高技术通讯2026，Vol.36Issue(2)：170-178,9.DOI:10.3772/j.issn.1002-0470.2026.02.006
基于GRA-DBN的机器人打磨工艺多目标优化方法
A multi-objective optimization method for robot polishing process based on GRA-DBN
陶永 1薛蛟 2杨林 2王潇桐 2刘亚醉 2魏洪兴1
作者信息
- 1. 北京航空航天大学机械工程及自动化学院 北京 100191
- 2. 北京航空航天大学航空发动机研究院 北京 100191
- 折叠
摘要
Abstract
Robot polishing has become an important machining method for improving the surface quality of civil aircraft fuselages due to its flexibility,extensive workspace,and high operability.In the polishing process of civil aircraft fuselages,there exists a mutual constraint relationship between surface roughness and material removal rate.There-fore,achieving optimal balance between surface roughness and material removal rate is a significant challenge in current research.To address this issue,this paper proposes a multi-objective optimization method for robot polis-hing processes based on GRA(grey relational analysis)-DBN(deep belief networks).Initially,principal component analysis(PCA)is employed to optimize GRA and establish the mapping relationship between multi-objectives and process parameters.Subsequently,the multi-objective optimization is transformed into a single-objective optimiza-tion problem.Furthermore,integrating DBN facilitates the construction of an optimization model for robot polishing process parameters.Experimental validation demonstrates that the proposed method not only effectively balances surface roughness and material removal rate but also enhances the stability and reliability of the polishing process.关键词
机器人打磨/民用飞机机身/多目标优化/灰色关联分析/深度置信网络Key words
robot polishing/civil aircraft fuselage/multi-objective optimization/grey relational analysis/deep belief network引用本文复制引用
陶永,薛蛟,杨林,王潇桐,刘亚醉,魏洪兴..基于GRA-DBN的机器人打磨工艺多目标优化方法[J].高技术通讯,2026,36(2):170-178,9.
