【作者】 李强；

【导师】 董海；

【作者基本信息】 沈阳大学 ， 工业工程（专业学位）， 2021， 硕士

【摘要】 随着全球经济的高速发展,生产制造业也进入了一个发展的新时期,随之而来的是企业间越来越激烈的市场竞争。工艺规划与车间调度作为制造系统中非常重要的两个部分,直接制约着整个制造系统的生产能力以及资源的有效利用率。由于客户个性化需求的不断提高,“种类多样化,生产批量更小”的这种生产方式被越来越多的生产制造类企业所使用。面对这种具有动态变化特征的市场需求,工艺规划系统与车间调度系统的有效集成已经变得越来越重要,本文将从这两个关键环节的集成策略以及相应算法设计展开研究。本文提出基于果蝇优化算法的柔性工艺规划研究方法,针对该问题,设计了三部分的编码方式分别处理加工次序柔性、加工工艺柔性以及加工机器柔性;对果蝇优化算法的嗅觉搜索阶段进行改进,采用插入、交换、变异三种操作构造邻域,并且引入全局协作机制,增强种群间的协作,从而对算法性能进行提升。通过算例验证果蝇优化算法在该问题上的有效性,为工艺规划与车间调度集成优化问题的研究奠定了基础。在柔性工艺规划问题的研究基础上,提出基于多目标果蝇优化算法的工艺规划与车间调度集成优化办法。本文以最小化最大完工时间、最小化最大机器负荷以及最小化总加工成本为研究目标,并且根据多目标特点将Pareto方法与果蝇优化算法相结合,提出了基于Pareto方法的多目标果蝇优化算法。在嗅觉搜索阶段采用插入的方式产生邻域;为了解决视觉搜索阶段个体更新的问题,采用相异度比较的办法对父代果蝇个体进行替换;全局协作阶段则使用POX交叉的方法进行全局协作。为了提高Pareto解集的质量,采用计算拥挤度距离的办法对外部档案中的非劣解进行更新维护。根据所提出的研究成果,进行数值实例验证,通过与多种算法的对比验证本文所提算法的有效性。并通过对某减速器生产企业的制造车间进行实际调研,发现该企业制造车间中真实存在加工时间过长、总加工成本过高等问题后,将本文研究成果实际应用于该案例,取得良好效果。更多还原

【Abstract】 With the rapid development of the global economy,the manufacturing industry has also entered a new period of development,followed by increasingly fierce market competition among enterprises.As two very important parts of the manufacturing system,process planning and job shop scheduling directly restrict the production capacity of the entire manufacturing system and the effective utilization of resources.Due to the continuous improvement of customers’ individual needs,the production method of "diversified types and smaller production batches" is used by more and more manufacturing enterprises.In the face of this dynamic market demand,the effective integration of the process planning system and the job shop scheduling system has become more and more important.This article will conduct research on the integration strategy of these two key links and the corresponding algorithm design.This paper proposes a research method for flexible process planning based on fruit fly optimization algorithm.In response to this problem,a three-part coding method is designed to deal with processing process flexibility,processing sequence flexibility and processing machine flexibility.The olfactory search stage of the fruit fly optimization algorithm is improved.Three operations of insertion,exchange,and mutation are used to construct neighborhoods,and a global collaboration mechanism is introduced to enhance the collaboration between populations,thereby improving the performance of the algorithm.An example is used to verify the effectiveness of the fruit fly optimization algorithm on this problem,which lays a foundation for the research on the integrated optimization problem of process planning and job shop scheduling.Based on the research of flexible process planning,an integrated optimization method of process planning and job shop scheduling based on multi-objective fruit fly optimization algorithm is proposed.This paper takes minimizing the maximumcompletion time,minimizing the maximum machine load and minimizing the total processing cost as the research goals,and combines the Pareto method with the fruit fly optimization algorithm according to the multi-objective characteristics,and proposes a multi-objective fruit fly optimization algorithm based on the Pareto method.In the olfactory search stage,the neighborhood is generated by inserting.In order to solve the problem of individual update in the visual search stage,the method of comparing the degree of dissimilarity is used to replace the individual of the parent fruit fly.In the global collaboration stage,the POX crossover method is used for global collaboration.In order to improve the quality of Pareto solution set,the method of calculating the congestion distance is used to update and maintain the non-inferior solutions in the external files.According to the proposed research results,numerical examples are verified,and the effectiveness of the algorithm proposed in this paper is verified by comparison with various algorithms.And through the actual investigation of the manufacturing workshop of a reducer manufacturer,it was found that the manufacturing workshop of the company had problems such as long processing time and high total processing cost.The research results of this article are actually applied to this case,and good results are achieved.更多还原