面向滚齿机故障诊断的知识管理方法及应用研究

发布时间：2018-01-23 17:59

本文关键词： 知识管理故障诊断滚齿机　出处：《重庆大学》2011年硕士论文　论文类型：学位论文

【摘要】：在工业企业转型升级的迫切需求下,生产设备的故障诊断技术正成为制造业服务化的重要研究方向,成为企业提高用户满意度、提升竞争力的重点内容。信息技术的发展推动着传统制造业向服务的转型,为故障诊断技术提供了有效手段。在这种背景下,信息和知识已经成为竞争性资产,所以,在信息化的基础上针对设备故障诊断领域开展知识管理,不仅会提升服务质量,也会带来显著的经济效益。滚齿机作为公认的最复杂的工作母机之一,在各种机械制造业中应用广泛,其故障维修具有涉及地域广、技术要求高、对工作经验依赖性强等特点,而由于滚齿机生产厂家管理水平不高,造成了大量的维修数据与记录处于闲置或低利用状态。因此,本文以滚齿机的故障诊断领域为研究对象开展知识管理的研究,以达到维修知识共享和知识利用最大化的目的,满足用户简化工作量、及时有效排除故障的需求。首先,系统地分析了滚齿机的结构及其工作状态,针对故障的层次性、多源性和相关性等特点,提出了一种面向滚齿机故障诊断领域的知识管理框架,该知识管理框架以资源层为支撑,通过实施层予以实现,在应用层得到验证。在此基础上,设计了知识管理的流程。其次,从滚齿机故障诊断的知识表示、知识获取、知识存储和知识维护等四个方面对知识管理的关键技术进行研究。针对滚齿机故障诊断维修的特点,提出了基于本体的知识表示方法,形式化地定义了滚齿机本体概念、属性、关系、实例和规则等基本要素,建立了滚齿机故障诊断的本体模型;在该本体模型的基础上,本文使用人工获取和计算机识别更新结合的半自动化知识获取方法,将故障信息结构化;采用了基于OWL文档的知识存储方法,更易于知识维护和更新;为适应本体动态性的变化,采用本体推理机FaCT++工具对知识库进行推理检验,确保了知识库的完备性和一致性。最后,针对以上理论和实现方法的研究,对相关关键技术进行了实现,开发了面向滚齿机故障诊断领域的知识管理原型系统,验证本文提出方法的可行性和实用性。
[Abstract]:Under the urgent need of the transformation and upgrading of industrial enterprises, the fault diagnosis technology of production equipment is becoming an important research direction of manufacturing service-oriented, and it has become an enterprise to improve customer satisfaction. The development of information technology promotes the transformation of traditional manufacturing industry to service, and provides an effective means for fault diagnosis technology. In this context, information and knowledge have become competitive assets. Therefore, on the basis of information technology, knowledge management in the field of equipment fault diagnosis will not only improve the quality of service, but also bring significant economic benefits. Gear hobbing machine is recognized as one of the most complex working mother machine. It is widely used in various mechanical manufacturing industries, and its fault maintenance has the characteristics of wide area, high technical requirements and strong dependence on working experience. However, the management level of gear hobbing machine manufacturers is not high. As a result of a large number of maintenance data and records are idle or low utilization. Therefore, this paper takes the fault diagnosis field of gear hobbing machine as the research object to carry out the research of knowledge management. In order to achieve the purpose of sharing maintenance knowledge and maximizing the utilization of knowledge, it can meet the needs of users to simplify the workload and to troubleshoot faults in a timely and effective manner. Firstly, the structure and working state of gear hobbing machine are analyzed systematically. According to the characteristics of fault hierarchy, multi-source and relativity, a knowledge management framework for fault diagnosis of gear hobbing machine is proposed. The knowledge management framework is supported by the resource layer, implemented by the implementation layer, and verified in the application layer. On this basis, the process of knowledge management is designed. Secondly, the key technologies of knowledge management are studied from four aspects: knowledge representation, knowledge acquisition, knowledge storage and knowledge maintenance. The knowledge representation method based on ontology is proposed, and the basic elements of gear hobbing machine ontology such as concept, attributes, relations, examples and rules are formally defined, and the ontology model of gear hobbing machine fault diagnosis is established. Based on the ontology model, this paper uses semi-automatic knowledge acquisition method which combines manual acquisition and computer identification update to structure fault information. The knowledge storage method based on OWL document is adopted, which is easier to maintain and update knowledge. In order to adapt to the change of ontology dynamics, the ontology reasoning engine (FaCT) is used to test the knowledge base, which ensures the completeness and consistency of the knowledge base. Finally, according to the research of the above theory and implementation method, the related key technologies are implemented, and a knowledge management prototype system for hobbing machine fault diagnosis field is developed. The feasibility and practicability of the proposed method are verified.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH165.3

【参考文献】