柔性机器人制造单元调度算法研究及应用

发布时间：2018-05-09 03:37

本文选题：机器人制造单元 + FJSP　；参考：《广东工业大学》2017年硕士论文

【摘要】：随着市场竞争的加剧和先进制造技术不断发展,机器人制造单元被越来越广泛地应用于生产制造行业,生产调度方案作为机器人制造单元安全高效运转的基础,近年来逐渐成为学者研究的重点。为了进一步减低制造成本,均衡利用企业资源和适应小批量、多品种的生产模式,往往需要考虑工序在不同加工加床之间的搬运工序,对多个可能存在相互冲突的目标进行优化。因此,本文在经典车间调度理论基础上,对两种具有重要应用意义的机器人制造单元调度问题进行研究。重点研究问题的调度机制和求解算法。本文首先对柔性多机器人制造单元单目标调度问题进行研究,提出一种求解问题的混合蚁群算法。首先建立问题析取图模型,接着对传统蚁群算法的状态转移规则和信息素更新规则进行优化,在此基础上加入遗传算子和多机器人调度算法,得到考虑搬运工序的解集,通过迭代优化搜索最优解。通过多组基准算例测试,验证了所提算法的有效性和稳定性。并且通过对比实验证明了MMAS信息素更新方式比ACS信息素更新方式更能提升算法求解大规模问题的能力。本文第二个研究的问题是柔性机器人制造单元多目标调度问题。在分析了NSGA-II算法的局限性后,提出一种改进NSGA-II算法,对调度问题的最大完工时间,机床总负荷和瓶颈机床负荷三个目标进行优化。通过种群预筛选机制提升初始解的质量,设计了一种改进带循环拥挤度计算的精英选择策略,提升解的分布性的同时保留种群多样性,防止算法过早收敛。同样利用基准算例对算法进行测试,验证了所提算法的有效性和可靠性。最后,基于提出的混合蚁群算法和改进NSGA-II算法,设计开发机器人制造单元原型调度系统用于指导实际车间生产。
[Abstract]:With the intensification of market competition and the continuous development of advanced manufacturing technology, robot manufacturing units are more and more widely used in the manufacturing industry. The production scheduling scheme is the basis for the safe and efficient operation of robot manufacturing units. In recent years, scholars have gradually become the focus of research. In order to further reduce manufacturing costs, make balanced use of enterprise resources and adapt to small batch, multi-variety production patterns, it is often necessary to consider the handling process between different processing and adding beds. Optimize multiple objectives that may conflict with each other. Therefore, based on the classical job-shop scheduling theory, this paper studies two kinds of robot manufacturing cell scheduling problems which have important application significance. The scheduling mechanism and solving algorithm of the problem are studied in detail. In this paper, the single-objective scheduling problem for flexible multi-robot manufacturing cells is studied, and a hybrid ant colony algorithm is proposed to solve the problem. Firstly, the problem disjunctive graph model is established, then the state transition rules and pheromone updating rules of the traditional ant colony algorithm are optimized. On this basis, genetic operators and multi-robot scheduling algorithms are added to get the solution set considering the handling process. The optimal solution is searched by iterative optimization. The validity and stability of the proposed algorithm are verified by a number of benchmark examples. The comparison experiments show that the MMAS pheromone updating method is better than the ACS pheromone updating method to improve the ability of the algorithm to solve large-scale problems. The second problem in this paper is the multi-objective scheduling problem of flexible robot manufacturing cells. After analyzing the limitation of NSGA-II algorithm, an improved NSGA-II algorithm is proposed to optimize the maximum completion time of scheduling problem, total load of machine tool and bottleneck load of machine tool. By means of population pre-screening mechanism to improve the quality of initial solution, an elite selection strategy with cyclic congestion calculation is designed to improve the distribution of solution while preserving population diversity, so as to prevent the algorithm from converging prematurely. A benchmark example is also used to test the algorithm, which verifies the validity and reliability of the proposed algorithm. Finally, based on the proposed hybrid ant colony algorithm and improved NSGA-II algorithm, a prototype scheduling system for robot manufacturing units is designed and developed to guide actual workshop production.
【学位授予单位】：广东工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH165

【参考文献】