With the rapid development of integral casting and 3D printing technology, the geometric structure of cylindrical thin-walled part is becoming more and more complex by integrating various accessory parts. The complex cylindrical thin-walled part has been widely used in manufacturing field with high requirement for face milling accuracy, especially for wall thickness accuracy. However, due to the existence of initial uncertain manufacturing deviations, such as blank casting errors, initial geometric deformations and clamping deformations, there is a large geometric deviation between actual blank workpiece and theoretical design model. While the milling process is designed based on the theoretical model rather than the actual model, the final wall thickness of milled surface is easy to be out of tolerance and even be cut through. In order to solve this problem in lacking of reliable geometrical model for face milling of complex cylindrical thin-walled parts, a method for accurate modeling and characterization of wall thickness considering the uncertain geometric deviations based on point cloud reconstruction is proposed.The proposed point cloud reconstruction-based method can accurately represent the geometric features of milling areas of complex cylindrical thin-walled part. Through feature extraction and meshing of the measured point cloud based on high definition metrology, the subregions of point cloud are adaptively fitted into small curved surfaces. And then NURBS surfaces are used to connect adjacent subregions to form a high-precision solid model as a whole. Furthermore, a method for identifying the wall thickness in milling areas of thin-walled part is proposed. The discrete mesh nodes of internal surface are defined as the sampling points representing the milling areas and being the interesting points for wall thickness characterization. The wall thicknesses are calculated as the distances between the sampling nodes and the external meshed surface using an improved KD-Tree algorithm. &nbsp;The method proposed in this paper can efficiently realize geometric modeling and wall thickness characterization of parts with initial deviations, and thus providing model support for milling trajectory design and wall thickness control subsequently.The proposed method contains such following key points: (1) The point cloud is placed in a cylindrical coordinate system, and the point cloud splicing datum is used as the origin of cylindrical coordinate system to perform coordinate conversion. The cartesian coordinates of point cloud are converted to cylindrical coordinates to facilitate point cloud segmentation and thickness measurement. (2) The internal surface and external surface of the point cloud are separated to reduce wrong feature identification. The features of the complex-shaped internal surface are extracted and structural features such as ribs and bosses are used as the segmentation boundary to extract milling areas of internal surface. (3) Fine grids are used to divide the milling areas of the internal surface, and external surface and non-milling areas of internal surface are divided by coarse grids. The point cloud data is stored in the grids, each fitting a small curved surface. The mesh edges between mesh surfaces are used as control lines and connected by NURBS surfaces. (4) The mesh nodes of the milling areas on internal wall are used as discrete point sets to measure the distance to external surfaces. The improved KD-Tree is used to find the nearest mesh surface on external wall and calculate the distance, and the results are used to generate discrete wall thickness data of milling areas and thickness cloud maps. (5) Based on the proposed method, a CAD solid model with error within 0.1mm compared with the original point cloud is generated. The generated model is used to plan milling trajectory, and the error after machining is within 0.3mm, which verifies the accuracy of the model.

Geometric Modeling and Characterization of Wall Thickness for Complex Cylindrical Thin-Walled Parts With Uncertain Manufacturing Deviations

Paper Proceedings: https://asmedigitalcollection.asme.org/IMECE/IMECE2021/volume/85543

ASME 2021 International Mechanical Engineering Congress and Exposition (IMECE2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.