Abstract: The machining and measurement accuracy of the blades directly affect structural strength and aerodynamic performance of the turbofan.To improve the sampling accuracy and efficiency of on-machine measurement for blade quality inspection,a curved surface sampling method based on bending moment theory was proposed to solve the problems of sharp curvature change at leading edge or trailing edge and global profile sampling.According to the geometric characteristics change trend of blade surface,the arrangement of sampling points was confirmed automatically.The effectiveness and feasibility of the proposed method was verified by case simulations and on-machine measurement experiments.In contrast to the bidirectional isoparametric method,improved partition isoparametric method and Hammersley sequence method,the results showed that the proposed method realized the fine and accurate adjustment of the position and number of sampling points.It had the excellent performance of tiny sampling error and high surface reconstruction accuracy,which could well meet the requirements of on-machine measurement on time and accuracy.

Key words: turbofan blade, non-uniform rational B-spline, surface modeling, bending moment theory, adaptive sampling

摘要： 涡扇叶片的加工和检测精度直接影响到涡扇的结构强度和气动性能。为提高叶片的在机测量精度和效率,提出一种基于弯矩理论的自适应曲面采样方法,以解决叶片曲面在前缘和后缘位置处曲率急剧变化以及轮廓面的全局采样问题。根据曲面的几何特征变化,自适应确定检测点的分布密度和数量。仿真和在机测量实验验证了该方法的有效性和可行性。通过与双向等参数法、改进的分区等参数法和Hammersley序列法的采样结果对比表明,基于弯矩理论的自适应采样方法实现了对采样点位置及数量的微细精准调整,采样误差小,重构表面精度高,满足在机测量对时间和精度的要求。

关键词: 涡扇叶片, 非均匀有理B样条, 曲面建模, 弯矩理论, 自适应采样