涡轮叶片形状检测中的模型配准控制点集选取

叶片模型的配准定位是叶片形状检测与分析的一个核心环节。为解决基于迭代最近点方法的叶片密集测量数据与其CAD模型的可靠配准问题，提出一种基于CAD模型的配准控制点集选取方法。结合模型配准可靠性分析技术，预先在叶片CAD模型规划出配准基准点集；通过测量数据与配准基准点集的配准运算，把配准基准点集映射到叶片测量数据上，获取测量模型上对应的配准控制点集；滤除其中可能降低模型配准精度的坏点，得到最终的叶片模型配准控制点集。进行基于仿真与实测数据的叶片模型配准试验及对比分析，总结出一些可用于指导叶片模型配准控制点集选取的策略。结果表明，所提出的叶片模型配准控制点集选取方法是可行的。

Model Registration Control Point Set Selection for Turbine Blade Shape Inspection

Model registration is a key technique for turbine blade shape inspection. In order to solve the problem of reliable registration based on the iterative closest point method for the density measurement data of the turbine blade and its CAD model, one kind of model registration control point set selection method based on the CAD model is brought forward. For certain blade, the model registration base points are firstly pre-planned according to its CAD model and the model registration reliability index; then, align the measurement data with the base point set, find out the measurement points relating to the base points as the alignment control points and filtering the bad points. Experiments based on both simulation data and real measurement data are performed and several guidelines for blade registration control point set selection are summarized. The feasibility of the discussed method for turbine blade shape inspection is verified by the experimental results.