基于数字孪生与深度学习技术的制造加工设备智能化方法研究

针对传统制造加工设备在生产加工过程中存在设备与数据信息联系不紧密,设备使用维护多依赖于人工经验等问题,提出了一种新的设备智能化方法。首先,在信息层建立能反映制造加工设备真实状态的数字孪生体;其次,基于历史加工大数据,通过数字孪生体对加工过程的行为进行建模及深度学习和训练,并利用训练好的人工神经网络根据采集到的实时数据来预测制造加工设备下一时刻的状态,使制造加工设备实现物理层与信息层数据的深度融合,拥有自我感知、自我预测的能力,最终实现智能化;最后,以浆料微流挤出成型设备挤出结构系统的智能化实施过程为例,验证了所提出方法的可行性。实例结果表明该设备智能化方法可有效地对挤出结构系统的运行状态进行监测及预测,为后续提高挤出成型精度提供了有效的数据信息。研究表明数字孪生和深度学习技术能够提升制造加工设备的智能化程度,可为未来智能制造的发展提供理论支撑。     

Digital twin of composite assembly manufacturing process

Industry is facing the management of geometrical deviations along the entire lifecycle of the product. It is helped by digital twin tools that may minimise the geometrical deviations from nominal of products. The new digital twin tools allow to manage geometrical variations through a set of steps fully related by modern information and communication technologies that establish a continuous and unambiguous flow of information among the different steps of this digital process along the whole product lifecycle. They are based on data coming from manufacturing, assembly and inspection. The available large data sets from manufacturing and inspection allow to develop new and more accurate simulation models that realistically consider form deviations and process signature, i.e. the pattern left by the manufacturing process on the produced part surfaces. The present work introduces a digital twin tool to support the lightweight design of assemblies in composite material. It establishes a continuous and unambiguous flow of variation information from the part design to assembly, passing through manufacturing by considering the manufacturing signature. It was applied to a case study and the obtained results agree with the experimental ones.     

复杂型材智能加工制造系统的设计

目的 实现新一代信息技术背景下传统铝门窗幕墙型材加工行业的转型升级,以应对复杂型材加工制造存在的成本高、工序繁多等诸多挑战。方法 根据型材加工工艺流程及该行业定制化生产的特点,提出一种涵盖网上下单、订单自动处理、机床智能加工生产的复杂型材智能加工制造系统架构,重点针对自主开发的门窗幕墙型材一站式加工智能机床,研发出一套复杂型材智能加工制造系统。结果 研究了加工信息数字化模型、工艺数据库等关键技术。通过工艺数据库的构建,实现了自动编程系统的搭建。结合Web Service与XML技术,研发出订单自助处理系统、机床智能操作管理系统及其与ERP系统的集成互连,打通了生产各环节之间的技术壁垒,形成了复杂型材一体化加工工艺。结论 实际测试表明,经复杂型材智能加工制造系统一体化制造的复杂型材从接受订单到产品加工完成只需40分钟,大幅提高了生产效率和产品质量,降低了加工成本。本研究为复杂型材智能加工及其他传统制造行业的转型升级提供了有益的借鉴和参考。     

数字孪生驱动的物联型混单包装线多工位联动优化

目的 解决生产扰动给混单包装线带来的交货期延误、成本额外增加等问题.方法 将传统的混单包装线物联网化,实现生产资源的实时感知,及时获取生产现场的真实信息;面向物联型混单包装线的生产环境,基于数字孪生体系构建起物理世界与数字世界的联系,通过物联网的实体感知和网络的信息传输形成孪生数据集,建立孪生数据驱动的混单包装线多工位联动优化的信息架构;依据包装任务的工序约束,建立起多个工位的动态关联,实现工位间生产信息的实时协调,进而通过算法实现工位任务队列的联动式智能优化.结果 通过在某摩托车外胎包装生产线上的应用,优化后工序任务队列的目标函数值相比最高值下降了36.1％,验证了文中提出的物联型混单包装线多工位联动优化方法的有效性和可行性.结论 文中提出的混单包装线物联网化能为智能制造的实时提供硬件支撑,基于数字孪生的多工位联动优化提供了智能化决策工具支持.     

面向冷链智能立体仓库的数字孪生系统构建

目的 当前冷链物流行业立体仓库存在货品损耗率高、失温严重、信息化水平低等问题,确保冷链智能立体仓库的食品质量安全,并实现降本增效。方法 面向冷链智能立体仓库的运行场景,设计全映射与多维仿真的数字孪生五维模型,通过对多源异构数据的融合,实现立体仓库全要素的状态实时可视化与监测预警,并提出一种支持系统联动决策的模型优化策略,对立体库负载与温控区热力分布进行智能调节,最大限度提高运行效率和降低运营成本。结果 构建了一套冷链智能立体仓库数字孪生系统,通过对优化前后仓库运行数据进行对比,同等仓储环境条件下生鲜蔬菜的食品保鲜度损耗率同比下降了25%~30%,实现了冷链智能立体仓库实时数据与动态虚拟模型的精准映射与联动运转。结论 以广东某供应链管理有限公司佛山智能冷链仓储物流中心的冷链立体仓库为实践对象,形成了冷链智能立体仓库一体化管控新模式。为冷链智能立体仓库运营管理、调度联动决策与协同作业提供了数字孪生技术的应用参考,为建立更完善的保鲜产品包装与冷链食品质量安全提供了新路径。     

基于数字孪生的单兵作战系统设计研究

目的 当前国际形势日益复杂严峻,大国之间军事竞争不断加剧,各国加速进行国防装备研发.信息技术飞速发展催生未来战场的信息化、智能化转变,亟需设计研发新型单兵作战系统以应对未来战场环境.为有效地提升单兵作战系统设计效率、解决传统单兵作战系统设计过程的问题,提出数字孪生驱动的单兵作战系统设计方法.方法 基于文献研究和案例研究,总结归纳单兵作战系统设计过程和存在问题,提出数字孪生单兵作战系统设计模型,分析数字孪生单兵作战系统设计模型的组成要素,研究并呈现了几种数字孪生单兵作战系统设计关键技术.结论 随着数字孪生技术日益成熟并得以广泛应用,将数字孪生技术应用于单兵作战系统设计过程具有可行性.数字孪生驱动的单兵作战系统设计过程丰富了国防装备设计研发的理论及方法,为新型单兵作战系统创新设计提供了新的路径.     

面向智能制造的数字仓储系统设计

目的 从设计的角度探究如何运用数字仓储系统满足智能制造发展需求的相关理论,以期为制造企业提供智能化生产服务.方法 运用文献研究和典型案例分析法,通过系统架构、管理模块、数据流等方面对数字仓储系统进行全面设计,揭示出数字化仓储技术与装备的功能特点,进而提出数字仓储系统实现路径.结果 数字仓储系统实现了对物料、成品从订单发出到在途、上架、入库、出库全流程的追踪监控,对仓储、物流资源进行了一体化管理,实现了智能制造信息深度自感知、智慧优化自决策、精准控制自执行.结论 在智能制造的大环境下,数字仓储系统能够满足制造企业大规模定制生产要求,具备高柔性的全流程数字化管理模式,助推智能制造进一步发展.     

智能交互时代设计赋能智能制造创新发展路径研究

目的 总结智能交互时代设计赋能智能制造创新发展的策略和路径。方法 基于智能交互时代背景,以创新设计思维为指导,从智能制造的发展现状、动态趋势以及与创新设计的关联性等方面对智能制造的创新设计发展策略及路径进行探讨。结果 在创新设计思维的指导下,总结智能制造的创新设计模型与发展策略,以及设计赋能智能制造创新发展的可行性路径。结论 以智能制造创新设计思维为指导,总结了创新设计流程、数字化转型、柔性设计制造、协同集成平台、创新服务系统以及设计教育新范式等六大设计赋能智能制造创新发展的可行性路径。     

Convolutional and generative adversarial neural networks in manufacturing

Manufacturing is undergoing transformation driven by the developments in process technology, information technology, and data science. A future manufacturing enterprise will be highly digital. This will create opportunities for machine learning algorithms to generate predictive models across the enterprise in the spirit of the digital twin concept. Convolutional and generative adversarial neural networks have received some attention of the manufacturing research community. Representative research and applications of the two machine learning concepts in manufacturing are presented. Advantages and limitations of each neural network are discussed. The paper might be helpful in identifying research gaps, inspire machine learning research in new manufacturing domains, contribute to the development of successful neural network architectures, and getting deeper insights into the manufacturing data.     

Reconfiguration: a key to handle exceptions and performance deteriorations in manufacturing operations

During a manufacturing operation, exceptions may occur dynamically and unpredictably. Their occurrence may lead to the degradation of system performance or, in the worst case scenario may interrupt the production process by causing errors in the schedule plan. This paper classifies three families of exceptions: (1) out-of-order events such as machine breakdowns, (2) operational out-of-ordinary events such as rush orders and (3) deteriorations of manufacturing resource performance such as reductions of machines’ utilization. In all cases, in order to maintain an adequate level of system performance, it is necessary to detect exceptions, to diagnose them quickly and to recover them by taking corrective actions to avoid fault propagations. Decisions concerning how to deal with exceptions, i.e. which strategy to implement, depend on the manufacturing environment (dedicated line, flexible system, reconfigurable system or a mix of them) and the advantages arising from using a certain exception handling policy vary from one production system to another. The activity of reconfiguring manufacturing resources has been demonstrated to be a powerful operation strategy to handle machine breakdowns. This paper extends the concept of ‘reconfiguration for exception handling’ to other families of exceptions and proposes reconfiguration for their recovery. The reconfiguration process is handled by an agent-based control system that implements four negotiation processes among manufacturing resource agents.     

Application of blockchain in the field of intelligent manufacturing: Theoretical basis,realistic plights,and development suggestions

Blockchain technology is considered one of the promising technologies of the information technology era. The core features of blockchain, such as decentralization, transparency, high security, and tamper-proof nature, bring great convenience for large-scale social cooperation and data sharing. Blockchain has a broad application prospect in the field of intelligent manufacturing. The key issues of this field, such as distributed collaborative production, industrial big data sharing and security, transparent logistics, and supply chain, are naturally consistent with the core characteristics of the blockchain technology. This study aims to analyze the application of blockchain in the field of intelligent manufacturing. First, we introduce the basic connotation and applications of blockchain. Then, we propose the theoretical basis for the application of blockchain in the field of intelligent manufacturing. Finally, we point out the realistic plights and provide some suggestions to promote the application of blockchain in the field of intelligent manufacturing.     

基于五轴加工中心智能生产线的数字孪生应用研究

智能制造是我国近年来大力倡导的重点发展方向之一。随着智能制造试点的不断建立、发展和完善,数字孪生技术逐步得到应用。基于此,以五轴加工中心智能生产线为例,对数字孪生技术的应用情况进行总结。基于数字孪生技术,按1∶1的比例构建五轴加工中心智能生产线上实体设备的数字模型并配置对应的物理属性,使得数字模型具有与实体设备相同的动作状态。然后,对数字模型进行信号点设置并与外部控制信号进行映射与连接,实现对建成前五轴加工中心智能生产线的虚拟调试,并通过不断迭代获得最优的建设方案。实践结果表明,利用数字孪生可实现智能生产线组合、调试和生产的虚拟仿真;基于数字孪生的虚拟仿真功能,可实现产品物理属性参数和设计精准度的可视化展示,使得产品在设计阶段就能通过适应性验证,实现快速更新;在智能生产线建成后,利用数字孪生可实现对生产线上实体设备的管理及其工作状态的实时监控,以更好地了解各设备的运行状态。综上,数字孪生技术贯穿了智能生产线的全生命周期,可在智能制造领域推广应用。     

Empowering production workers with digitally facilitated knowledge processes – a conceptual framework

Recent digital advancements, including social software, mobile technologies and augmented reality, offer promising opportunities to empower knowledge workers in their production environment by leveraging their knowledge processes, decision-making skills and social interaction practices. This paper proposes a conceptual framework for empowering workers in industrial production environments with digitally facilitated knowledge management processes. The framework explores four concrete facets of digital advancements that apply to a wide range of knowledge processes and production strategies in manufacturing companies. Each of these advancements are capable of supporting one specific facet of the individual knowledge management processes of workers; knowledge transfer, discovery, acquisition and sharing. The study contributes to the production research community by aligning emerging digital technologies and current trends in advanced manufacturing environments to benefit workers and improve job satisfaction, efficiency and productivity. The paper also contains suggestions about developing innovative solutions for production environments that support workers with digital technologies for flexible production.     

Integrated and Intelligent Manufacturing: Perspectives and Enablers

With ever-increasing market competition and advances in technology, more and more countries are prioritizing advanced manufacturing technology as their top priority for economic growth. Germany announced the Industry 4.0 strategy in 2013. The US government launched the Advanced Manufacturing Partnership (AMP) in 2011 and the National Network for Manufacturing Innovation (NNMI) in 2014. Most recently, the Manufacturing USA initiative was officially rolled out to further “leverage existing resources... to nurture manufacturing innovation and accelerate commercialization” by fostering close collaboration between industry, academia, and government partners. In 2015, the Chinese government officially published a 10-year plan and roadmap toward manufacturing: Made in China 2025. In all these national initiatives, the core technology development and implementation is in the area of advanced manufacturing systems. A new manufacturing paradigm is emerging, which can be characterized by two unique features: integrated manufacturing and intelligent manufacturing. This trend is in line with the progress of industrial revolutions, in which higher efficiency in production systems is being continuously pursued. To this end, 10 major technologies can be identified for the new manufacturing paradigm. This paper describes the rationales and needs for integrated and intelligent manufacturing (i²M) systems. Related technologies from different fields are also described. In particular, key technological enablers, such as the Internet of Things and Services (IoTS), cyber-physical systems (CPSs), and cloud computing are discussed. Challenges are addressed with applications that are based on commercially available platforms such as General Electric (GE)’s Predix and PTC’s ThingWorx.     

Bi-objective dynamic control of batch processor with non-identical jobs in mould manufacturing

The heat-treatment operation in dynamic mould manufacturing often involves non-identical jobs, which allow for simultaneous processing yet with different weights and due dates. Effective production control of this operation is essential to improve the on-time delivery and decrease the manufacturing cost of the mould. This paper considers the dynamic control of a batch processor for dealing with such non-identical jobs. A new look-ahead batching strategy called MLAB (mould: look-ahead batching) has been proposed. In MLAB, the control decisions are made by the joint use of both near-future arrival information of upstream operations and workload level information of downstream operations. MLAB strategy is used to control two kinds of conflicting objectives related to the delivery and utilisation performances and finally achieve trade-off based on compromise programming method. Computational experiments are conducted to verify the effectiveness of the MLAB strategy and show that the results are promising as compared to benchmark control strategies.     

Integrated predictive maintenance strategy for manufacturing systems by combining quality control and mission reliability analysis

Predictive maintenance (PdM) is an effective means to eliminate potential failures, ensure stable equipment operation and improve the mission reliability of manufacturing systems and the quality of products, which is the premise of intelligent manufacturing. Therefore, an integrated PdM strategy considering product quality level and mission reliability state is proposed regarding the intelligent manufacturing philosophy of ‘prediction and manufacturing’. First, the key process variables are identified and integrated into the evaluation of the equipment degradation state. Second, the quality deviation index is defined to describe the quality of the product quantitatively according to the co-effect of manufacturing system component reliability and product quality in the quality–reliability chain. Third, to achieve changeable production task demands, mission reliability is defined to characterise the equipment production states comprehensively. The optimal integrated PdM strategy, which combines quality control and mission reliability analysis, is obtained by minimising the total cost. Finally, a case study on decision-making with the integrated PdM strategy for a cylinder head manufacturing system is presented to validate the effectiveness of the proposed method. The final results shows that proposed method achieves approximately 26.02 and 20.54% cost improvement over periodic preventive maintenance and conventional condition-based maintenance respectively.     

Modelling reconfigurable manufacturing systems with coloured timed Petri nets

Reconfigurable manufacturing systems (RMSs) have been acknowledged as a promising means of providing manufacturing companies with the required production capacities and capabilities. This is accomplished through reconfiguring system elements over time for a diverse set of individualised products often required in small quantities and with short delivery lead times. Recognising the importance of dynamic modelling and visualisation in decision-making support in RMSs and the limitations of current research, we propose in this paper to model RMSs with Petri net (PN) techniques with focus on the process of reconfiguring system elements while considering constraints and system performance. In view of the modelling challenges, including variety handling, production variation accommodation, machine selection, and constraint satisfaction, we develop a new formalism of coloured timed PNs. In conjunction with coloured tokens and timing in coloured and timed PNs, we also define a reconfiguration mechanism to meet modelling challenges. An application case from an electronics company producing mobile phone vibration motors is presented. Also reported are system analysis and application results, which show how the proposed formalism can be used in the reconfiguration decision making process.     

面向智能发电的数字孪生技术及其应用模式

发电过程作为电力系统的重要组成部分,其智能化建设是推进我国能源电力转型发展的重要方向。结合数字孪生技术的研究发展现状及发展趋势,剖析了智能发电的概念与体系,并给出面向智能发电的数字孪生定义,通过功能映射,构建了智能发电系统的数字孪生架构。提出了在智能发电系统中部署数字孪生技术的一般性方法,对数字孪生技术的应用领域和应用模式进行了归纳,分析了应用过程中面临的主要技术挑战。研究结果可为数字孪生技术在发电领域产业化发展提供理论和方法参考。     

模具数字化设计与快速制造

目的为缩短模具产品开发周期,降低制造成本。方法构建了模具的数字化设计与快速制造系统,从造型、结构分析、虚拟设计、成形工艺仿真与优化等方面,阐述了数字化设计的过程,并介绍了RP和RT制造的方法。最后以玩具车齿轮为例,阐述了模具数字化设计与快速制造的应用。结论模具的数字化设计与快速制造,为新产品开发提供了一种快速而又经济的方法。