面向生产过程调度的数字孪生系统体系架构研究  

Research on Digital Twin System Architecture for Production Scheduling

作　　者：崔冰[1] 张林鍹[1] 欧阳亮 傅永康 CUI Bing;ZHANG Lin-xuan;OUYANG Linag;FU Yong-kang(Dapartment of Automation,Tsinghua University,Beijing 100084,China;Weichai Power Co.,Ltd.,Weifang Shandong 261061,China)

机构地区：[1]清华大学自动化系,北京100084 [2]潍柴动力集团,山东潍坊261000

出　　处：《计算机仿真》2023年第6期9-12,共4页Computer Simulation

摘　　要：针对大规模混流制造车间生产过程的调度问题,本文提出一种基于数字孪生系统的架构体系,结合复合调度模型,对大规模、复杂约束、不确定性的调度优化提供了较好的解决思路。明确了物理车间进行数据采集、清洗和融合的层次结构;提出了多层次多维度虚拟车间的构建机理,对多维度融合的基础单元模型构建进行了阐述;对复杂环境下的生产过程调度,提出了孪生数字驱动的传统模型、预测模型和知识模型的复合调度模型;最后对某大型发动机制造企业的复杂混流生产调度进行了实例验证。This paper presents the architecture of digital twin system for production scheduling in large-scale mixed flow manufacturing workshop,which provides a good solution to the complex scheduling optimization.In the pa-per,the hierarchical structure of data acquisition,data cleaning and data fusion in the physical workshop were defined;The construction mechanism of multi-level and multi-dimensional virtual workshops was proposed,and the construction of the basic unit model for multi-dimensional fusion was elaborated.A composite scheduling model of twin digital driven traditional models,predictive models,and knowledge models has been proposed for production process scheduling in complex environments.Finally,an example of mixed flow production scheduling in a large en-gine manufacturing enterprise was verified.

关 键 词：数字孪生 虚拟仿真 生产调度 离散制造 虚拟车间 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]