基于异步智能算法的生产调度问题的研究

发布时间：2018-03-29 11:13

本文选题：Flow　切入点：Shop调度问题　出处：《华东理工大学》2012年硕士论文

【摘要】：生产调度在计算机集成制造系统中是连接管理层与监控层的枢纽,它通过传递决策层的经营管理决策,向监督控制层下达指令,以保证企业生产有条不紊的进行,是流程工业中能否成功实施CIMS的关键。调度问题与企业的利益最大化是紧密相关的,对于我国现代化生产制造过程的发展起着至关重要的作用,其中Flow Shop调度问题是一个非常典型的生产调度问题。本文通过引入异步进化策略设计与改进现存的智能优化算法用于解决Flow Shop调度问题。根据大量的实验结果证明,异步智能算法对于解决生产调度问题是非常有效的。对于以总流程时间为目标的置换Flow Shop问题,提出一种改进的粒子群算法。该算法通过引入最小位置值(SPV)规则,把粒子的各位置分量按照由小到大排序,将粒子位置映射到置换Flow Shop问题的解空间。同时采用变邻域局部搜索机制对父代个体执行不同代数的搜索以实现异步行为来增加种群的多样性,从而获得更快的收敛速度及更好的解。仿真结果表明了该算法的有效性。对于以Makespan为目标函数的置换Flow Shop调度问题,在改进蛙跳算法NSFLA的基础上,结合异步的理念,提出了一种全新的算法异步蛙跳算法(ASFLA)用于解决置换Flow Shop调度问题。ASFLA全局通过采用洗牌策略加强全局信息交换,局部搜索则采用异步概念增强种群的多样性。仿真实验表明蛙跳算法经过异步化之后求解性能显著提高。对于一类加工时间不确定的以总流程时间为目标的置换Flow Shop调度问题,应用模糊数学的方法来表示加工时间的不确定性,采用一种改进的智能算法异步遗传局部搜索算法(AGLA)。该算法的一个初始解是由构造型启发式算法产生,其他则是随机产生,然后通过引入变邻域搜索机制和简单交叉算子,对种群执行异步进化操作,算法最后加入重启策略防止陷入局部极小。仿真实验结果验证了AGLA解决模糊Flow Shop问题的有效性。
[Abstract]:Production scheduling is a key link between management and monitoring layer in the computer integrated manufacturing system. By transmitting the decision of management and management at the decision-making level, the production scheduling can give instructions to the supervision and control layer, so as to ensure that the production of the enterprise is carried out methodically. Scheduling problem is closely related to the profit maximization of enterprises and plays an important role in the development of modern manufacturing process in China. The Flow Shop scheduling problem is a typical production scheduling problem. This paper introduces asynchronous evolutionary strategy design and improves the existing intelligent optimization algorithm to solve the Flow Shop scheduling problem. Asynchronous intelligent algorithm is very effective to solve production scheduling problem. An improved particle swarm optimization (PSO) algorithm is proposed for the permutation Flow Shop problem with total flow time as the target. By introducing the minimum position value (Flow) rule, each position component of the particle is sorted from small to large. The particle position is mapped to the solution space of the permutation Flow Shop problem. At the same time, variable neighborhood local search mechanism is used to perform different algebraic searches on parent individuals to achieve asynchronous behavior to increase population diversity. In order to obtain faster convergence speed and better solution, the simulation results show the effectiveness of the algorithm. For the permutation Flow Shop scheduling problem with Makespan as the objective function, the idea of asynchronous is combined with the improved leapfrog algorithm NSFLA. In this paper, a novel asynchronous leapfrog algorithm (ASFLAA) is proposed to solve the replacement Flow Shop scheduling problem. The global shuffle strategy is used to enhance the global information exchange. The local search uses the asynchronous concept to enhance the diversity of the population, and the simulation results show that the performance of the leapfrog algorithm is improved significantly after the asynchronous algorithm. For a class of permutation Flow Shop scheduling problems with uncertain processing time, the uncertainty of processing time is expressed by fuzzy mathematics method. An improved intelligent algorithm named Asynchronous genetic Local search algorithm is used. One of the initial solutions of the algorithm is generated by the constructive heuristic algorithm, and the others are randomly generated, and then by introducing variable neighborhood search mechanism and simple crossover operator, The asynchronous evolution operation is performed on the population, and the restart strategy is added to prevent the population from falling into local minimization. The simulation results show that AGLA is effective in solving the fuzzy Flow Shop problem.
【学位授予单位】：华东理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH186;TP18

【参考文献】