基于遗传算法的电容器智能制造系统设计与实现

发布时间：2018-01-31 15:18

本文关键词： 智能制造制造执行系统生产调度遗传算法　出处：《广东工业大学》2017年硕士论文　论文类型：学位论文

【摘要】：在“中国制造2025”的发展战略背景下,我国制造业正面临新一轮的转型升级。智能制造借助物联网、大数据、云计算、人工智能等新一代信息技术正在快速发展,智能制造系统的应用更是为实现智能工厂奠定了坚实的基础。实现信息化与工业化的深度融合将推动我国制造业更上一个台阶。在企业信息化管理系统体系中,制造执行系统(MES)占据着关键位置。本文总结分析了电容器制造行业的信息化发展现状及生产特点,对制造系统模型进行了研究分析。以电容器制造过程中的卷材切割最优化问题、生产调度问题和系统实现作为研究重点,针对这些关键问题和相关技术进行了系统的研究。本文主要研究内容如下:(1)分析了智能制造和MES研究现状,分析了制造系统体系模型,总结了MES功能体系三层模型及各层之间的信息交互,对生产管理模型和生产调度模型进行了深入研究,对基础静态信息定义区域、生产调度指令下达区域和生产绩效统计反馈区域做了详细的阐述,解释了生产调度中各信息流在模块间的传递,并对MES功能需求和性能需求进行了分析。(2)针对电容器制造过程中整卷原材料切割方案优化问题进行抽象,建立了数学模型,模型考虑了实际生产过程中原材料规格不统一和裁切目标多样化的特点,设计了多目标评价函数和约束条件。通过改进的基于偏好的遗传算法对该实际问题进行理论推导,最终通过实例数据进行仿真试验,仿真结果验证了改进算法的有效性和稳定性。(3)结合电容器实际生产车间的特点,总结分析了电容器多工序生产流程,多个工序都存在并行机的特点,建立了电容器生产车间调度模型。该模型属于典型流水车间调度模型。考虑到电容器实际生产过程中,由于让同一台机器生产不同规格的产品是需要通过调整机器设备硬件结构和软件运行参数,这会使生产调度中除了生产时间外还需考虑换批次时的改机时间,所以建模时已经把改机时间影响因素考虑到调度模型中。研究分析了流水车间调度求解的思路,采用遗传算法对该模型进行求解。由于传统遗传算法在收敛速度和全局搜索能力都存在一些缺陷,本文利用改进的自适应遗传算法,让选择概率和交叉概率随种群进化过程的最优适应度和平均适应度进行自适应调整。实例数据仿真运算结果证明本文改进的算法对于电容器制造企业生产车间调度问题求解的有效性与稳定性。(4)设计并实现了电容器MES,对系统功能框架和网络结构拓扑图进行了详细分析。对系统中涉及的各功能子系统做了详细阐述,通过实际使用效果分析了该MES系统能满足电容器制造企业的需求,给企业带来了经济效益与实际使用价值,提升了电容器制造企业的管理水平。
[Abstract]:In the context of the development strategy of "made in China 2025", our manufacturing industry is facing a new round of transformation and upgrading. Intelligent manufacturing with the help of the Internet of things, big data, cloud computing. New generation of information technology, such as artificial intelligence, is developing rapidly. The application of intelligent manufacturing system has laid a solid foundation for the realization of intelligent factory. Realizing the deep integration of information technology and industrialization will push the manufacturing industry to a higher level in the enterprise information management system. Manufacturing execution system (mes) occupies a key position. This paper summarizes and analyzes the current situation of information development and production characteristics of capacitor manufacturing industry. The model of manufacturing system is studied and analyzed. The research emphasis is on the optimization of coil cutting, production scheduling and system realization in the process of capacitor manufacture. The main contents of this paper are as follows: (1) the research status of intelligent manufacturing and MES is analyzed, and the model of manufacturing system is analyzed. This paper summarizes the three-tier model of MES function system and the information exchange between each layer, deeply studies the production management model and production scheduling model, and defines the basic static information region. The region of production scheduling and the region of statistical feedback of production performance are elaborated in detail, and the transfer of information flow between modules in production scheduling is explained. The functional and performance requirements of MES are analyzed. (2) aiming at the optimization of the cutting scheme of whole roll raw materials in capacitor manufacturing process, the mathematical model is established. The model takes into account the disunity of raw material specifications and the diversification of cutting targets in the actual production process. The multi-objective evaluation function and constraint conditions are designed. The theoretical derivation of the practical problem is carried out through the improved genetic algorithm based on preference. Finally, the simulation experiment is carried out through the example data. The simulation results verify the effectiveness and stability of the improved algorithm. Combined with the characteristics of the actual capacitor production workshop, the multi-process process of capacitor production is summarized and analyzed, and the characteristics of parallel machines are found in many processes. A workshop scheduling model for capacitor production is established. The model belongs to the typical income workshop scheduling model and takes into account the actual production process of capacitors. Because it is necessary to adjust the hardware structure and software operating parameters for the same machine to produce products of different specifications, this will make the production scheduling in addition to the production time also need to consider the time of changing the machine when changing batches. Therefore, the influence factors of machine modification time have been taken into account in the scheduling model, and the idea of income job shop scheduling solution has been studied and analyzed. Genetic algorithm (GA) is used to solve the model. Because the traditional genetic algorithm has some defects in convergence speed and global search ability, the improved adaptive genetic algorithm is used in this paper. The selection probability and crossover probability are adaptively adjusted with the optimal fitness and average fitness of the population evolution process. The simulation results of the example data show that the improved algorithm is suitable for the production workshop of capacitor manufacturing enterprises. The validity and Stability of solving the degree problem. 4) the capacitor MES is designed and implemented. The functional framework and topology diagram of the system are analyzed in detail, and the functional subsystems involved in the system are described in detail. The MES system can meet the needs of capacitor manufacturing enterprises, bring economic benefits and practical use value to the enterprises, and improve the management level of capacitor manufacturing enterprises.
【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】：TP18;TM53

【参考文献】