基于分割球面逼近和差分进化的复杂曲面轮廓度误差评定

Evaluation of Complex Surface Profile Errors Based on Subdivision and Sphere Approximation Method and Differential Evolution Algorithm

为了评定复杂曲面轮廓度误差,提出了一种快速简便的分割球面逼近方法。该方法用球面逐步逼近测点垂足所在的曲面片来求得测点到设计曲面的最短距离,然后利用差分进化算法优化测点位置,结合分割球面逼近方法最终求得复杂曲面轮廓度误差值。将分割球面逼近方法和差分进化算法相结合,用于计算蜗杆齿面的轮廓度误差,仿真结果表明,该方法计算精度高于最小二乘法,能满足高精度误差评定的需求。

In order to evaluate the profile errors of complex surface, a subdivision and sphere ap- proximation method, fast and convenient, was firstly presented. Approximating the surface patch con- taining the perpendicular foot of measuring point by a sphere,the shortest distance between a measur- ing point and surface was obtained. Secondly, a differential evolution algorithm was selected to opti- mize the position of measuring points. Finally, the evaluation of complex surface profile errors was re- alized by using subdivision and sphere approximation method combined with differential evolution al- gorithm. The proposed method was used to evaluate worm tooth--surface profile errors and the simu- lation results show that the method can obtain higher calculation accuracy compared with least square method and can satisfy the requirements of high--precision error evaluation.