采用新型历史学习果蝇算法的机器人去毛刺多参数优化  

Optimization of Robot Deburring Process Parameters Based on Novel History Study Fruit Fly Optimization Algorithm

作　　者：唐方红[1] 胡华荣[2] TANG Fanghong;HU Huarong(School of Mechanical Engineering Dongguan polytechnic,Dongguan Guangdong 523808,China;School of Mechanical Kngineering Hunan University of Science and Technology,Xiangtan Hunan 411100,China)

机构地区：[1]东莞职业技术学院机电工程学院,广东东莞523808 [2]湖南科技大学机电工程学院,湖南湘潭411100

出　　处：《机械设计与研究》2021年第3期105-108,共4页Machine Design And Research

基　　金：湖南自然科学基金资助项目(2019JJ40084)。

摘　　要：针对机器人去毛刺时的工艺参数设置往往依靠经验的不足,采用新型历史学习果蝇算法(NHSFOA)的机器人去毛刺多参数。相比于果蝇算法(FOA),NHSFOA对果蝇个体位置更新方式进行了改进,变向最优个体所在位置飞去的方式为向最优个体位置和历史位置协同确定的位置飞去,在保证种群多样性的同时增加了算法的搜索能力。进行了去毛刺工艺的仿真分析和工程应用,试验结果表明NHSFOA缩短了单位长度毛刺的去除时间,提高了加工效率。Aiming at the inadequacy that the process parameter of robot deburring heavily depend on experience,a robot deburring process parameters optimization method based on novel history study fruit fly optimization algorithm(NHSFOA)is proposed.Compared with(FOA),fruit fly optimization algorithm,NHSFOA improved the updating mode of fruit individual position,and changes the way to fly to the location of the optimal individual to the location co-determined by the optimal individual position and historical position,which not only ensures population diversity,but also increases the searching ability of the algorithm.The simulation analysis and engineering application of deburring process are carried out.The experimental results show that NHSFOA can shorten the removal time of burring per unit length and improve the processing efficiency.

关 键 词：历史学习 果蝇算法 参数优化 机器人 去毛刺 

分 类 号：TH165.3[机械工程—机械制造及自动化]