基于组合拍卖机制的分布式柔性车间优化调度的研究

发布时间：2018-10-15 15:37

【摘要】：分布式生产制造系统由多个分处异地的车间组成，通过长期的或临时的合作，以松散的或者紧密的配合，完成共同的制造任务目标，因而具有较强的动态性、不确定性和复杂性。其生产调度是一个多对象、多目标优化问题，比常规的调度优化问题更难求解，需要寻求一种更加有效的调度策略，使其不仅能获得最优或接近最优的调度结果，还能对生产制造过程中的变化做出快速响应。本文以联盟企业制造模式为例，对分布式柔性生产车间生产调度问题进行研究，以期为制造车间优化生产过程、提高经济效益提供有益参考和工具。为解决动态环境下的生产调度问题，首先构建制造系统的控制结构模型，然后提出一种基于组合拍卖机制的分布调度策略，，设计蚁群优化算法。针对联盟制造模式的特点，在对分布式生产调度问题分析的基础上，提出一种分布-递阶混杂控制结构，结构内容主要包括任务的分配和柔性车间内优化生产调度。基于控制结构的运行原理，分别对任务分配问题及生产调度优化问题进行研究，重点分析任务分配问题。将组合拍卖机制运用到任务分配中，并详细地论述了任务分配过程中拍卖机制的运作过程，给出基于组合拍卖机制的任务分配模型。为进一步解决生产调度问题，根据每个分布企业具有柔性车间的特征，建立了柔性车间生产调度模型。为了验证所提出的基于组合拍卖机制的调度策略的可行性，充分应用蚁群算法在解决离散组合优化问题上的优势，本文针对任务分配及优化调度问题，设计蚁群优化算法。算法的设计主要从选择机制和信息素更新机制上进行优化。最后，本文以MATLAB R2010a为开发环境，采用MATLAB为编程语言，对研究的调度策略进行验证。结果既证明了算法的有效性，也说明本文针对分布式柔性车间提出的调度策略能够达到满意结果。
[Abstract]:The distributed manufacturing system is composed of many workshops located in different places. Through long-term or temporary cooperation and loose or close cooperation, the distributed manufacturing system accomplishes the common manufacturing task goal, so it has a strong dynamic character. Uncertainty and complexity. The production scheduling is a multi-object, multi-objective optimization problem, which is more difficult to solve than the conventional scheduling optimization problem. It needs to seek a more effective scheduling strategy, so that it can not only obtain the optimal or near optimal scheduling results. Can also make rapid response to the changes in the manufacturing process. This paper takes the manufacturing model of alliance enterprise as an example to study the production scheduling problem of distributed flexible production shop in order to provide useful reference and tools for optimizing production process and improving economic benefit of manufacturing shop. In order to solve the production scheduling problem in dynamic environment, the control structure model of manufacturing system is first constructed, and then a distributed scheduling strategy based on combinatorial auction mechanism is proposed to design ant colony optimization algorithm. Based on the analysis of distributed production scheduling problem, a distributed hierarchical hybrid control structure is proposed, which mainly includes task allocation and flexible production scheduling. Based on the operation principle of the control structure, the task assignment problem and the production scheduling optimization problem are studied respectively, and the task assignment problem is analyzed emphatically. The combination auction mechanism is applied to the task assignment, and the operation process of the auction mechanism in the task assignment process is discussed in detail, and the task assignment model based on the combination auction mechanism is given. In order to further solve the problem of production scheduling, according to the characteristics of each distributed enterprise with flexible workshop, the production scheduling model of flexible workshop is established. In order to verify the feasibility of the proposed scheduling strategy based on combinatorial auction mechanism and make full use of the advantages of ant colony algorithm in solving discrete combinatorial optimization problems, this paper designs an ant colony optimization algorithm for task allocation and optimal scheduling problems. The algorithm is designed from the selection mechanism and pheromone updating mechanism. Finally, using MATLAB R2010a as the development environment and MATLAB as the programming language, this paper verifies the proposed scheduling policy. The result not only proves the validity of the algorithm, but also shows that the scheduling strategy proposed in this paper for distributed flexible job shop can achieve satisfactory results.
【学位授予单位】：沈阳工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TH186;TP18

【参考文献】