【作者】 郭春芬；

【导师】 钟佩思；

【作者基本信息】 山东科技大学 ， 机械设计及理论， 2011， 博士

【摘要】 随着社会经济从工业经济向知识经济的转变,知识逐渐成为生产力中最活跃、最重要的因素,知识管理作为一种全新的管理理论和管理方法应运而生,成为一个重要的研究领域。在制造业企业中,工艺知识是企业知识的重要组成部分,工艺知识管理是企业提高生产效率,降低消耗,发展生产的重要手段和保证。为了充分发挥工艺知识在企业生产中的作用,实现知识的实时共享、利用和创新,论文结合工艺知识的特点,从工艺知识表示、工艺知识检索和工艺知识挖掘三个方面对工艺知识管理的相关技术进行了研究。为实现企业工艺知识的积累,推动企业工艺知识的系统化、标准化管理和高效利用奠定了理论基础。针对工艺知识复杂性、隐含性、多样性和动态性的特点,提出了一种基于本体的工艺知识管理系统框架,给出了以知识检索、知识维护、知识学习、知识交流、知识挖掘和用户管理为主要功能的系统功能模型。围绕企业的生产过程和知识过程,以实现知识积累、知识共享、知识交流、知识创新为目的,同时利用用户的权限管理保证知识的安全性,为企业用户提供可靠的工艺知识管理。从工艺知识的特点出发,将工艺知识分为事实性知识、选择性知识和决策性知识,将本体引入工艺知识表示领域,建立了工艺知识概念本体层次结构。综合目前主要的本体建模方法,提出一种改进的循环综合工艺知识本体建模方法,研究更适合工艺知识的本体建模过程。在此基础上,以机械制造工艺知识为对象,提取部分工艺知识的本体概念,确定概念属性和关系,建立了轴类零件的本体概念模型,并对部分工艺知识进行了OWL语言的形式化表示,为后续的语义知识检索和知识挖掘奠定基础。针对传统检索技术查准率和查全率较低的问题,对基于本体的工艺知识语义检索技术进行了研究,通过计算工艺本体概念间的语义相似度,利用概念间语义的匹配,提高检索的全面性和精确性。在分析现有的相似度计算模型的基础上,综合考虑了语义距离及其权重、语义重合度、语义深度、语义密度和语义属性等影响因素,提出一种基于工艺本体的语义相似度计算方法,并针对工艺知识对该计算方法进行了合理性和精确性的验证。针对工艺知识的关联知识,提出一种基于本体的改进的关联规则挖掘算法。算法利用本体概念对工艺知识进行关联规则的挖掘,可以层次化地更为清晰的概括,通过生成十字链表结构,快速地得出频繁项集,减少了扫描数据次数和算法的运行时间,简化了算法,提高挖掘的速度。提出了工艺知识管理系统的设计思想和功能结构,开发了相应的原型系统,介绍了系统的开发环境,对系统的基本功能进行了实现。随着网络技术和信息技术的高速发展,新工艺、新技术的不断出现,工艺知识管理系统需要不断的改进和完善。在工艺知识挖掘和工艺知识领域本体建立等方面都需要进一步的研究和改进。更多还原

【Abstract】 With the change of the social economy from an industrial economy to a knowledge economy, the knowledge is becoming the most active and important factor in the productivity. Knowledge management, as a new management theory and management methods, has emerged and became an important area of research. In manufacturing enterprises, process knowledge is an important part of enterprise knowledge, and the process knowledge management is the important means and assurance for the enterprise to improve production efficiency, reduce the consumption, and develop the production. In order to give full play to the role of the process knowledge in the enterprise production, and to achieve the real-time sharing and use of knowledge and its innovation, this dissertation, with the combination of characteristics of process knowledge, did an investigation on related technologies of the process knowledge in three aspects, i.e. process knowledge representation, process knowledge retrieval and process knowledge mining. The results of this research intend to lay a theoretical foundation to achieve the accumulation of process knowledge, promote the systematic and standardization management, and efficient utilization of the process knowledge.Regarding the complexity, implication, diversity and dynamic features of process knowledge, a process knowledge management ontology-based system framework was put forward. The system function model was established, with the main function given to knowledge retrieval, knowledge maintenance, knowledge learning, knowledge exchanges and knowledge mining and user management. Centered around the production process and knowledge process, this dissertation, by taking the advantage of user permission management to guarantee the safety of knowledge, made possible the realization of the knowledge accumulation, knowledge sharing, knowledge exchanges and knowledge innovation, and at the same time provided a reliable craft knowledge management. With a point of departure from the characteristics of technological knowledge, process knowledge was divided into factual knowledge, selective knowledge and decision-making knowledge, and the ontology introduced into the process knowledge representation field, and process knowledge in conception-ontology hierarchy was established. Given the current main body modeling methods, this dissertation put forward an improved ontology modeling method-circle synthesis, which could help to research more appropriate process knowledge for ontology modeling process. Based on the above study, by taking the machinery manufacturing process knowledge as an object, this research, by extracting ontic conceptions of parts of the process knowledge, determining the nature and relations of the concepts, established the corresponding ontology models, and represented parts of the process knowledge by the OWL language, which laid the foundation of subsequent semantic knowledge retrieval and knowledge mining. In the case of the defects of the low precision and recall ratio of the traditional retrieval technology, semantic knowledge retrieval technology based on ontology was researched. By calculating the semantic similarities between ontic conceptions, and by using semantic matching between concepts, the completeness and accuracy of retrieval were improved. On the basis of analyzing the current similarity-calculating models, and with the consideration of all the influencing factors including the degree of semantic overlap, semantic depth and density, and semantic attributes, and etc., this dissertation proposed a modified ontology-based semantic similarity calculation method, and checked its rationality and accuracy use in process knowledge. For the relating knowledge of the process knowledge, an improved ontology-based mining algorithm of the relating rules was proposed. The algorithm explored the relating rules of the process knowledge by using the ontic conceptions, which provided a clearer summary of the structure layered, thus generating cross-list, so as to quicken the things’support degree counting, shorten the process of scanning time, and simplify the mining process, the author proposed the design idea and functional structure of the process knowledge management system, developed the corresponding prototype, introduced the development environment of the system, and formed the system basic function.With the rapid development of network technology and information technology, and the emergence of new process and new technologies, the process knowledge management system needs continuous improvement and perfection. And it also needs investigation and promotion on process knowledge mining and ontology of process knowledge establishing, and. etc.更多还原