以叶片为实例，在迭代最近点（ICP）精配准的基础上，遵循新一代产品几何技术规范（GPS），提出一种引入改进灰狼优化算法（IGWO）的线轮廓度误差评定方法。首先，根据最小区域理论建立线轮廓度评定模型；其次，利用非均匀有理B样条（NURBS）插值拟合出理论轮廓线，并用分割逼近法求算测点到NURBS的最短距离；接着，采用Logistic混沌映射初始化狼群，融合莱维飞行策略更新狼群位置；最后，以ICP参数为变量，基于IGWO并结合分割逼近迭代计算轮廓度误差，同时对比了GA、PSO、WOA三种算法下的计算精度。通过实验计算表明：该方法可用于叶片线轮廓度误差评定，在ICP配准的基础上，轮廓度计算精度提高了26.79%。IGWO相较于GWO收敛速度和计算精度均有所提高，且优于其他算法。

Taking the blade as an example， based on the iterative closest point （ICP） fine registration and following generation product specification （GPS）， a method for evaluating the line profile error by introducing the Improved Grey Wolf optimizer （IGWO） is proposed. Firstly， a line profile evaluation model is established according to the minimum zone theory. Secondly， the theoretical profile line is interpolated and fitted by using Non-Uniform Rational B-Spline （NURBS）， and the shortest distance from the measurement points to the NURBS is calculated by the subdivision approximating method. Logistic chaotic mapping was used to initialize the wolves， and merge Levy flight strategy to update Wolves position. With ICP parameters as variables， the profile error is computed iteratively based on IGWO combined with subdivision approximating； and the calculation accuracy of GA， PSO and WOA algorithms is compared. The experimental results show that this method can be used to evaluate the line profile error of blade. Based on ICP registration， the calculation accuracy of the profile is increased by 26.79%， and compared with GWO， the convergence speed and calculation accuracy of GWO are improved， and it is better than other algorithms.