Research On Model Reconstruction Method Of The Defection Components For The Remanufacturing

Nowadays, with the earth resources have been getting waned, and the increasing environmental pollution, the Remanufacturing engineering is paid attention gradually. The reverse engineering technology has been becoming the important method to realize the remanufacturing. Aiming at the existing problems in the reverse engineering, such as point cloud patch don't have good ability of robustness, the speed of triangulating is slow, the defect model has lower slicing accuracy, this paper proposed some methods to solve these problems. In order to get the completely model, the defect clouds should be repaired and complete the grid reconstruction of cloud data. The intersection of slicing data should be optimized and then we obtain the high precision of the slicing model.The main study contents are as following:Firstly, in order to calculate the geometric properties of point cloud models, it's necessary to use the method of adaptive octree to finish the segmentation of point cloud model of the damaged part. According to the k neighbor search method complete the estimation of point cloud vector in every cuboid. In order to get the boundary points we use the weighted metrics to calculate the probability of every point becoming the boundary points, and extracted with high probability as the boundary point. Experiments show that this method not only easy realization and high efficiency, it can be extracted from point cloud model feature points and can complete the repair of the damaged point cloud.Secondly, according to the method of adaptive octree, we realize the point cloud segmentation, also proposed the partion of region growing method. Then the method was optimized by the PSO. We set the seed triangle to create standards, and the design criteria for mathematics evaluation model. And the introduction of PSO optimization algorithm with two points of an edge based search third points, so as to ensure the improvement in the speed of reconstruction of the case to ensure the accuracy and quality of surface model.Thirdly, the error messages for STL mesh model sections caused by the proposed thickness slicing method slicing processing for defect model. The coordinates of the intersection section compression based on wavelet transform and adaptive value of the data, thus to optimize and eliminating the false intersection points. This method can significantly improve the slice point data, the slice data accuracy is more accurate, accurate representation of the original model.Lastly, we put the damaged turbine engine blades as the research object for the test. The point cloud segmented by the method proposed in this paper. Adjusting the vector of point cloud and the feature points can be extracted by weighted metric to ensuring the efficiency and accuracy of feature point extraction. Using the Boolean difference operation got the broken blade. Slicing the damaged blade and its defects by the method proposed in this paper. And then we used the 3D printer to get the broken blade and its defection part, and the two model basic anastomosis. The comparing results verified the feasibility of this method.This paper not only provides a new method for the modeling of defection parts, but also optimized the method of equal thickness slicing, which has important significance to improve the precision of repairing the defection parts.