基于改进粒子群算法的阀门柔性装配线设计

发布时间：2018-12-10 09:24

【摘要】：本文所研究的课题来自于某企业的横向课题“分体式蝶阀的结构设计以及阀门柔性装配线的设计”。本文所研究的主要内容为阀门柔性装配线的设计,该问题包括两个方面的内容:阀门柔性装配线的平衡设计以及不同品种阀门投产排序的设计。本文首先对阀门柔性装配线进行数学建模。结合对企业实际调查的结果,本文设计了阀门柔性装配线的总体方案,并在此基础上,对阀门柔性装配线的技术参数进行计算,同时,完成对待装配阀门的结构以及装配工序的分析。针对粒子群算法在求解过程中出现的离散性差以及搜索能力低的缺点,本文提出了一种自适应的离散化算法,并结合模拟退火、禁忌搜索等算法的原理,提出了一系列提高其搜索能力的改进算法;通过不同算例的对比实验,验证了本文所提出的算法的有效性。由于不同品种阀门装配过程中装配任务以及装配时间存在差异,阀门柔性装配线平衡设计的难度大大增加,本文通过对不同阀门装配工序的分析与归并,建立了以工作站数目、工作站负载波动指数以及装配线耦合程度为优化目标的阀门柔性装配线模型;在算法求解过程中,结合启发式算法以及随机算法,设计了初始粒子群的生成算法,并通过归一化方法与多目标优化方法设计了适应度函数;在求解结果的基础上,通过EM-Plant仿真分析,完成缓冲区的设计。针对因阀门投产顺序不同造成的库存成本、时间成本的差异,本文在对设计完成的阀门柔性装配线分析的基础上,设计了以库存成本、工艺辅助时间、时间准备成本为优化目标的阀门投产排序模型;通过归一化与多目标优化的方法设计了适应度函数,完成投产排序的设计,并与只考虑工艺辅助时间的结果进行对比,说明了本文投产排序的优越性。通过上述的研究,本文最终设计了一条工作站数目最小、各工站负载波动较小、耦合性较好的阀门柔性装配线;并为企业设计了一个生产批次内的装配工作的安排,按照该投产顺序安排生产,企业的总成本最小。
[Abstract]:The research of this paper comes from the horizontal subject of a certain enterprise, "the structure design of the split butterfly valve and the design of the flexible assembly line of the valve". The main content of this paper is the design of valve flexible assembly line. This problem includes two aspects: the balance design of valve flexible assembly line and the design of sequencing of valve. In this paper, the valve flexible assembly line mathematical modeling. Combined with the result of the actual investigation, this paper designs the overall scheme of the valve flexible assembly line, and on this basis, calculates the technical parameters of the valve flexible assembly line, at the same time, Complete the analysis of assembly valve structure and assembly procedure. Aiming at the disadvantages of poor discreteness and low searching ability in particle swarm optimization (PSO), an adaptive discretization algorithm is proposed in this paper, which combines the principles of simulated annealing, Tabu search and so on. A series of improved algorithms to improve its search ability are proposed. The effectiveness of the proposed algorithm is verified by comparison experiments with different numerical examples. Due to the difference of assembly task and assembly time in different valve assembly process, the difficulty of balancing design of valve flexible assembly line is greatly increased. In this paper, the number of workstations is established by analyzing and merging different valve assembly procedures. The valve flexible assembly line model with the load fluctuation index of workstation and coupling degree of assembly line as the optimization objective; In the process of solving the algorithm, combining heuristic algorithm and stochastic algorithm, the algorithm of generating initial particle swarm optimization is designed, and the fitness function is designed by normalization method and multi-objective optimization method. On the basis of the solution, the buffer is designed by EM-Plant simulation. In view of the difference of inventory cost and time cost caused by different order of valve putting into operation, based on the analysis of the flexible assembly line of valve, this paper designs the cost of inventory and the auxiliary time of process. The sequencing model of valve commissioning with time preparation cost as the optimization objective; Through the method of normalization and multi-objective optimization, the fitness function is designed, and the design of commissioning sequencing is completed. Compared with the results with only considering the auxiliary time of process, the superiority of this paper is illustrated. Through the above research, this paper finally designed a valve flexible assembly line with minimum number of workstations, less load fluctuation and better coupling. An assembly arrangement in the production batch is designed for the enterprise. The total cost of the enterprise is minimum according to the sequence of production.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TH165

【相似文献】