基于描述逻辑的智能故障诊断方法研究

发布时间：2018-08-19 13:13

【摘要】：基于知识处理的智能故障诊断方法是当前故障诊断研究的热点。其关键在于智能诊断领域的知识表示以及知识推理上,即知识库的构建和推理机的运用。现如今的知识表示、知识处理方法在诊断领域都表现出一定的不足之处,如逻辑表示法、产生式表示法虽然知识的推理能力较强,但是对于结构复杂、故障关联性、交错性较强的故障领域,则呈现出表示能力不足的缺陷。基于网络的知识表示方法具有很强的知识知识表示能力,但推理的复杂度比较高,难以在合理时间内给出诊断答案,同时这些方法在处理蕴含性知识以及在检测经验性知识的逻辑正确性上面都不能很好的满足要求。针对现有的知识表示、知识处理方法在故障诊断领域的不足以及描述逻辑在知识表示、知识处理上所呈现出表达能力强、推理复杂度低的特性,本文将描述逻辑应用于故障诊断领域提出了基于描述逻辑的智能故障诊断方法,课题的应用背景为电梯系统的故障诊断。本论文的主要工作如下： 1)详细分析了故障诊断领域中各种知识获取、知识表示以及推理方法的优缺点针对性地提出了基于描述逻辑的智能故障诊断方法； 2)运用描述逻辑来表示故障领域知识、构建知识库、并进行推理； 3)针对电梯系统领域,构建了基于描述逻辑的电梯故障知识库并进行推理实现； 4)设计实现了电梯故障诊断及安全评价系统软件。本文将描述逻辑理论应用于故障诊断,利用分层分类方法构建知识库并进行推理,针对地处理了知识结构复杂、表现形式多样交错性强的故障问题,在推理过程中很好地利用了领域蕴含知识,知识库利用率高、系统诊断速度快,在知识表示能力和推理复杂度二者之间取得了很好的折中。最后开发实现了电梯故障诊断及安全评价系统,具有实际的应用推广价值。
[Abstract]:Intelligent fault diagnosis based on knowledge processing is a hot topic in fault diagnosis. The key lies in the knowledge representation and knowledge reasoning in the field of intelligent diagnosis, that is, the construction of knowledge base and the application of inference machine. Nowadays, knowledge representation and knowledge processing methods have some shortcomings in the field of diagnosis, such as logical representation, production representation, although the reasoning ability of knowledge is strong, but the structure is complex, and the fault is related to each other. The fault areas with strong interlacing have the defect of insufficient expression ability. The knowledge representation method based on network has strong knowledge representation ability, but the complexity of reasoning is high, so it is difficult to give the diagnosis answer in a reasonable time. At the same time, these methods can not meet the requirements in dealing with implied knowledge and detecting the logical correctness of empirical knowledge. In view of the shortcomings of existing knowledge representation, knowledge processing methods in fault diagnosis and description logic in knowledge representation, knowledge processing has the characteristics of strong expressive ability and low reasoning complexity. In this paper, an intelligent fault diagnosis method based on description logic is proposed in the field of fault diagnosis. The application background of this paper is the fault diagnosis of elevator system. The main work of this paper is as follows: 1) the knowledge acquisition, knowledge representation and reasoning methods in the field of fault diagnosis are analyzed in detail, and the intelligent fault diagnosis method based on description logic is put forward. 2) use description logic to express fault domain knowledge, construct knowledge base, and infer; 3) construct elevator fault knowledge base based on description logic and realize reasoning for elevator system domain; 4) the software of elevator fault diagnosis and safety evaluation system is designed and implemented. In this paper, the description logic theory is applied to fault diagnosis, the knowledge base is constructed by hierarchical classification method and reasoning is carried out, and the problems of complex knowledge structure and strong interleaving in various forms are dealt with in this paper. In the process of reasoning, it makes good use of the knowledge contained in the domain, has a high utilization rate of knowledge base, and makes a good compromise between the ability of knowledge representation and the complexity of reasoning. Finally, the elevator fault diagnosis and safety evaluation system is developed, which has practical application value.
【学位授予单位】：西安工业大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH165.3

【相似文献】