• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2016, Vol. 52 ›› Issue (15): 139-146.doi: 10.3901/JME.2016.15.139

• 数字化设计与制造 • 上一篇    下一篇

复杂曲面零件加工精度原位检测系统的残余误差补偿*

高健12, 黄沛霖1, 文章1, 张亚莉3, 邓海祥3   

  1. 1. 广东工业大学机械装备制造与控制技术教育部重点实验室 广州 510006
    2. 广东工业大学广东省计算机集成制造重点实验室 广州 510006
    3. 广东工业大学实验教学部 广州 510006
  • 出版日期:2016-08-05 发布日期:2016-08-05
  • 作者简介:

    高健(通信作者),女,1964年出生,博士,教授,博士研究生导师。主要研究方向为精密检测技术、曲面修复技术、微电子封装装备技术。

    E-mail:gaojian@gdut.edu.cn

  • 基金资助:
    * 国家自然科学基金(51175093)和广东省科技计划(2015B010104008,2015A030312008)资助项目; 20150901收到初稿,20160517收到修改稿;

Residual Error Compensation of In-situ Inspection System for Precision Machining of Complex Components

GAO Jian12, HUANG Peilin1, WEN Zhang1, ZHANG Yali3, DENG Haixiang3   

  1. 1. Key Laboratory of Mechanical Equipment Manufacturing & Control Technology of Ministry of Education, Guangdong University of Technology, Guangzhou 510006
    2. Guangdong Provincial Key Laboratory of Computer Integrated Manufacturing, Guangdong University of Technology, Guangzhou 510006
    3. Department of Experiment Teaching, Guangdong University of Technology, Guangzhou 510006
  • Online:2016-08-05 Published:2016-08-05

摘要:

复杂曲面零件数控加工后直接进行原位加工精度检测和误差补偿,是实现精密产品闭环制造模式的有效途径。原位检测系统的误差来源于测量系统误差和机床运动系统误差,经相关的误差分离与误差补偿后,仍存在较大的残余误差,影响检测精度及其推广应用。针对原位检测系统的检测精度问题,开展检测系统残余误差的回归建模与补偿研究,在机床几何误差、测头半径误差以及预行程等基本误差补偿的基础上,建立基于偏最小二乘回归分析算法的误差回归模型,实现曲面零件测点法矢方向的检测数据二次补偿。在算法实现的基础上,列举复杂曲面零件进行数控加工与在线检测的试验研究。试验结果表明,二次误差补偿方法可以进一步提高原位检测系统的检测精度。

关键词: 回归模型, 偏最小二乘法, 误差补偿, 原位检测, 加工精度

Abstract:

:In-situ inspection system used for the machining precision inspection of a complex surface part is an effective approach for the closen-loop manufacturing mode. Errors of an in-situ inspection system are often original from the measuring system and the machine tool system. Even with the process of error seperation and error compensation, the system still exists big residual error, which decreases the system inspection accuracy and affects its applications. In order to improve the inspection accuracy for the complex parts after they are machined on a machine tool, the residual error compensation method on the basis of previous compensation is focused in geometric error, the probe radius error and pre-travel error. An error regression model is generated to predict the residual error through the algorithm of Partial Least-Squares Regression. Along the normal vector direction of the measuring points on the machined part, the predicted error is compensated accordingly. Several testing parts are exampled through the machining, measuring, CMM inspecting and compensating operations. The results show that the error regression model can further improve the inspection accuracy for the in-situ inspection system.

Key words: error compensation, in-situ inspection, partial least-squares regression, regression analysis, precision machining