基于多智能体的制造系统生产控制建模研究

为提高制造系统生产控制的性能,建立了基于多智能体系统的混合控制模型.该模型把生产控制系统分为管理智能体层、单元智能体层和执行智能体层.管理智能体层负责调度和协调各单元智能体,并对所有智能体进行管理;单元智能体层中的各单元智能体间通过公用数据库相互协作;执行智能体对制造系统内的硬件负责,它们根据局部的本地资源信息及当前状态,接收发布的任务,并对其求解.同一层次的智能体之间是分布式结构.采用基于多智能体的混合控制模式,提高了制造系统生产控制的实时性和灵活性.通过激光拼焊生产系统中的试验,验证了该模型的有效性.     

基于多智能体的网络协同产品数据管理技术研究

为解决广义企业的协同工作环境构建知识集成平台的问题，提出了基于多智能体的网络协同产品数据管理技术的产品数据管理解决方案。该方案运用多智能体技术的原理和方法，建立了基于多智能体的网络协同产品数据管理的体系结构和网络产品全信息模型，研究了基于多智能体的网络协同产品数据管理系统实现的关键问题，并采用多智能体理论解决产品数据管理技术中数据对象复杂化、异构化、网络化和协同化等复杂问题，最后开发了一个基于多智能体的网络协同产品数据管理原型系统，并验证了该解决方案的可行性。     

并行工程中的冲突协商模型研究

产品并行开发过程中地理上分布的各智能体进行协同设计时不可避免地会产生冲突。为了获得一个各方一致接受的全局最优设计 ,各智能体必须通过协商来解决冲突 ,使冲突最终达成一致。本文提出并实现了一种并行环境下的多智能体模型 ,该模型中的各智能体采用多层混合推理策略进行冲突协商。该协商解决方案可以弥补单一解决方案存在的不足 ,充分发挥基于实例、基于规则和基于模型的推理方法解决冲突的长处     

基于多智能体的连铸-热轧一体化生产调度模型

连铸.热轧生产制造流程是钢铁生产的一个重要环节,该过程的生产计划与调度问题是影响钢铁工业生产效率和经济效益的重要因素.本文针对连铸.热轧过程的生产调度问题,建立了整个连铸.热轧一体化生产调度过程多智能体系统模型,该模型由6个智能体组成.讨论了多智能体模型各智能体的机能、任务以及各智能体之间的协同工作问题.最后,利用实际生产数据进行了仿真实验,表明所提出的模型能够有效提高连铸-热轧一体化生产调度的热装比,减少能源消耗.     

分布式CAPP中协作智能体的构造

基于智能体的分布式CAPP中 ,各功能智能体共同协调工作是解决网络化制造环境下CAPP自动化的关键。针对焊接结构件CAPP系统的开发 ,提出协作智能体的构造方法 ,建立其工作过程模型 ,实现了CAPP中多用户、多任务的并行协作     

基于水池故障模拟试验的传感器系统设计

基于智能体系统进行水池故障模拟试验,提出传感器系统设计方法,详细说明了适应于智能体水池试验的测速设备及方法,针对试验环境,建立坐标系,通过坐标变换,根据不同象限轨迹角计算公式,得到智能体在固定坐标系的各坐标轴上的速度分量,为进行智能体运动控制研究及状态监测研究奠定了基础.智能体系统水池故障模拟试验结果验证了文中所设计的传感器系统的有效性.     

企业信息化建设的多智能体编队模型研究

企业信息化可以看成一个大系统,包括输入、输出、反馈、内部运行、行政措施和激励措施等。企业内部可以看作是由多个智能体(Agent)组成的相互协作的人工社会组织单元。从多智能体编队的角度研究了企业信息化模型,以及智能体和企业元素之间的联系,引入了势场函数理论来描述企业元素之间的关系,即引力代表各单元意见一致,斥力则代表各单元有互斥意见,通过确定适当的势场函数可以满足各单元之间的功能和利益需求。     

多智能体模型在多机器人网络遥操作中的应用

通过Internet网络操作远端的机器人可以扩展人类的控制能力,但系统的复杂性和不确定性使如何更好地表现其良好性能成为目前的热点问题。通过对多机器人网络遥操作系统控制模式分析,阐明了以智能体技术为支撑,实现操作者智能与机器人局部自主能力的结合,完成多操作者多机器人Internet网络遥操作任务。在对多智能体技术概述基础上,重点分析智能体之间的联结特点,建立具有公共属性的智能体模块化层次结构,给出各模块的功能描述。在此基础上,实现基于多层分布式黑板模型和智能体节点的多机器人网络遥操作系统体系结构。最后设计一组辅助医学操作试验,通过测试实时避碰智能体的工作状态和效率验证了多智能体结构框架下多机器人网络遥操作的可行性和有效性。     

起重机系列产品装配图明细表自动更新技术研究

为了使起重机系列产品三维变型设计中,各模块明细表属性能够随着产品变更后的信息一致,属性链接零件配置信息,开发了基于三维明细表自动更新的智能模块体程序,通过遍历装配体,提取零部件数量、名称等知识,利用知识推理技术,自动更新明细表行数、行距、列宽等属性信息,同时对于大型复杂产品,明细表超出图框以外能够自动折断换行,实现了三维转向二维明细表信息的无缝对接。该模块通用性、适用性强,解决了三维设计中明细表信息修改繁琐、易错的难题。     

基于Agent的芯片制造生产线动态调度方法研究

基于智能体技术,提出了芯片制造生产线动态调度新方法,实现了投料调度、工件调度与设备维护调度的集成。首先,给出了基于智能体的动态调度模型,该模型中包括管理智能体、投料智能体、工件智能体、设备智能体、运输智能体与人员智能体。投料智能体用于实现投料控制,工件调度通过工件智能体与设备智能体之间的协商实现,设备维护调度由设备智能体实现。通过智能体间的合作,能够实现投料控制、工件调度与设备维护调度的协同进行。然后给出了工件智能体与设备智能体问协商使用的协商协议（即带有时间约束的单步协商协议）,以及智能体决策中使用的调度算法（包括投料调度算法,工件调度算法与设备维护调度算法）。最后,通过实例,给出了方法的使用过程,并进行了模拟仿真。仿真结果表明,集成的调度方法能更好地优化模型的生产率、加工周期与设备利用率,最终提高准时交货率。     

面向成本的设计中的成本构成及其估算框架的研究

认为面向成本的设计（DFC）是指在满足用户不前提下,对尽可能降低成本的的产品设计提供技术支持。在分析DFC中成本构成及其估算方法的基础上,提出了综合运用多种估算方法并与成本管理工作相结合的DFC的成本估算策略。运用人工智能技术建立DFC的成本估算框架。框架采用多级多阶段结构并综合运用多种成本估算方法对产品全生命周期成本进行估算。     

A Multi-Agent-Based Agile Shop Floor Control System

The ability of an enterprise to deliver new products quickly and efficiently to market is often the chief determinant of competitive success. The shop floor control system must be an open dynamic system with the capability of adapting and accepting radical unpredictable changes in its structures and industrial practices. This paper presents a new architecture for an agile shop floor control system. The architecture is based on the methodology of multi-agent systems in distributed artificial intelligence (DAI). The multi-agent system has some common characteristics such as: distribution, autonomy, interaction, and openness, which are helpful for transferring traditional architecture to a distributed, cooperative architecture for a shop floor control system. A bidding method based on the required production cost and processing time is also proposed. Using a distributed object-oriented technique, a CORBA-based multi-agent framework for an agile shop floor control system is constructed to integrate all the activity of the shop floor into a distributed intelligent open environment. To implement the framework, a coordination model between agents and behavioural models of some representative agents are established.     

用于机加工产品设计过程降低成本求解的知识表示方法

为解决由于评价对象的不确定性造成的无法有效应用降低成本知识的问题,针对机加工类产品降低成本知识进行分类,指出知识之间的相互嵌套关系.分析降低成本知识的特点,提出一种适用于降低成本求解过程的集成成本知识元-规则-框架(CKE-R-F)的混合知识表示方法,用产生式规则作为知识表示主体,内嵌框架定义和成本知识元用以表达知识内容之间的相互关系.建立降低成本知识库,实现知识的交互获取与知识库的维护功能,并将其应用于设计过程的降低成本求解过程中.     

An ontology-based method for knowledge integration in a collaborative design environment

A collaborative design environment can be viewed as a multi-agent system where each design agent has knowledge about specific domains and can solve different problems. Several agents combine to solve a complex problem through knowledge sharing and inter operation. It is essential to construct an integrated knowledge base to improve the efficiency and consistency of complex problem-solving. An ontology-based knowledge integration framework in a collaborative design environment is presented in this paper. Ontology provides standard vocabulary, technical terminology and a domain model for knowledge integration. The representation and construction of ontology are critic problems for knowledge integration. An object-oriented concept graph representation is presented to represent ontology. It has advantages of both object-oriented methodology and Sowa’s conceptual graph. It can be translated into predicate calculus conveniently, and can represent the intension of concepts. In order to improve the efficiency of ontology construction, a semantic-based approach is presented. Firstly, a semantic affinity based clustering algorithm is presented to generate concepts from the term set of multi-agent systems. Then, ontology is constructed based on knowledge.     

EDM COLLABORATIVE MANUFACTURING SYSTEM BASED ON MULTI-AGENT TECHNIQUES

A framework for building EDM collaborative manufacturing system using multi-agenttechnology to support organizations characterized by physically distributed, enterprise-wide, heteroge-neous intelligent manufacturing system over Internet is proposed. Expert system theory is introduced.Design, manufacturing and technological knowledge are shared using artificial inielligence and webtechniques by EDM-CADagent, EDM-CAMagent and EDM-CAPPagent. System structure, design     

RESEARCH ON CAPP／SCHEDULING INTEGRATION MULTI-AGENT SYSTEM MODEL AND IMPLEMENTATION

A design methodology fOr multi-agent systems is proposed. The systemic framework ofCAPP and scheduling integrated multi-agent system according to design methodology is researched.Agent model, composition model and cooperation model are discussed respectively in the multi-agentsystem.Static composition model and dynamic running model of CAPP and scheduling integrated     

基于多智能体的现代制造系统建模技术

多智能体技术在制造领域的应用是一个新兴的研究热点.文章对智能体的特性进行了全面阐述,提出了一种新的、动态的基于多智能体的制造系统建模方法.基于多智能体的制造系统模型是一个动态的、分布式智能体网络模型,可以对环境变化做出快速响应,动态生成最优的产品制造方案,实现系统模型重构和资源重新配置.     

基于MAS的虚拟企业框架结构设计

虚拟企业具有动态性、分布性、自治性、异构性等特点，传统的企业信息系统已经不能满足虚拟企业的这些特殊要求。在分析了多代理技术在虚拟企业中应用的必要性和可行性基础上，将多代理技术引入到虚拟企业信息系统中，构造了一个基于MAS技术的虚拟企业框架结构，并对于各功能Agent，给出了一种通用的Agent结构。     

Bidding-based multi-agent system for integrated process planning and scheduling: a data-mining and hybrid tabu-SA algorithm-oriented approach

This paper conceptualizes a bidding-based multi-agent system for solving integrated process-planning and scheduling problem. The proposed architecture consists of various autonomous agents capable of communicating (bidding) with each other and making decisions based on their knowledge. Moreover, in contrast to the traditional model of integrated process-planning and scheduling problem, a new paradigm has been conceptualized by considering tool cost as a dynamic quantity rather than a constant. Tool cost is assumed to comprise tool-using cost and its repairing cost. The repairing cost is considered to depend on the tool-breaking probability, which is predicted by the data-mining agent equipped with the virtues of C-fuzzy decision tree. When a job arrives at the shop floor, the component agent announces a bid for one feature at a time to all the machine agents. Among the machine agents capable of producing the first feature, one comes forward to become a “leader”, and groups other machine agents for the processing of remaining features of the job. Once all features are assigned to the appropriate machines, the leader then sends this allocation information to the optimization agent. The optimization agent finds optimal/near-optimal process plans and schedules via the hybrid tabu-SA algorithm.     

Decision fusion system for fault diagnosis of elevator traction machine

Fault detection and diagnosis is critical for healthy operation of an elevator system. In order to realize a real-time and convenient diagnosis and satisfy the requirement of advanced maintenance of an elevator system, this paper proposes an intelligent fault diagnosis approach of induction motor for elevator traction machine using a developed decision fusion system. First, the basic knowledge of fusion techniques is briefly introduced which consists of classifier selection and decision fusion. Then a developed decision fusion system is presented. Next, four fusion algorithms—majority voting, Bayesian belief, multi-agent and modified Borda count—are employed for comparison in a real-world diagnosis experiment of a faulty elevator motor system. Based on the satisfactory results shown in the experiment, a big potential in real-world application is presented that is effective and cost saving only by analyzing stator current signals using proposed decision fusion system.