【作者】 彭超；

【导师】 方艺霖；

【作者基本信息】 武汉理工大学 ， 信息与通信工程， 2019， 硕士

【摘要】 在离散企业的生产制造过程中,车间生产是其基本单位,而车间调度是影响生产制造效率的决定性因素。随着智能制造新模式的出现,以及各种信息技术的融合应用,传统的车间调度模式不足以满足实时性、鲁棒性和准确性等需求,无法对车间生产制造过程中异常事件进行实时动态调整。同时由于数字孪生技术的兴起,其虚实交互、实时映射、共生演化的特点,引起了广泛的关注。因此将数字孪生应用于车间调度,是提高制造业生产力的一个重要途径。本文基于生产车间的数据感知和异常监测,实现数字孪生的实时映射与交互,考虑机器设备和操作工人两种资源的多目标柔性作业车间调度问题,提出一种改进的多目标优化算法,通过实时监测生产中的异常情况,进行调度方案的动态调整,实现车间生产调度的动态优化。主要研究内容如下:(1)面向车间的数据感知和虚实交互研究。针对生产制造车间数据种类多、来源广、结构多样性的特点,提出一种面向生产过程的全要素车间生产信息分析方法,在此基础上提出一种面向车间人机物环的全要素车间信息感知框架。研究基于车间感知数据的生产动态异常事件监测,提出基于RFID的工件异常事件监测和基于设备运行数据的状态识别,从而为车间动态调度调整提供基础。在此基础上,通过数字孪生虚实空间的信息差异分析,建立虚实空间的映射框架,提出一种基于事件监听和触发机制的虚实车间信息交互方法。(2)虚实演进的车间调度方法研究。针对传统的车间调度方法在实时响应等方面的不足,提出一种基于数字孪生的车间调度机制,并对该调度机制的运行原理进行研究。结合虚实交互动态调整,实现车间调度的虚实共同演进。考虑车间生产中的机器设备和操作工人两种资源的柔性,并以完成时间、加工成本、加工能耗、机器负载和交货延期为优化目标,提出一种双资源柔性的多目标作业车间调度问题,并通过引入改进的精英保留策略和邻域搜索算子,提出一种改进的遗传算法进行调度问题求解。基于生产异常事件,研究偏离小且稳定的车间生产动态重调度。最后,通过算法对比实验和实例仿真实验的结果分析,验证本文所提方法的有效性。(3)基于数字孪生的车间调度管理系统的设计与实现。在系统整体设计的基础上,以面向模块化开发的思想,设计车间生产调度管理系统,实现信息管理模块、信息监测模块以及车间调度模块的开发,并通过数字孪生车间模型实现动态调度仿真功能,最后对每个功能模块进行实现和测试,验证系统的有效性,实现车间生产调度的有效管理。更多还原

【Abstract】 In the manufacturing process of discrete enterprises,workshop production is the basic unit and job shop scheduling is the decisive factor affecting the manufacturing efficiency.With the emergence of intelligent manufacturing and the integration of various information technologies,the traditional job shop scheduling is not enough to meet the requirements of real-time,robustness and accuracy.And it is also impossible to dynamically adjust abnormal events in the process of workshop manufacturing.At the same time,due to the characteristics of virtual-real interaction,real-time mapping and co-evolution,digital twin has attracted wide attention.Therefore,the application of digital twin to job shop scheduling is an important way to improve manufacturing productivity.In this study,real-time mapping and interaction of digital twins are achieved based on data perception and abnormal events monitoring in production workshop.Considering the double resource multi-objective flexible job shop scheduling problem with machine and labor resources,an improved multi-objective optimization algorithm is proposed.Through real-time monitoring of abnormal conditions in production,the dynamic adjustment and optimization of scheduling plan are accomplished.The main research contents are as follows:(1)Research on workshop-oriented data perception and virtual-real interaction.Since that the workshop data have many types,wide sources and diverse structures,a total-elements workshop information analysis method is proposed.According to the characteristics of workshop data,a perception framework for workshop-oriented totalelements workshop information is proposed.The monitoring of abnormal production events based on workshop perception data is studied.The monitoring of abnormal production events based on RFID and the state recognition based on equipment operation data are proposed,which provide the basis for dynamic scheduling adjustment of workshop.The mapping framework between virtual and physical space is established through the analysis of information difference between the two space in digital twin,and an information exchange method in digital twin based on event monitoring and triggering mechanism is proposed.(2)Research on evolutionary workshop scheduling method between virtual and physical space.Aiming at the shortcomings of traditional job shop scheduling methods in real-time response,a digital twin-based job shop scheduling mechanism is proposed and its operation principle is studied.Combined with the dynamic adjustment and interaction,the virtual and physical evolution of job shop scheduling is achieved.Taking completion time,processing cost,processing energy consumption,machine load and delivery delay as optimization objectives,a double resource flexible multiobjective job shop scheduling problem considering the flexibility of machine and operator is proposed.By introducing improved elitism and neighborhood search operator,an improved multi-objective genetic algorithm is proposed to solve the scheduling problem.Based on abnormal production events,stable dynamic rescheduling of workshop production is studied.Finally,the effectiveness of the proposed method is verified by simulation experiments and result analysis.(3)Design and implementation of job shop scheduling management system based on digital twin.On the basis of the overall design of the system and the idea of modular development,the job shop scheduling management system is designed to achieve information management module,information monitoring module and job shop scheduling module.The dynamic scheduling simulation is accomplished through the digital twin workshop model.Finally,each module is implemented and tested to verify the effectiveness of the system and achieve effective management of the job shop scheduling.更多还原