文献类型：期刊文献

中文题名：基于遗传算法的复杂平面曲线轮廓度误差评定

英文题名：Error Evaluation on Planar Free-Form Curve Profile Based on Genetic Algorithm

作者：郭慧[1];马永有[1];潘家祯[1]

机构：[1]华东理工大学机械与动力工程学院,上海200237

年份：2007

卷号：33

期号：6

起止页码：888

中文期刊名：华东理工大学学报（自然科学版）

外文期刊名：Journal of East China University of Science and Technology

收录：CSTPCD;;EI(收录号：20080411059301);Scopus;北大核心:【北大核心2004】；CSCD:【CSCD2011_2012】；

语种：中文

中文关键词：遗传算法;平面曲线;轮廓度误差;误差评定;NURBS

外文关键词：enetic algorithm; planar free-form curve; profile error; error evaluation~ NURBS

摘要：针对不规则曲线轮廓度误差评定中存在的问题,提出了一种基于非均匀有理B样条(NURBS)插值与遗传算法相结合的复杂曲线轮廓度评定方法,对离散数据点表示的理论轮廓进行3次NURBS插值反算控制点,建立了理论轮廓曲线的数学模型;采用实数编码的遗传算法求解测量点与理论轮廓曲线位置偏差,消除了由于位置偏差引起的轮廓度评定的不精确问题;阐述了测量点到理论轮廓最短距离的求解算法和步骤。实验结果表明该方法能够快速获得较好的误差评定结果。
Aiming at the error evaluation on planar free-form curve profile, a method combined NURBS interpolation and real coding genetic algorithm is proposed for evaluating planar free-form curve profile. Theoretical curve was built by interpolating measure points and reversed calculation control vertexes. The genetic algorithm is taken for matching measure points and theoretical curve to avoiding the evaluation error derived from position error. The algorithm of the minimum distance between measure points and theory profile curve is presented in detail in the paper. The experiment results show the proposed method can rapidly acquire a better result.