基于工图DXF文件的圆柱齿轮三维重建  

Research on 3D Reconstruction of Cylindrical Gear Based on DXF File

作　　者：张昱 张平宽[1] 干佳金 ZHANG Yu;ZHANG Pingkuan;GAN Jiajin(School of Mechanical Engineering,Taiyuan University of Science and Technology,Taiyuan 030024,China)

机构地区：[1]太原科技大学机械工程学院,太原030024

出　　处：《自动化与仪表》2024年第11期146-150,153,共6页Automation & Instrumentation

摘　　要：针对当前三维模型在表达公差时存在表达不清的问题,该文提出了一种基于DXF文件的齿轮零件重建方法。首先,该方法以DXF文件为研究对象,通过提取其中的实体信息和标注信息,采用K-means对实体所属视图进行聚类;然后,利用BP神经网络识别和提取零件的重要图元信息,特别是关键标注信息和标注的实体信息;最后,通过基于VBA的参数化建模,在三维零件上表达关键形状公差和粗糙度信息。这一方法为解决三维模型表达公差问题提供了有效的解决方案,对于进一步研究具有指导意义。Aiming at the problem that the current three-dimensional model has unclear expression when expressing tolerances,a gear part reconstruction method based on DXF file is proposed.This method takes the DXF file as the research object,and uses K-means to cluster the views of the entities by extracting the entity information and annotation information.Then,the BP neural network is used to identify and extract the important primitive information of the parts,especially the key annotation information and the annotated entity information.Finally,through the parametric modeling based on VBA,the key shape tolerance and roughness information are expressed on the three-dimensional parts.This method provides an effective solution to solve the tolerance problem of 3D model expression,and has important guiding significance for further research.

关 键 词：DXF文件 特征识别 BP神经网络 三维重建 

分 类 号：TH16[机械工程—机械制造及自动化] TP391.4[自动化与计算机技术—计算机应用技术]