基于混合算法的车间布局多目标优化及仿真研究

发布时间：2018-08-13 09:56

【摘要】：一直以来,制造系统的设备布局规划设计都被视为工业系统中最关键和最困难的任务之一,是提高企业生产效率、降低运营成本的重要因素。改革开放以来,我国制造业持续快速发展,但与世界先进水平仍存在一定差距,尤其在设备布局规划设计上缺乏足够的重视,工厂的布局和改善主要依据设计者的经验。因此,本文以车间设备布局优化为研究对象,研究了该类问题的建模、求解以及仿真优化。本文的主要研究工作如下:(1)设备布局优化问题在理论上属于组合优化问题,在求解上属于NP-hard问题。本文借鉴博弈论中的相关思想,提出一种双子群—双策略的混合算法。该算法通过支付效用矩阵,灵活控制算法的分配和调用顺序,为参与寻优的每个子群选择最佳的搜索算法,实现双子群独立并行优化。将所提出的混合机制应用到函数优化中,仿真结果表明:当混合策略中的单种优化算法具有不同的搜索特性时,所提出的混合算法是行之有效的,且在搜索精度和收敛速度等方面均表现出良好的性能。(2)设备布局问题在工程实际中往往有多个设计优化目标,这些目标相互联系,彼此制约,有的还相互矛盾,一个目标性能的改善可能导致其他目标性能的下降,即不存在一个方案能使所有的目标同时达到最优。因此,权重设置是多目标优化中的难点。基于此,本文提出采用多准则决策领域中的优先算子来计算多目标优化中的权重,使得权重的大小和候选方案的属性值密切相关,且有效避免了不同目标之间的补偿交易机制。(3)针对多行设备布局多目标优化问题,本文以Y公司的设备布局为研究对象,建立了以物料搬运费用最小化和面积利用率最大化为优化目标的数学模型,并采用基于博弈论的PSO-CS混合算法进行了求解。同PSO、CS原单种优化算法以及GA算法相比,PSO-CS在物料搬运费用和面利用率两个目标上都取得了更好的优化效果,从而得到了Y公司机加区设备布局的初始方案。(4)由于在建模过程中作出了一系列的简化和假设,使得基于数学模型求解的布局方案与工程实际存在一定的差异。因此,本文采用优化算法求解与计算机仿真技术相结合的方式,对PSO-CS求解的布局方案运用Plant Simulation进行仿真分析,并针对存在的问题提出相应的改善意见,通过仿真验证表明,改进后的布局方案在提高产能、降低物流量和平衡机器负荷等方面均有较好的改进效果。
[Abstract]:The equipment layout planning and design of manufacturing system has been regarded as one of the most critical and difficult tasks in industrial system, which is an important factor to improve the production efficiency and reduce the operating cost. Since the reform and opening up, the manufacturing industry of our country has been developing continuously and rapidly, but there is still a certain gap with the advanced level of the world, especially the lack of enough attention to the layout and design of the equipment, the layout of the factory and the improvement of the main designer's experience. Therefore, the modeling, solving and simulation optimization of this kind of problem are studied in this paper. The main work of this paper is as follows: (1) the equipment layout optimization problem belongs to combinatorial optimization problem in theory and NP-hard problem in solving. In this paper, we propose a hybrid algorithm of two subgroups and two strategies for reference to the related ideas in game theory. By means of payment utility matrix and flexible control of the allocation and call order of the algorithm, the algorithm selects the best search algorithm for each subgroup participating in the optimization, and realizes the two-subgroup independent parallel optimization. The proposed hybrid algorithm is applied to function optimization. The simulation results show that the proposed hybrid algorithm is effective when the single optimization algorithm in the hybrid strategy has different search characteristics. And it shows good performance in search accuracy and convergence speed. (2) equipment layout problems often have multiple design optimization objectives in engineering practice, which are related to each other, restrict each other, and some of them contradict each other. The improvement of the performance of one target may lead to the degradation of the performance of other targets, that is, there is no one scheme to achieve the optimal performance of all objectives at the same time. Therefore, weight setting is a difficult problem in multi-objective optimization. Based on this, this paper proposes to use the priority operator in the field of multi-criteria decision-making to calculate the weight in the multi-objective optimization, so that the size of the weight is closely related to the attribute value of the candidate scheme. And the compensation transaction mechanism between different targets is avoided effectively. (3) aiming at the multi-objective optimization problem of multi-row equipment layout, this paper takes the equipment layout of Y Company as the research object. A mathematical model with minimization of material handling cost and maximization of area utilization ratio is established and solved by PSO-CS hybrid algorithm based on game theory. Compared with PSO-CS single optimization algorithm and GA algorithm, PSO-CS has better optimization effect on material handling cost and surface utilization ratio. Thus, the initial layout scheme of Y driver plus area equipment is obtained. (4) due to a series of simplification and assumptions made in the modeling process, the layout scheme based on mathematical model is different from the actual engineering. Therefore, in this paper, the optimization algorithm is combined with computer simulation technology to simulate and analyze the layout scheme of PSO-CS solution using Plant Simulation, and the corresponding improvement suggestions are put forward in view of the existing problems. The simulation results show that, The improved layout has a good effect in increasing production capacity, reducing material flow and balancing machine load.
【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TB491

【相似文献】