中国机械工程 ›› 2014, Vol. 25 ›› Issue (12): 1645-1650.

• 智能制造 • 上一篇    下一篇

基于误差转换及图像域的圆度评定方法

张学昌1;梁涛2;唐艳梅1   

  1. 1.浙江大学宁波理工学院,宁波,315100
    2.浙江大学,杭州,310058
  • 出版日期:2014-06-26 发布日期:2014-06-27
  • 基金资助:
    国家自然科学基金资助项目(51075362);浙江省自然科学基金资助项目(Y1100073);宁波市产业技术创新及成果产业化重点资助项目(2013B10022) 

Roundness Evaluation Method Based on Error Transformation and Image Domain

Zhang Xuechang1;Liang Tao2;Tang Yanmei1   

  1. 1.Ningbo Institute of Technology of Zhejiang University,Ningbo,Zhejiang,315100
    2.Zhejiang University,Hangzhou,310058
  • Online:2014-06-26 Published:2014-06-27
  • Supported by:
    National Natural Science Foundation of China(No. 51075362);Zhejiang Provincial Natural Science Foundation of China(No. Y1100073)

摘要:

针对工程应用中圆度误差评定方法存在理论深奥、计算复杂、检测效率低且不适用于大容量采样点的问题,提出了一种基于误差转换及图像域的圆度误差评定方法。该方法首先将图像域测量得到的原始圆度误差进行转换,使其满足误差评定的要求;然后以最小二乘圆为起始圆,寻求半径或半径差的“极大中的极小”,通过对最小二乘圆进行小尺度平移,并用遗传算法得到该平移规划坐标,从而获得平移后的理想圆并求得圆度误差值;最后对某型号零件进行试验,试验结果与用三坐标测量得到的结果相吻合,表明该方法可以有效、正确地进行圆度误差的评定。

关键词: 圆度评定, 误差转换, 图像域, 遗传算法

Abstract:

For solving the problems of roundness error assessment methods with profound theory, complex computation, low detection efficiency and inapplicable to sample points with large capacity in engineering application, a new method was proposed based on error transformation and image domain. The method transformed the roundness errors from the field measurements to the image domain and made it meet the requirements of error estimation. The method took the least square circle as the initiative round to seek the  “minimumin maximum” of its radius or radius difference. And then through small scale translation of the least square circle and getting the translation planning coordinate by genetic algorithm, the ideal circle after translation and roundness error value were obtained. Experimental results of a certain type of part show that the roundness error can be evaluated effectively and exactly by using this method because it is very close to the results measured by coordinate measuring machine.

Key words: roundness evaluation, error transformation, image domain, genetic algorithm

中图分类号: