摘要
以复杂薄壁形曲面零件为研究对象,以提高复杂曲面零件的检测精度为目的,基于“DE/rand/2/hin”和“DE/best/2/bin”变异策略,提出一种新的混合式差分进化算法。针对曲面零件三维匹配问题的特性,本算法利用旋转和平移变量构成的转换矩阵为个体,以两个点云模型间的最小欧氏距离为目标函数并逐代进化,能够保证计算结果为全局最优解。实验结果表明,所提出的算法具有收敛速度快、搜索能力强、鲁棒性好等优点,该方法可有效降低测量误差对曲面零件质量检测的影响。
The complex and thin-walled surface parts are taken as the research object,and the aim is to improve the detection accuracy of complex surface parts.Based on the mutation strategies of"DE/rand/2/bin"and"DE/best/2/bin",a new hybrid differential evolution algorithm is proposed.Considering the characteristics of 3D matching problem of surface parts,in this algorithm,the transformation matrix composed of rotation and translation variables is used as individuals.In addition,the minimum Euclidean distance between the two point cloud models is taken as the objective function,and the algorithm evolves generation by generation,which can ensure the calculation result is the global optimal solution.Experimental results show that the proposed method has the characteristics of fast convergence,strong search ability and good robustness.The algorithm can effectively reduce the influence of measurement error on the quality detection of complex parts.
作者
李太峰
高亮
LI Tai-feng;GAO Liang(The 718th Research Institute of CSIC,Handan 056000,China;School of Mechanical Science and Engineering,Huazhong University of Science and Technology,Wuhan 430074,China)
出处
《测控技术》
2019年第8期89-94,共6页
Measurement & Control Technology
基金
国家自然科学基金项目(51721092)
关键词
曲面零件
模型匹配
差分进化算法
质量检测
free-form surface parts
model matching
differential evolution algorithm
quality inspection
