【作者】 郭建文；

【导师】 于德介；

【作者基本信息】 湖南大学 ， 机械工程， 2010， 博士

【摘要】 设备维护是保证制造企业经营成功的关键因素。提升企业的设备维护水平不仅有助于企业实现其经营目标,也将推动生产、环境与社会的和谐发展。近年来兴起的先进制造方式极大地改变了设备维护的工作对象和工作环境,使得先进的设备维护模式得到研究与应用。e-维护作为新兴的设备维护模式,能够整合维护联盟的核心维护能力,通过维护资源的高效管理和维护流程的优化配置实现维护效益的最大化。维护知识作为e-维护的核心资源和维护流程的关键支撑要素,需要高效的管理,使得其能够在业务需求的驱动和权限的约束下在维护联盟中共享,以提高e-维护工作的质量。知识网格作为构建在网络上的知识集成平台,能够为知识资源的高效管理提供支持。为了有效提升e-维护知识资源管理的水平,论文在国家自然科学基金项目“设备e-维护模式理论及支撑技术研究”(项目编号:70601010)的资助下,对基于知识网格的e-维护知识资源管理问题进行了深入系统的研究。本文主要研究工作和创新性成果有:(1)结合e-维护知识资源管理的需求,提出了基于知识网格的e-维护知识资源管理(Knowledge Grid based e-Maintenance Knowledge Resources Management, KG-EMKRM)模型。KG-EMKRM模型以知识网格体系为基础,结合本体、维护知识空间和网格服务对知识资源进行集成,以业务过程为中心对知识资源进行管理。对模型进行了形式化定义,辨识了模型的组成要素;探讨了模型的内涵、特征和应用模式;从工程应用的角度设计了模型的体系结构和支持系统。(2)建立了面向e-维护知识资源管理的领域本体模型,为知识资源管理提供了语义基础。该领域本体模型包括e-维护知识元本体和e-维护联盟本体。其中,e-维护知识元本体为e-维护知识资源管理领域提语义清晰的理解基础;e-维护联盟本体表达维护联盟的运作知识,为联盟成员之间的知识协作提供共同的语义基础。提出了基于本体集成的e-维护知识元本体构建方法;详细探讨了e-维护联盟本体中的维护组织本体类、维护业务本体类和设备本体类的设计方法。(3)提出了面向e-维护知识管理的知识资源建模框架。为了实现结构化维护知识的规范收集和存储,采用本体对e-维护中的结构化知识进行表示,并以维护案例本体表示为例,探讨了本体的表示方法;引入资源空间模型,并将其扩展为多个面向知识资源管理的子空间,建立了维护知识空间模型,对知识资源进行描述,以支持资源的有序化管理;设计了知识网格服务模型,并构建了基于扩展OWL-S(Web Ontology Language for Services)的知识网格服务本体,以支持知识网格服务的语义描述。(4)提出了面向维护业务过程的知识支持模型,以实现知识资源在e-维护业务驱动下的共享。对模型实现的关键技术进行了研究:设计了面向e-维护联盟的知识关联网络模型,实现知识资源与维护业务的集成,提高知识索引的效率;建立了面向业务的知识供应模式,将知识需求集成到业务上下文中,以支持按需、及时的知识供应;提出了以服务功能匹配和服务质量(Quality of Service, QoS)匹配为基础的知识网格服务发现方法,以支持知识资源服务的搜索。(5)为了使得e-维护的知识资源能够权限的约束下共享,结合基于上下文的安全管理和知识管理技术,提出了面向e-维护联盟的知识访问控制(e-Maintenance Alliance Knowledge Access Control,EMAKAC)模型。与传统的基于角色的访问控制模型相比,EMAKAC模型以e-维护联盟的组织结构为基础,以维护业务上下文作为访问控制的授权中介来实现业务驱动的知识安全管理。(6)为使KG-EMKRM模型具有实践指导意义,结合某石化企业的e-维护平台知识资源管理的需求,设计开发了基于KG-EMKRM模型的原型系统,并进行验证;结合背景企业e-维护的实施,设计了KG-EMKRM模型的实施方法。原型系统的应用实践表明KG-EMKRM模型在实践中是可行的,可有效支持e-维护平台的知识资源管理。本文的初步研究工作表明,将知识网格的理论和技术体系应用于e-维护知识资源管理领域是可行、有效的。本文的研究成果能够为e-维护知识资源管理提供新的思路和实践经验。更多还原

【Abstract】 Plant maintenance is a critical factor in the success of manufacturing enterprises. Upgrading the level of plant maintenance will not only help enterprises achieve their business objectives, but also promote the harmonious development of production, environment and the society. In recent years, advanced manufacturing pattern has greatly changed the working subjects and environment of plant maintenance. Accordingly, many advanced maintenance modes are being studied and developed. As a new equipment maintenance mode, e-maintenance can integrate the core maintenance capability in e-maintenance alliance to achieve maximum maintenance efficiency by efficient management for maintenance resources and optimal allocation for maintenance process.Maintenance knowledge is one of core e-maintenance resources and critical support element of maintenance process. To improve the quality of e-maintenance work, maintenance knowledge requires efficient management. As an integrated knowledge platform in the Internet, knowledge grid has the ability to support the efficient management for knowledge resources. In order to upgrade the level of knowledge resources management for e-maintenance, sponsored by the National Natural Science Foundation of China (NO. 70601010), the present dissertation did a profound and systematic research on the key technologies of knowledge grid based knowledge resources management for e-maintenance.The main contributions of this dissertation are summarized as follows:(1) Considering the demands of the knowledge resources management for e-maintenance, a knowledge grid based e-maintenance knowledge resources management (KG-EMKRM) model was introduced in the present dissertation. In the KG-EMKRM model, knowledge grid architecture was adopted, ontology, maintenance knowledge space and grid service were employed in integration of knowledge resources in knowledge grid, and then knowledge resources management was performed by business-process-centered method. A formal definition of model was introduced to identify the components of KG-EMKRM model. The connotation, characteristic and application mode of model were discussed. The architecture and support system of model were designed from the view point of practice.(2) A domain ontology model for knowledge resources management of e-maintenance was established to provide semantic infrastructure for knowledge resources management. The domain ontology model consisted of e-maintenance knowledge meta-ontology (EMKMO) and e-maintenance alliance ontology (EMAO). The EMKMO provided semantic basis of understanding for knowledge resources management of e-maintenance; The EMAO expressed the knowledge of e-maintenance alliance operation, and then provided semantic basis for knowledge collaboration between the alliance members. A method based ontology integration was proposed to build the EMKMO. The design method of maintenance organization ontology (MOO), maintenance business ontology (MBO) and equipment ontology (EO) was present.(3) A knowledge resources modeling framework for e-maintenance knowledge resources management was proposed. In order to capture and storage structured maintenance knowledge resources, ontology was adopted to express structured maintenance knowledge resources, and then taking the ontology expression of maintenance case as an example, a method of ontology expression was presented. To achieve ordering management of knowledge resources, by introducing the resource space model and extending it to multiple sub-spaces for knowledge resources management, a maintenance knowledge space (MKS) model was set up for description of knowledge resources. A knowledge grid service model was established, and then knowledge grid service ontology based OWL-S (Web Ontology Language for Services) was also set up for semantic description of knowledge grid service.(4) To achieve the knowledge sharing under business-driven, a knowledge support model for maintenance business process was proposed and key techniques of the model were studied. In order to improve the efficiency of knowledge index, a e-maintenance knowledge relation network was designed to integrate maintenance knowledge into maintenance business process. Integrating requirements of knowledge into maintenance business context, a business-oriented knowledge supply mode was set up to support on-demand and timely knowledge supply. A service discovery method based service function matching and Quality of Service (QoS) matching was proposed to support service search.(5) To achieve the knowledge sharing under authority constraints in e-maintenance alliance, an e-maintenance alliance knowledge access control (EMAKAC) model was proposed based upon context-based security management and context-based knowledge management technology. Comparing with role-based access control (RBAC) model, The EMAKAC model was on the basis of organization structure of e-maintenance alliance and used maintenance business context (MKC) as the authorization intermediary to achieve business-driven knowledge security management.(6) In order to make the KG-EMKRM model has a practical significance, considering the demands of knowledge resources management in the e-maintenance platform of cooperation enterprise, a corresponding prototype system had been developed, validated, and improved. Together with the implementation of e-maintenance platform, the implementation method of KG-EMKRM model had been designed。The application of the prototype system showed the KG-EMKRM model was feasible in practice and can be effectively support knowledge resource management in e-maintenance platform.The preliminary study of this dissertation showed that applying the theory and technical architecture of the knowledge grid to knowledge resource management in e-maintenance is feasible and effective. The results of this study can provide new ideas and practical experience for knowledge resource management in e-maintenance.更多还原