提取是进行表面综合形貌评定的核心和关键。本文将灰色误差理论应用到表面综合形貌评定中，提出了表面综合形貌误差的灰色分离方法，应用灰色动态滤波将表面形貌误差分离为表面粗糙度、表面波纹度和形状误差三项之和。该方法不要求被测表面的原始采样数据服从典型分布，可以对少数据表面轮廓进行误差分离，而且分离过程中不损失原始数据。仿真结果表明，灰色动态滤波是进行表面综合形貌误差分离的有效方法，其误差分离的结果与现行高斯滤波法具有良好的一致性。该方法可以作为国际标准的高斯滤波法的一种补充。

英文摘要：

The comprehensive topography errors of engineering surfaces are composed of surface roughness, surface waveness and profile error. The three topography errors influence the workpieces＇ functions and performances in varying degrees. To pick up the three compositions without distortion is the most important for surface comprehensive topography evaluation. In this paper, the grey error theory is applied in the surface comprehensive topography evalua- tion, and a grey separation method for surface comprehensive topography is brought forward. In the method, a grey dynamic filtering surface roughness based on dynamic GM （ 1,1 ） is used to separate the surface comprehensive topography errors into , surface waveness and the typical distributions, and the surface profile error. The primary sampled data of measured surface need not obey comprehensive topography with less data can also be separated without losprimary data. Through the sample and simulation analysis, the method is effective for separating the surface comprehensive topography errors, and the error separation results are well consistent with the Gaussian filtering method. The method can be one of complements for Gaussian filtering.