基于混合灰狼算法实现第Ⅱ类机器人 U型装配线能耗优化
张子凯1,2 唐秋华1,2 张利平1,2 李梓响1,2 1.武汉科技大学冶金装备及其控制教育部重点实验室,武汉, 430081 2.武汉科技大学机械传动与制造工程湖北省重点实验室,武汉, 430081摘要:面向作业工序分配和机器分配,提出一种基于随机键编码的灰狼算法,以实现能耗最小化。算法在解码中,考虑工序间的优先关系约束和节拍约束,将工序和机器分配到工位中。该算法基于灰狼个体间的社会等级信息,选择3只最优狼指引剩余个体进化,以实现种群的更新。同时,该算法依据作业工序分配和机器人分配,混合了两种交叉方式以增强最低等级狼群间的交流。最后通过U型装配线的标杆案例,验证灰狼算法的有效性和优越性。
关键词:机器人U型装配线;灰狼算法;能耗;优化
中图分类号: TH16
DOI:10.3969/j.issn.1004⁃132X.2018.16.015 开放科学(资源服务)标识码(OSID) :
Hybrid Gray Wolf Optimization for Energy-efficient Type- Ⅱ Robotic
U-shaped Assembly Line
ZHANG Zikai1,2 TANG Qiuhua1,2 ZHANG Liping1,2 LI Zixiang1,2
1.Key Laboratory of Metallurgical Equipment and Control Technology,the Ministry of Education,
Wuhan University of Science and Technology,Wuhan,430081
2.Hubei Key Laboratory of Mechanical Transmission and Manufacturing Engineering,
Wuhan University of Science and Technology,Wuhan,430081
Abstract : Considering task assignment and robot assignment,gray wolf algorithm was proposed based on random key code to minimize energy consumption. Considering precedence relations and cycle time constraint,the algorithm allocated the tasks and robots into workstations in decoding phase.Based on social hierarchy in the population of gray wolf,3 optimal wolves were selected to guide other wolves to update the population.Simultaneously,based on the task assignment and robot allocation,2 crossover op⁃ erators were designed to enhance the communication among low ⁃ grade wolves. The proposed algorithm was proved to have great performance under benchmarks of U⁃shaped assembly lines.
Key words : robotic U⁃shaped assembly line;gray wolf algorithm;energy consumption;optimization
0 引言装配线是一种常用于汽车、电子等行业的产品制造系统。在装配线上,产品从入口线的工位依次进入下一工位进行装配,直到从出口线的工位出来。该过程中,工序的装配顺序需满足工序间的优先关系,且工位的加工时间应小于节拍。不同于传统的直线型装配线, U型装配线的同一工位既可以包含入口线工序,也可以包含出口线工
序,这使得U型装配线具有更高的生产效率和柔性。如今,自动化技术已逐渐应用到企业的装配系统,机器人开始逐渐替代工人在装配线上进行作业的装配。
MILTENBURC等 提出并构建了U型装配
[] 1线平衡的数学模型后,众多学者开始研究U型装配线平衡问题。GÖKÇEN等 采用最短路径法、
[] 2
OGAN等 提出分支定界法来求解简单U型装配
[] 3
线平衡问题。LI等 提出多规则启发式算法求解
[] 4
第二类U型装配线平衡问题。RABBBANI等同
[] 5时考虑混流U型装配线的工位负载和线效率,提