节点文献

面向中小企业的云制造平台关键技术研究

Research on Key Technologies of Cloud Manufacturing Platform Facing to Small and Medium-sized Enterprise

【作者】 杨男

【导师】 李东波;

【作者基本信息】 南京理工大学 , 机械制造及自动化, 2013, 博士

【摘要】 网络化制造的快速发展以及越来越激烈的市场竞争,为中小企业的生存带来了严峻的考验。不同于集团企业,中小企业的研发能力较弱,制造能力受限较多,如何帮助中小企业在目前的市场环境下获得更大的发展空间,云制造这一先进的制造模式为我们带来了新的研究思路。本文通过介绍面向中小企业的云制造模式及其相关关键技术的研究背景和意义,指出了云制造模式能帮助中小企业更快地实现网络化制造从而提高企业的竞争力;通过分析目前国内外针对云制造和传统网络化制造的研究现状,总结了目前关于中小企业云制造平台应用研究中尚存在的不足和局限性,并据此确定了本文的主要研究内容如下:(1)分析了面向中小企业的云制造模式的应用特点,提出了基于动态联盟的云制造应用模式,给出了该模式下参与云制造的三个角色各自的职责和提供/获得/管理制造服务的方法和步骤。搭建了基于动态联盟和资源共享的五层云制造平台功能结构,针对每层进行了功能介绍和模块分析。建立了面向服务的基于ESB的集成架构并对该架构下实现功能的工具进行了分析和选取,在此基础上建立了实现面向中小企业的云制造平台的关键技术体系。(2)分析了云制造环境下云制造服务的属性和特点,对制造资源进行了分类和形式化地描述,提出了一种基于动态描述逻辑(Dynamic Description Logic, DDL)的云制造服务描述方法并且建立了一种基于动态描述逻辑的匹配过程原子动作理论——将云制造服务匹配过程视为一组存在逻辑关系的原子动作集合,在此基础上建立了一种基于DDL的云制造服务匹配策略,给出了该策略下的匹配判定规则和算法。(3)分析了云制造环境下制造服务的优选需求和关键问题,建立了以组合服务QoS为依据的制造服务优选方案和“基本QoS模型→关联QoS模型→静态组合QoS模型→动态组合QoS模型→优选模型→算法求解”的研究思路。在该研究思路指导下,利用统计逻辑关系这一新概念来描述制造服务之间存在的关联关系,建立了基于统计逻辑关系的关联QoS模型;对QoS的更新机制进行了研究,提出了基于企业自主更新和历史数据统计加权求和的更新方法,引入了更新参数τ用以控制历史数据对实际数据的影响程度,并最终建立了动态QoS模型。最后,对基于动态组合QoS的制造服务优选问题进行了形式化描述和数学建模,并采用粒子群算法对优选问题进行了求解和验证。(4)基于云制造平台中可能出现多任务并行制造的情况,建立了以案例为导向的云制造服务配置思路,通过分析案例导向的云制造服务配置特点提出了基于制造能力负载均衡的云制造服务配置方法。该方法的核心思想可以概括为:以多案例协同进行为前提、以制造案例为导向、以制造云服务使用状态监控为主线,以制造能力为衡量手段,以相同制造能力负载均衡为目标,最终实现拥有同质制造能力的云制造服务提供商能够最大程度上获利均衡。作为对配置过程的支持,形式化地描述了案例执行状态监控过程,并提出了一种新的描述方法——调整域△,利用△确定制造能力冲突位置和具体表现。在进行制造能力冲突消解的过程中,考虑到云制造服务独占性和地理分布性的特点提出了制造地域隶属度的概念,建立了基于制造地域隶属度的制造能力消解方法,给出了该方法实现的步骤。最后给出了基于制造能力负载均衡的云制造服务配置方法的数学模型,建立了问题求解的若干假设、目标函数和约束条件,并用实例验证了本章方法的正确性和可行性。(5)通过对云制造环境下制造服务交易过程的分析建立了基于多属性协商的云制造服务交易方法,多属性协商考虑了云制造服务交易过程中表现出的动态性、高分布性、复杂性和自治性,并且协商交易符合中小企业实际交易需求和交易中遵循的低价高利润的原则。建立基于多属性协商的交易模型时首先针对多属性协商进行研究,建立了衡量不同属性表现的效用函数U(x),利用U(x)建立了基于同步让步的云服务协商策略,包括协商提议的确定与驳回,可接受区域、可协商区域和驳回区域的确定,基于时间的让步策略模型的建立。在此基础上给出了基于多属性协商的云制造服务交易方法的操作步骤。(6)根据论文的研究成果搭建了面向锻造行业的云制造服务平台原型系统,介绍了平台所使用的集成开发环境和工具,向读者介绍了原型系统各模块的原理和实现功能。

【Abstract】 Unlike group enterprises, small and medium-sized enterprises’Research&Development capabilities are weaker, and manufacturing capacity is limited by many elements. How to help SMEs get more development spaces in the current market environment now is a crucial challenge and cloud manufacturing, a new advanced manufacturing mode, brings us a new research idea. This paper introduces the research background and significance of the cloud manufacturing model and its key technologies for SMEs, pointing out that the cloud manufacturing model can help SMEs to achieve the networked manufacturing faster, enhance the competitiveness of SMEs. After analyzing the current research situations of cloud manufacturing and networked manufacturing at home and abroad, the deficiencies and limitations of the current researches under the cloud manufacturing environment for SMEs are summed up.The main research contents in this paper are as follows:Firstly, according to the characteristics of cloud manufacturing model faced to the SMEs, the application mode based on dynamic alliance under the cloud manufacturing environment is proposed. In this mode, there are three typical characters named CMS/CMM/CMG, and their respective responsibilities and functions are introduced. And then a five-layer cloud manufacturing architecture based on dynamic alliance emphasizing resources sharing is built up, and features and modules of every level are introduced in details. In order to solve the difficulties of System Integration, a service-oriented mode based on the ESB is put forward. Established on the basis, key technologies for SMEs’cloud manufacturing platform are discussed.Secondly, the attributes and characteristics of manufacturing services under the Cloud manufacturing environment are studied. The classification and description of cloud manufacturing resources are researched; a formal description method and a new motor theory based on the Dynamic Description Logic are proposed. According to this theory, the matching process is considered as a set of logical atomic actions. On this basis, a strategy of manufacturing service matching based on DDL is proposed, and the rules and algorithms are discussed as well.Thirdly, after analyzing the needs and key issues of the combination selection questions of the cloud manufacturing services, establish the research approach of "the basic QoS model→associated QoS model→static combination QoS model→dynamic combination QoS model→selection model→algorithm".A new concept called "Statistics logical relationship" is proposed to describe the relationship between manufacturing services and a associated QoS model based on Statistics logical relationship is built up. At the same time, QoS update mechanism is researched and a update method considering both enterprise independent update and historical statistics with the update parameter τ is proposed, dynamic combination QoS model is built up as well. Then, the formal description and mathematical modeling of the combination selection model is built up and solved.Fourthly, multiple manufacturing tasks may exist at the same time under the cloud manufacturing environment, in order to solve the assignment of cloud manufacturing service, a case-directed method based on manufacturing capacity using balance is proposed. A new description method called "Adjust domain-△" is built up for the manufacturing capacity conflict in the process of collaborative manufacturing, and a concept named "manufacturing area membership" is proposed in order to describe the exclusive and geographical distribution characteristic of cloud manufacturing services. Then the mathematical modeling of assignment problem based on manufacturing capacity using balance is built up and solved.Fifthly, a trading method based on multi-attribute negotiation of cloud manufacturing services is proposed because of the characteristics, requirement and current situation of the SMEs. A utility function U(x) is built up in order to describe the affect of every attribute during the process of negotiation. A strategy based on U(x) and synchronous concessions which concluding the received or rejected of negotiation suggestion, the divide of received/rejected/discussed area and the model of time-based concessions is proposed. And then,the steps of trading method based on multi-attribute negotiation are introduced in detail.Finally, on the basis of above research, a system of cloud manufacturing platform face to forging industry is developed. The implementation environment, realization method and function of every module are introduced.

  • 【分类号】TH166;TP393.09
  • 【被引频次】6
  • 【下载频次】1730
  • 攻读期成果
节点文献中: 

本文链接的文献网络图示:

本文的引文网络