(1. AECC Shenyang Liming Aero-Engine Co., Ltd., Shenyang 110043, China; 2. School of Mechanical Engineering and Automation, Beihang University, Beijing 100191, China; 3. School of Energy and Power Engineering, Beihang University, Beijing 100191, China)
摘要航空发动机叶片长期服役在高温、高压和高速的环境下，极易出现损伤，而损伤叶片的修复技术一直是国内航空领域的研究重点和难点。以双联高导涡轮叶片为例，研究了一种基于逆向工程的航空发动机叶片再制造修复方法。该方法不依赖原始设计模型，而是基于同期服役非损伤叶片，这样重构的目标修复曲面更接近损伤叶片的变形状态。首先，用 3D 蓝光扫描仪扫描已损伤叶片及同期服役非损伤叶片以获取其点云模型；其次，对两叶片点云模型进行最佳拟合配准并分析识别出修复加工区域和边界；然后，基于非损伤区域配准后的误差情况，用多项式方程拟合曲线预测损伤区域的误差数据，再以非损伤叶片截面曲线为基础，叠加预测误差值求取修复区域的目标截面线；最后，通过蒙皮方法构建出目标修复曲面，编制数控加工程序，完成修复加工。以非损伤叶片的逆向模型为参考，对损伤叶片修复结果进行精度检测和误差分析。结果表明：该叶片修复轮廓精度基本上在 0.03mm 以内，满足修复公差要求，且修复区域与非修复区域能实现光滑搭接过渡，利用提出的方法对双联高导涡轮叶片有很好的修复效果。
Abstract：Aero-engine turbine blades are vulnerable to be damaged due to the serving environment of high temperature, high pressure and high speed. At present, the repair technologies of damaged blades have been a hotspot in the aircraft maintenance. In this paper, a method of remanufacturing and repairing aero-engine blades based on reverse engineering is studied by taking a turbine blade as an example. The target repairing surface reconstructed in this method based on non-damaged blade serving at the same time instead of the original design model will more accord with deformation state of damaged blades. Firstly, the damaged blade and non-damaged blade serving at the same time are scanned by the 3D blue-light scanner and the point cloud model of two blades can be acquired. Secondly, the repairing area and boundary can be identified after best fitting. And then, the errors between these two blades in repaired area will be predicted according to polynomial curves fitted by these errors in non-repaired area after registration. Furtherly, the section curves of target repairing surface can be obtained based on the predicted errors and section curves of non-damaged blade. Finally, the target repairing surface can be constructed using the skin method. And the constructed surface is used for the generation of NC tool paths which will achieve repair processing. The repaired results of the damaged blade are analyzed according to the reverse model of non-damaged blade. And the results show that the profile errors are basically within 0.03mm which meet the tolerance requirements of repairing blades and there are smooth transition between repaired and non-repaired areas. The method presented in this paper has a good effect on repairing aero-engine blades.
2020, 
