%0 Journal Article
%T 基于XGBoost的双目标零件类别划分技术与优化<br>Classification Technology and Optimization of Double Objective Parts Based on XGBoost
%A 朱露
%A 仲梁维
%J Modeling and Simulation
%P 734-743
%@ 2324-870X
%D 2022
%I Hans Publishing
%R 10.12677/MOS.2022.113069
%X 针对目前复杂产品装配对象内部零件结构层次识别困难的情形，以提高装配工艺文件编制的合理性、准确性以及其效率为目标。研究基于贝叶斯优化的XGBoost算法，以五套不同型号的生物器皿消毒机三维模型与装配工艺卡作为工艺数据源，对零件进行了关于结构类型和功能类型的双目标多类别的自动识别。得到结构类型和功能类型识别的准确率分别可达到96%和90%。实现自动识别并划分零件类型，丰富了装配体内部信息。<br />
In view of the difficulty in identifying the internal part structure hierarchy of complex product assembly object, the goal is to improve the rationality, accuracy and efficiency of assembly process documentation. The XGBoost algorithm based on Bayesian optimization algorithm is studied. Five sets of three-dimensional models and assembly process cards of different types of biological utensils disinfection machines are used as the process data source to automatically identify the structure type and function type of parts. The recognition accuracy of structure type and function type can reach 96% and 90% respectively. Automatic identification and classification of parts are realized, which enriches the internal information of assembly.
%K 装配工艺，零件划分，XGBoost，贝叶斯优化算法<br>Assembly Process
%K Part Classification
%K XGBoost
%K Bayesian Optimization Algorithm
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=51497