Intelligent scheduling of a feature-process-machine tool supernetwork based on digital twin workshop

Modern manufacturing enterprises are shifting toward multi-variety and small-batch production. By optimizing scheduling, both transit and waiting times within the production process can be shortened. This study integrates the advantages of a digital twin and supernetwork to develop an intelligent scheduling method for workshops to rapidly and efficiently generate process plans. By establishing the supernetwork model of a feature-process-machine tool in the digital twin workshop, the centralized and classified management of multiple data types can be realized. A feature similarity matrix is used to cluster similar attribute data in the feature layer subnetwork to realize rapid correspondence of multi-source association information among feature-process-machine tools. Through similarity calculations of decomposed features and the mapping relationships of the supernetwork, production scheduling schemes can be rapidly and efficiently formulated. A virtual workshop is also used to simulate and optimize the scheduling scheme to realize intelligent workshop scheduling. Finally, the efficiency of the proposed intelligent scheduling strategy is verified by using a case study of an aeroengine gear production workshop.     

数字车间生产调度原型系统及调度仿真

文章介绍所开发的一个基于Agent的数字车间的生产调度原型系统,包括它的组成、结构、平台选择和软件运行环境等方面的内容;并结合一个实际的生产任务,对车间调度系统对生产任务的招投标过程进行了仿真。     

应用Agent理论的生产调度系统研究

生产调度问题,一般可根据生产流程的不同分为Job-shop调度和Flowshop调度两大类(也有学者认为,存在两者相结合的第三类—混合调度)。该文研究以最小化Makespan为目标的Flowshop调度问题。基于Agent理论,提出采用Flowshop复合代理体(Flowshop-Compound-Agent,FSCA)求解Flowshop调度问题的方法。在给出FSCA的结构及其实现的基础上,通过毛纺企业制条车间的实例说明了使用FSCA解决Flowshop调度问题的有效性。     

网络管理中多agent的半在线调度算法

多agent调度算法在基于多agent的网络管理中对任务执行效率起着至关重要的作用.现有的多agent调度算法由于缺乏考虑任务间的依赖关系,使得面对复杂任务系统时会产生大量的网络负载和等待时间.为此,在建立一个适合网络管理任务特点的多agent调度框架的基础上,提出了一种基于任务依赖关系的多agent半在线调度算法.理论分析和测试结果表明,这种半在线调度算法优于已有的全在线调度算法,其性能更接近离线最优调度算法,从而为网络管理任务中多agent的动态调度提供了一种新的途径.     

动态二进制翻译中的指令调度技术研究与实现

动态二进制翻译提供了无需重新编译源代码就能将源机器生成的可执行代码自动转换到目标机器的方法,很好地解决了代码兼容性问题.其核心思想是根据程序的动态运行信息找到反复执行的代码序列,对代码序列进行翻译和优化,并将结果多次重用.指令调度作为一种有效的编译优化手段,也适用于动态二进制翻译.在对gcc的指令调度器分析研究的基础上,结合动态二进制翻译的实时性特点,提出了适合动态二进制翻译的效率高、开销小的指令调度算法.     

Digital Twin Enhanced Dynamic Job-Shop Scheduling

For dynamic scheduling, which is daily decision-making in a job-shop, machine availability prediction, disturbance detection and performance evaluation are always common bottlenecks. Previous research efforts on addressing the bottlenecks primarily emphasize on the analysis of data from the physical job-shop, but with little connection and convergence with its virtual models and simulated data. By introducing digital twin (DT), further convergence between physical and virtual spaces of the job-shop can be achieved, which greatly enables dynamic scheduling. DT fuses both real and simulated data to provide more information for the prediction of machine availability on one hand; and on the other hand, it helps to detect disturbances through comparing the physical machine with its continuously updated digital counterpart in real time, triggering timely rescheduling when needed. It also enables comprehensive performance evaluation for rescheduling using multiple-dimension models, which can describe geometric properties, physics parameters and behaviors of the machines. In the paper, a five-dimension DT for a machine in the job-shop is introduced first, then the DT-based machine availability prediction, disturbance detection and performance evaluation methods are explored. Based on this, a DT-enhanced dynamic scheduling methodology is proposed. A scheduling process of making hydraulic valves in a machining job-shop is taken as a case study to illustrate the effectiveness and advantages of the proposed method.     

基于PI3000平台的分布式任务调度研究与实现

在PI3000平台中任务调度是应用系统中很重要的部分,应用环境的复杂程度导致各式各样调度的框架和算法.文中提出一个新的分布式负载均衡的任务调度框架,以解决在并行任务中特定的应用环境下的任务调度问题.这个框架是动态的、可重用的,通过提供给各种不同的资源环境对应的接口,来进行跨服务的调度.整个框架关注于实际应用环境下任务、资源的动态不稳定性和任务计算的快速响应.     

Edge computing-based real-time scheduling for digital twin flexible job shop with variable time window

Production scheduling is the central link between enterprise production and operation management and is also the key to realising efficient, high-quality and sustainable production. However, in real-world manufacturing, the frequent occurrence of abnormal disturbance leads to the deviation of scheduling, which affects the accuracy and reliability of scheduling execution. The traditional dynamic scheduling methods (TDSMs) cannot solve this problem effectively. This paper presents a real-time digital twin flexible job shop scheduling (R-DTFJSS) method with edge computing to address the issue. Firstly, an overall framework of R-DTFJSS is proposed to realise real-time scheduling (RS) through real-time interaction between physical workshop (PW) and virtual workshop (VW). Secondly, the implementation process of R-DTFJSS is designed to realise real-time operation allocation. Then, to obtain the optimal RS result, an improved Hungarian algorithm (IHA) is adopted. Finally, a case simulation from an industrial case of a cooperative enterprise is described and analysed to verify the effectiveness of the proposed R-DTFJSS method. The results show that compared with the TDSMs, the R-DTFJSS method can effectively deal with unexpected and frequent abnormal disturbances in the production process.     

Framework of agent-based intelligence system with two-stage decision-making process for distributed dynamic scheduling

The advent of multiagent systems, a branch of distributed artificial intelligence, introduced a new approach to problem solving through agents interacting in the problem solving process. In this paper, a collaborative framework of a distributed agent-based intelligence system is addressed to control and resolve dynamic scheduling problem of distributed projects for practical purposes. If any delay event occurs, the self-interested activity agent, the major agent for the problem solving of dynamic scheduling in the framework, can automatically cooperate with other agents in real time to solve the problem through a two-stage decision-making process: the fuzzy decision-making process and the compensatory negotiation process. The first stage determines which behavior strategy will be taken by agents while delay event occurs, and prepares to next negotiation process; then the compensatory negotiations among agents are opened related with determination of compensations for respective decisions and strategies, to solve dynamic scheduling problem in the second stage. A prototype system is also developed and simulated with a case to validate the problem solving of distributed dynamic scheduling in the framework.     

面向数字孪生流域的云网融合架构研究

数字孪生流域建设是智慧水利建设的重要标志,是推动流域水利高质量发展的必然要求。为了应对当前数字孪生流域建设过程中面临的流域物联感知能力、通信传输网络资源、模型算法仿真计算水平等方面的不足而产生的挑战,本文结合云网融合的云网协同、按需互联、敏捷智能的服务能力特性,基于云网融合层次化的设计思想,把流域云和网充分融合,有效调度并应用计算、通信、存储等基础设施资源,提出了面向数字孪生流域的云网融合逻辑架构,梳理并阐述了水利感知网云网融合服务实现、水利信息网云网融合服务实现、数字孪生平台算力服务实现等可应用于数字孪生流域建设的云网融合服务架构,总结分析了面向数字孪生流域的云网融合关键技术。研究对充分发挥新一代信息技术对数字孪生流域建设的支撑驱动作用具有重要参考意义。     

A multi-agent architecture for dynamic scheduling of steel hot rolling

Steel production is a complex process and finding coherent and effective schedules for the wide variety of production steps, in a dynamic environment, is a challenging task. In this paper, we propose a multi-agent architecture for integrated dynamic scheduling of the hot strip mill (HSM) and the continuous caster. The scheduling systems of these processes have very different objectives and constraints, and operate in an environment where there is a substantial quantity of real-time information concerning production failures and customer requests. Each process is assigned to an agent which independently, seeks an optimal dynamic schedule at a local level taking into account local objectives, real-time information and information received from other agents. Each agent can react to real-time events in order to fix any problems that occur. We focus here, particularly, on the HSM agent which uses a tabu search heuristic to create good predictive–reactive schedules quickly. The other agents simulate the production of the coil orders and the real-time events, which occur during the scheduling process. When real-time events occur on the HSM, the HSM agent might decide whether to repair the current schedule or reschedule from scratch. To address this problem, a range of schedule repair and complete rescheduling strategies are investigated and their performance is assessed with respect to measures of utility, stability and robustness, using an experimental simulation framework.     

Design Alternatives of Multithreaded Architecture

This paper compares two possible implementations of multithreaded architecture and proposes a new architecture combining the flexibility of the first with the low hardware complexity of the second. We present performance and step-by-step complexity analysis of two design alternatives of multithreaded architecture: dynamic inter-thread resource scheduling and static resource allocation. We then introduce a new multithreaded architecture based on a new scheduling mechanism called the semi-static. We show that with two concurrent threads the dynamic scheduling processor achieves from 5 to 45 % higher performance at the cost of much more complicated design. This paper indicates that for a relatively high number of execution resources the complexity of the dynamic scheduling logic will inevitably require design compromises. Moreover, high chip-wide communication time and an incomplete bypassing network will limit the dynamic scheduling and reduce its performance advantage. On the other hand, static scheduling architecture achieves low resource utilization. The semi-static architecture utilizes compiler techniques to exploit patterns of program parallelism and introduces a new hardware mechanism, in order to achieve performance close to dynamic scheduling without significantly increasing the static hardware complexity. The semi-static architecture statically assigns part of the functional units but dynamically schedules the most performance-critical functional units on a medium-grain basis.     

基于强化学习的智能车间调度策略研究综述

智能制造是我国制造业发展的必然趋势,而智能车间调度是制造业升级和深化“两化融合”的关键技术。主要研究强化学习算法在车间调度问题中的应用,为后续的研究奠定基础。其中车间调度主要包括静态调度和动态调度;强化学习算法主要包括基于值函数和AC（Actor-Critic）网络。首先,从总体上阐述了强化学习方法在作业车间调度和流水车间调度这两大问题上的研究现状;其次,对车间调度问题的数学模型以及强化学习算法中最关键的马尔可夫模型建立规则进行分类讨论;最后,根据研究现状和当前工业数字化转型需求,对智能车间调度技术的未来研究方向进行了展望。     

A survey of dynamic scheduling in manufacturing systems

In most real-world environments, scheduling is an ongoing reactive process where the presence of a variety of unexpected disruptions is usually inevitable, and continually forces reconsideration and revision of pre-established schedules. Many of the approaches developed to solve the problem of static scheduling are often impractical in real-world environments, and the near-optimal schedules with respect to the estimated data may become obsolete when they are released to the shop floor. This paper outlines the limitations of the static approaches to scheduling in the presence of real-time information and presents a number of issues that have come up in recent years on dynamic scheduling. The paper defines the problem of dynamic scheduling and provides a review of the state-of-the-art of currently developing research on dynamic scheduling. The principles of several dynamic scheduling techniques, namely, heuristics, meta-heuristics, multi-agent systems, and other artificial intelligence techniques are described in detail, followed by a discussion and comparison of their potential.     

一种混合式拓扑的P2P VOD系统设计

针对P2P视频点播系统的扩展性问题,从媒体源服务器、索引服务器两方面考虑,提出一种混合式拓扑结构的P2PVOD系统的设计模型。在mesh结构系统普遍采用的gossip拓扑层之上构建一层有结构的P2P网络拓扑,使得DHT可以用于拖动操作后的合作节点定位,以有效减轻索引服务器的压力。在该混合式拓扑的基础上进行多服务节点数据调度算法的设计,能很好地适应网络的异构性与节点的动态性,并有效降低媒体源服务器的负载,同时满足系统扩展性与视频服务质量的需求。     

遗传算法在刹车片生产计划排产中的应用与实践

生产计划的排产问题是影响车间生产效率的主要因素之一.合理的排产计划能够高效地使用车间现有的生产资源,提高车间生产能力,降低生产成本.本文首先分析了刹车片生产运行状况,针对热压成型生产车间的多品种变批次生产模式,建立了排产数学模型;其次,设计了一种联合均值-遗传算法,对热压成型车间的排产数学模型进行求解;最后通过实验仿真,对本文设计的算法进行验证,实验结果表明联合均值-遗传算法能够很好的解决当前刹车片生产计划的排产问题.     

基于价值密度的Job-shop车间生产计划控制

针对离散型生产中工艺路线不确定,工时不准确,加工中变化快,物流速度快的生产现状,提出计划期内生产的价值作为追求目标,排产不必追求任务完成的时间节点,而应制定投产的顺序,以保证企业在单位时间内实现的价值最高,即价值密度作为排产依据。确定车间的瓶颈设备之后,通过排产保证该设备前面有足够的零部件等待加工,将人力资源和设备混合排产,在动态中实现任务计划的改变、时间窗口的锁定或拒绝,兼顾了订单的重要性和交货时间,发挥出了车间设备的最大能力。每个任务的计算量仅有一次,特别适合动态中的快速排产。     

基于网格计算池模型的作业调度策略研究

在现阶段,网格的计算池模型是一种较实用的网格实现方式,其资源的查找和分配机制也相对简单。文章根据计算池模型的特点,参照织女星网格的体系结构,将基于最短路由转发的资源发现机制与任务调度算法相结合,提出了一种新的网格任务调度策略,并给出了简化模型清晰地描述了任务在此策略下全部的执行过程。     

A digital twin-driven approach towards traceability and dynamic control for processing quality

Processing quality is the basis for ensuring product quality, and reflects the development needs and application value of realizing intelligent manufacturing. Aiming at the low efficiency of quality problems traceability, poor timeliness and unpredictability of quality control in the machining process, digital twin technology can provide a new intelligent solution based on interaction and integration between physical workshop and virtual workshop. Therefore, a digital twin-driven approach towards traceability and dynamic control for processing quality is proposed in the paper. Firstly, a Bayesian network model for the analysis of factors affecting processing quality (BN_PQ) is introduced, which determines the relevance and influence weight of each factor to processing quality. Secondly, in order to integrate multi-source heterogeneous data to trace the processing quality, a multi-level scalable information model and association mechanism are established. Moreover, the construction method of the IoT system in manufacturing unit for dynamic control of processing quality are introduced, in which the collection method of real-time data is discussed. The contents of digital twin data for processing quality constraints (DTD_PQ) and the management method are elaborated. Then, the digital twin-driven dynamic control method of processing quality is proposed. The conceptual model of the digital twin database and the operating logic for dynamic control of processing quality are described in detail. Finally, the interactive operation and core technologies of DTD_PQ towards traceability and dynamic control of processing quality are analyzed. By choosing examples of machining the connecting rod of diesel engine and the prototype system that has been developed, the effectiveness of the proposed method is verified.     

动态不确定环境下的决策：一种分层决策模型

本文提出一种智能体分层决策结构模型，试图通过分层决策技术有效地解决动态、不确定环境中的智能体的实时决策问题。本模型的高层采用BDI结构，以便为较长期任务的规划和推理提供充分的支持；模型的底层采用反应式结构，以保证对短期实时任务的及时响应。实验结果表明了这种分层模型在某些复杂任务领域中的有效性。