基于多智能体的数字孪生及其在工业中应用的综述

数字孪生是一种将物理实体数字化的技术,通过建立虚拟的数字孪生模型模拟实际的物理过程,以便进行模拟仿真、数据分析和优化设计等操作.鉴于此,分析数字孪生技术在复杂工业生产中的发展历程和研究现状,并重点讨论其概念、国家相关重点研究的政策,以及数字孪生使能技术在各行业的应用.主要途径是分析和综述基于多智能体的数字孪生、基于数字孪生的设计、制造和运维、数字孪生的集成在智能制造中的应用相关的研究成果.此外,提出高炉连续生产数字孪生方案和大飞机多智能体离散制造方案,高炉模型包括成分场大模型和增量学习小模型,该模型可以为数字孪生在复杂流程工业中的应用提供带有增量补偿的机理与计算机视觉相结合的解决方案.在复杂工业制造中,数字孪生和多智能体技术可以提高生产效率和质量,减少能源消耗和废品产生,同时也能够降低复杂度、安全风险和成本.     

智能制造下产品数字孪生体全生命周期研究

针对信息物理融合系统在智能制造领域中的应用新范式,提出了产品数字孪生体全生命周期的理论框架.数字孪生的出现为信息物理融合理念的实现提供可行的思路和途径.从数字孪生概念的发展背景出发,分别从物理对象和物理对象的发展过程分析数字孪生和数字孪生体的定义,在此基础上进一步分析了数字孪生体全生命周期的内涵和特征.数字孪生体全生命...     

数字孪生: 跨界赋能于多领域智能的新应用

随着大数据、5G、人工智能、CPS、云计算、物联网技术的发展与交叉融合, 使得世界朝着数字化、智能化方向发展. 数字孪生是以物理实体为原型建立多维虚拟模型, 通过安装在物理本体上的传感器实时反馈数据, 并结合以往的历史数据和人工智能技术, 最后利用软件分析并呈现. 由于数字孪生技术能与多个先进理念, 如: 工业4.0、航空航天、智慧城市、智慧医疗等良好的融合并应用, 这使其成为多个行业的热门研究方向与主要驱动技术, 在各行各业都有很大的发展空间. 本文首先阐述了数字孪生技术的基本概念, 梳理了数字孪生技术的发展脉络, 进一步理清了数字孪生技术与CPS技术之间的关系, 并介绍了数字孪生技术的研究现状. 其次, 介绍了数字孪生的关键技术即多维多尺度建模, 孪生数据管理和虚拟呈现. 最后, 探讨了数字孪生技术在智慧工厂领域、智慧城市领域、孪生医疗领域、航空航天领域的应用发展和方向, 并从方案、特点、关键技术等角度介绍了本研究团队在智慧工厂领域对原稳加热炉设备的数字孪生应用案例.     

数字孪生与生产线仿真技术研究

随着新一代信息技术与制造技术的深度融合,仿真技术正向着信息系统与物理系统深度融合和高效协同的方向发展,应运而生的数字孪生技术成为近年来的新兴研究热点.为探索采用新一代信息技术促进离散制造行业数字化转型升级,从仿真技术的发展脉络和应用场景入手,引入基于数字孪生技术开展生产线仿真分析的研究,介绍了数字孪生技术的概念和内涵,...     

基于数字孪生技术的船舶管件加工智能车间研究

数字孪生技术的研究成果在智能化制造领域得到推广应用.这已经引起国内外相关学术界与企业的高度关注.为研究数字孪生技术与智能车间生产融合应用效果,以船舶管件加工车间为研究对象,分析了传统管件加工车间的生产特点.采用数字孪生技术,建立了从管材入库、上料、管线切割、打磨、贴标、焊接到管零件制作完成的数字模型,创新构建了虚实结合...     

流程工业数字孪生关键技术探讨

流程工业是制造业的重要组成部分, 是国民经济发展的重要基础, 主要包括化工、冶金、石化等行业, 其安全高效的生产对国家而言具有重要的战略意义. 然而, 流程工业物理化学变化反应复杂、流程间能质流严重耦合、多目标冲突、在线实验风险大, 给生产流程系统建模与高效协同优化带来极大困难, 严重制约了生产质量和资源利用率的进一步提升. 随着信息技术与人工智能的发展, 建立虚实结合、协同优化运行的流程工业数字孪生生产线所需技术逐渐成熟, 其在流程工业的应用价值与潜力日益凸显. 本文首先阐述数字孪生在流程工业应用的必要性与重要性, 并通过边界定义法将数字孪生与信息物理系统(Cyber-physical system, CPS)、工业互联网等概念进行对比分析,从而明确数字孪生的基本内涵与功能边界. 其次描述流程工业抽象模型和数字孪生理论模型间的映射关系, 并分析了如何用数字孪生技术解决流程工业系统建模与高效协同优化的瓶颈问题. 最后, 从数字孪生系统构建的角度探讨数字孪生发展的关键技术, 并以一条炼铁生产线为例, 展示数字孪生技术在实际工业中的应用解决方案.     

数字孪生驱动的火箭控制系统故障诊断研究综述

在第四次工业革命中,智能制造成为各国工业发展的重点方向,数字孪生技术作为一项新兴技术,能够有效实现物理信息的融合;将其应用于火箭控制系统的故障诊断和健康管理,能够进一步提高故障诊断的事前准确性,提升火箭发射的可靠性;本文对数字孪生技术在航天领域的研究现状进行归纳整理;首先梳理了NASA的数字孪生目标,国内领域按照设计、生产、支持服务阶段对数字孪生应用进行分类;其次,按照故障树、专家系统、神经网络、数据驱动的方法阐述控制系统故障诊断的研究现状;在介绍数字孪生驱动的健康管理方法的基础上,提出数字孪生驱动的火箭控制系统的故障诊断方法;详细介绍其基本组成框架,分析关键技术及应用难点,并提出数字孪生健康管理平台的基本流程;该方法预期实现火箭控制系统的事前诊断和维修策略的制定     

可靠性数字孪生概念内涵和技术框架研究

在武器装备可靠性与数字孪生技术的基础上,提出了覆盖"全系统、全特性、全过程"的数字孪生体系结构,明确了可靠性数字孪生的定义,界定了可靠性数字孪生与性能数字孪生的相互关系.并且,从系统层级、系统特性、系统过程3个维度分析了产品数字孪生框架,以此为基准构建了包括故障物理模型、传感器数据监测、健康信息感知、使用与维修映射、数...     

基于数字孪生的力能控制式压力机虚拟调试

为降低调试周期和成本,提前预知系统级风险,基于数字孪生的虚实一致特性,提出了一种基于数字孪生的力能控制式压力机虚拟调试方法,设计了虚拟调试系统.根据力能控制式压力机的多系统协同调试任务,建立了其多物理领域统一模型,即机械、电气等的多领域数字孪生模型,并构建了虚拟控制系统,包括虚拟控制器和控制器-模型接口.通过应用于压力机虚拟调试的实例,展示了基于此数字孪生模型进行虚拟调试的有效性,构建的虚拟调试系统能够验证机械系统、电气系统和控制程序的设计,并确定和优化运行参数,有效避免了实机调试成本高、风险大、周期长等弊端.     

面向智能矿山与新工科的数字孪生技术研究

将数字孪生与人工智能(AI)技术相结合,提出了基于数字孪生+AI的智能矿山建设新思路。探索了智能矿山技术发展路径,研究了数字孪生技术的特征、应用领域及发展趋势,指出数字孪生是数字化矿山发展的必然趋势。提出了基于数字孪生+AI的智能矿山理论架构,构建了矿山数字孪生模型,模型自下而上分别为矿山全要素物理实体、矿山信息物理融合层、矿山数字孪生模型、矿山孪生数据交互层、矿山应用智能服务层,据此实现智能矿山的泛在感知、协同控制和智能决策与优化。从应用实际需求出发,探讨了智能矿山模型构建技术、智能开采数字孪生体技术、矿山智能控制技术、矿山设备故障预测、基于数字孪生的人机交互等关键技术。通过研究数字孪生在智能矿山中的应用,为AI技术在智能矿山应用落地提供思路,为未来智能矿山新工科建设提供理论借鉴。     

Control Strategies for Digital Twin Systems

With the continuous breakthrough in information technology and its integration into practical applications, industrial digital twins are expected to accelerate their development in the near future. This paper studies various control strategies for digital twin systems from the viewpoint of practical applications.To make full use of advantages of digital twins for control systems, an architecture of digital twin control systems, adaptive model tracking scheme, performance prediction scheme, perfo...     

Automated manufacturing system discovery and digital twin generation

The latest developments in industry involved the deployment of digital twins for both long and short term decision making, such as supply chain management, production planning and control. Modern production environments are frequently subject to disruptions and consequent modifications. As a result, the development of digital twins of manufacturing systems cannot rely solely on manual operations. Recent contributions proposed approaches to exploit data for the automated generation of the models. However, the resulting representations can be excessively accurate and may also describe activities that are not significant for estimating the system performance. Generating models with an appropriate level of detail can avoid useless efforts and long computation times, while allowing for easier understanding and re-usability. This paper proposes a method to automatically discover manufacturing systems and generate adequate digital twins. The relevant characteristics of a production system are automatically retrieved from data logs. The proposed method has been applied on two test cases and a real manufacturing line. The experimental results prove its effectiveness in generating digital models that can correctly estimate the system performance.     

Towards a sustainable interoperability in networked enterprise information systems: Trends of knowledge and model-driven technology

In a turbulent world, global competition and the uncertainty of markets have led organizations and technology to evolve exponentially, surpassing the most imaginary scenarios predicted at the beginning of the digital manufacturing era, in the 1980s. Business paradigms have changed from a standalone vision into complex and collaborative ecosystems where enterprises break down organizational barriers to improve synergies with others and become more competitive. In this context, paired with networking and enterprise integration, enterprise information systems (EIS) interoperability gained utmost importance, ensuring an increasing productivity and efficiency thanks to a promise of more automated information exchange in networked enterprises scenarios. However, EIS are also becoming more dynamic. Interfaces that are valid today are outdated tomorrow, thus static interoperability enablers and communication software services are no longer the solution for the future. This paper is focused on the challenge of sustaining networked EIS interoperability, and takes up input from solid research initiatives in the areas of knowledge management and model driven development, to propose and discuss several research strategies and technological trends towards next EIS generation.     

Modeling and implementation of a digital twin of material flows based on physics simulation

Cyber-physical production systems enable adaptivity and flexibility when manufacturing customized products in small batches. Due to varying routes and a high variance of workpieces, material flows in cyber-physical production systems can get highly complex, which can lead to physically induced disturbances that can result in accidents or decreased throughput and high costs. This issue can be addressed by applying a physics engine to simulate the physical interaction between workpieces and the material handling systems during the operation. Connecting such a digital model to a real material handling system in order to derive simulation-based decision support leads to the concept of digital twins. To date, few practical implementations of digital twins in manufacturing outside the machine tool domain were reported. Therefore, this paper describes the modeling and the subsequent implementation of an integrated system that consists of a real material handling system and its digital twin, based on physics simulation. A practical use case demonstrates the versatile advantages of the implemented solution for a manufacturing system with respect to the three digital twin functions prediction, monitoring and diagnosis.     

平行智能与CPSS: 三十年发展的回顾与展望

社会物理信息系统 (Cyber-physical-social systems, CPSS)在传统物理信息系统 (Cyber-physical systems, CPS)的基础上纳入对社会信号及社会关系的考虑, 利用网络世界近乎无限的人力、数据和信息资源, 突破物理世界有限的资源约束以及时空的限制. 然而, CPSS中人类和社会行为的复杂性加剧了实际系统和其模型之间的建模鸿沟, 使得系统的形态演变为“默顿系统”. 对此, 以ACP方法为核心的平行智能 (Parallel intelligence, PI) 框架通过组合人工系统 (Artificial systems, A)、计算实验 (Computational experiments, C)、平行执行 (Parallel execution, P)三个过程, 为跨越这一鸿沟提供了可行的路径. 具体而言, ACP将模型从系统解析器转变为数据生成器, 使原本难以控制的“默顿系统”可测试、可计算、可验证, 为复杂系统中“涌现”和“收敛”的对立统一确立了方法基础. 本文从平行控制与智能控制、平行机器人与平行制造、平行管理与智能交通、平行医学与智慧健康、平行生态与平行社会、平行经济系统与社会计算、平行军事系统以及平行认知与平行哲学这八个方面阐述面向CPSS的平行智能应用成果. 最后, 对CPSS未来的发展方向和技术趋势进行了讨论与展望.     

软件定义的系统与知识自动化:从牛顿到默顿的平行升华

知识自动化是智能化、人机化、自动化等的有机融合.从Cyberspace、社会信号和默顿系统等的兴起这一时代特征为出发点, 讨论知识自动化的意义与相关发展问题, 从智能算法到知识机器人讨论其技术基础及趋势, 从软件定义的系统和流程探讨其与系统工程的内在关联, 从平行系统的角度阐述其在复杂控制和管理过程中的重要作用.     

The metadatabase approach to integrating and managing manufacturing information systems

Integration has become a self-evident goal in today's manufacturing enterprises. Since the late 1970s, numerous efforts have been launched world-wide to develop, employ, and deploy integration technology as a strategic weapon to compete in the global market-place. Important results have been obtained, including computer integrated manufacturing (CIM), concurrent engineering, and enterprise re-engineering. The field is poised to tackle some fundamental barriers of integration and thereby effect a new quantum jump in overall productivity, where information will become a fourth basic factor of production along with land, labor and capital. Among the barriers are the lack of information technology enabling multiple systems operating concurrently over different geographical regions. This paper discusses a unique approach to the integration problem regarding information resources management, global query and concurrent systems administration. This approach is based on the metadatabase model which incorporates both data and knowledge in order to accomplish information integration across manufacturing functions. Basic concepts, methods and techniques have been developed for the approach and verified via a prototype metadatabase system. The system's primary components and major functionalities are discussed and illustrated in detail through the use of an extended example of a pilot manufacturing system comprising order processing, process design and shop floor control. It will also provide the reader with a simulated system demonstration.A preliminary version of this paper was presented to the Third International Conference on Computer Integrated Manufacturing, May 1992.