基于协同进化的混合智能优化算法在SMT中的应用研究
发布时间：2019-04-11 17:42
【摘要】：在电子产品制造过程中,利用先进的优化技术对生产线及其设备进行合理调度和优化控制,是提高生产效率和资源利用率的重要方法。因此,探索和研究表面组装生产线及其关键设备——贴片机优化运行的相关理论问题及优化方法,具有重要的理论意义和应用价值。本论文研究了表面组装生产过程中以最小化PCB板组装生产时间为目标的贴片机优化问题。该优化问题涉及多个优化子问题,这些问题之间高度关联,使得整个优化问题非常复杂和难以求解。因此,在建立问题数学模型的基础上,如何设计有效的求解算法,缩短贴片机的贴装时间,提高贴片机和SMT生产线的生产效率,是本文研究的主要内容。论文对目前有关表面组装生产优化问题的研究思路和方法进行了系统的分析、分类、比较和总结,对以最小化PCB板组装时间的单个贴片机优化问题进行了分析,提出了基于协同进化的单机PCB板组装时间问题的求解方法。我们将单机优化问题分解为多个,并由此建立了二阶段的数学模型:第一阶段建立吸嘴分配的整数规划模型,并运用启发式的方法进行求解。第二阶段建立喂料器分配问题和元器件拾贴顺序问题的混合整数规划模型,运用协同进化的思想,结合多智能体进化算法的核心思想和局部搜索算法的优势对其进行求解,最终得出问题的最优解。
[Abstract]:In the manufacturing process of electronic products, it is an important method to improve the production efficiency and resource utilization rate by using the advanced optimization technology to reasonably schedule and control the production line and its equipment. Therefore, it is of great theoretical significance and application value to explore and study the theoretical problems and optimization methods of the optimal operation of the surface assembly line and its key equipment-the placement machine. In this paper, the optimization problem of the placement machine with the goal of minimizing the production time of PCB board assembly is studied in the process of surface assembly. The optimization problem involves a number of optimization sub-problems, which are highly related to each other, which makes the whole optimization problem very complex and difficult to solve. Therefore, on the basis of establishing the mathematical model of the problem, how to design an effective algorithm to solve the problem, shorten the mounting time of the mounting machine, and improve the production efficiency of the mounting machine and the SMT production line are the main contents of this paper. This paper systematically analyzes, classifies, compares and summarizes the current research ideas and methods on the optimization of surface assembly production, and analyzes the optimization problem of single mount machine with the aim of minimizing the assembly time of PCB board. A co-evolution-based method for solving the assembly time problem of stand-alone PCB boards is presented in this paper. We decompose the single-machine optimization problem into several, and then establish a two-stage mathematical model: in the first stage, the integer programming model of nozzle allocation is established, and the heuristic method is used to solve the problem. In the second stage, the mixed integer programming model of feeder assignment problem and component pick-up sequence problem is established, which is solved by the idea of co-evolution, combined with the core idea of multi-agent evolutionary algorithm and the advantages of local search algorithm. Finally, the optimal solution of the problem is obtained.
【学位授予单位】：华南理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TN05
本文编号：2456627
[Abstract]:In the manufacturing process of electronic products, it is an important method to improve the production efficiency and resource utilization rate by using the advanced optimization technology to reasonably schedule and control the production line and its equipment. Therefore, it is of great theoretical significance and application value to explore and study the theoretical problems and optimization methods of the optimal operation of the surface assembly line and its key equipment-the placement machine. In this paper, the optimization problem of the placement machine with the goal of minimizing the production time of PCB board assembly is studied in the process of surface assembly. The optimization problem involves a number of optimization sub-problems, which are highly related to each other, which makes the whole optimization problem very complex and difficult to solve. Therefore, on the basis of establishing the mathematical model of the problem, how to design an effective algorithm to solve the problem, shorten the mounting time of the mounting machine, and improve the production efficiency of the mounting machine and the SMT production line are the main contents of this paper. This paper systematically analyzes, classifies, compares and summarizes the current research ideas and methods on the optimization of surface assembly production, and analyzes the optimization problem of single mount machine with the aim of minimizing the assembly time of PCB board. A co-evolution-based method for solving the assembly time problem of stand-alone PCB boards is presented in this paper. We decompose the single-machine optimization problem into several, and then establish a two-stage mathematical model: in the first stage, the integer programming model of nozzle allocation is established, and the heuristic method is used to solve the problem. In the second stage, the mixed integer programming model of feeder assignment problem and component pick-up sequence problem is established, which is solved by the idea of co-evolution, combined with the core idea of multi-agent evolutionary algorithm and the advantages of local search algorithm. Finally, the optimal solution of the problem is obtained.
【学位授予单位】：华南理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TN05
【参考文献】
相关期刊论文 前1条
1 周泓;王建;谭小卫;;一种求解集成生产计划的混合协同进化算法[J];计算机集成制造系统;2007年07期
，本文编号：2456627
本文链接：https://www.wllwen.com/kejilunwen/dianzigongchenglunwen/2456627.html
