【摘要】 为了获得某型号发动机叶片的三维建模及其相关数据，给逆向设计提供一些零件的三维建模尺寸。由于获得建模的三维点云数据时必然会产生孔洞，文中使用基于Genetic Algorithm（基因遗传算法）优化的Back Propagation（反向传播)神经网络(又称GA-BP神经网络)作为一个回归预测算法，来对产生的散乱点云孔洞加以修复和点云处理，最后再通过Geomagic Wrap对某型号发动机叶片进行逆向建模。通过对比BP神经网络可知，GA-BP修补孔洞的误差明显降低，满足对三维模型精度较高的要求，可应用于逆向工程。更多还原

【Abstract】 To obtain the 3D modeling of a certain type of engine blade and its related data, the 3D modeling dimensions of some parts are provided for reverse design. Due to the inevitable creation of holes when obtaining 3D point cloud data for modeling, this paper uses back propagation neural network(also known as GA-BP neural network) optimized based on genetic algorithm as a regression prediction algorithm to repair the generated random point cloud holes, then carry out point cloud processing, and finally uses Geomagic Wrap to build reverse modeling of an engine blade. By comparing with BP neural network, it can be seen that the error of hole repair by GA-BP is significantly reduced, which meets the requirement of high accuracy of 3D model and can be applied to reverse engineering.更多还原