自由曲面几何偏差的空间统计分析  

Spatial Statistical Analysis of Geometric Deviations of Free-form Surfaces

作　　者：靳龙[1] 陈岳坪[1] 葛动元 JIN Long;CHEN Yueping;GE Dongyuan(School of Mechanical and Transportation Engineering,Guangxi University of Science and Technology,Liuzhou,Guangxi,545006)

机构地区：[1]广西科技大学机械与交通工程学院,柳州545006

出　　处：《中国机械工程》2020年第4期445-450,共6页China Mechanical Engineering

基　　金：国家自然科学基金资助项目(51565006,51765007);广西自然科学基金资助项目(桂科自2018GXNSFAA050085,2014GXNSFAA118337);2016年广西高校高水平创新团队及卓越学者计划资助项目;广西科技大学创新团队支持计划资助项目。

摘　　要：为深入分析数控加工自由曲面几何偏差的内在联系,采用空间统计分析方法将这些几何偏差分解为确定性偏差和随机性偏差来确定误差源的影响。采用B样条方法对确定性偏差形成的曲面进行迭代建模,并用莫兰指数在迭代的每一步检验模型残差的空间统计特性,若模型残差具有空间独立性,则模型残差可作为随机性偏差。通过对一个数控加工后的自由曲面进行几何偏差分解,验证了所提方法的有效性。To analyze the intrinsic relationship among the geometric deviations of machined free-form surfaces,a spatial statistical analysis method was applied to decompose these deviations into deterministic and random deviations,which would help to determine error sources.The B-spline method was applied to iteratively build a surface that forms deterministic deviations.And the Moran I was used to test the spatial statistical characteristics of the model residuals in every iterative step.If the residuals were spatially independent,then they might present the random deviations.To verify the presented method,the geometric deviations on a CNC machined free-form surface were successfully decomposed into the deterministic and random deviations.

关 键 词：自由曲面 几何偏差 空间统计分析 空间独立性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]