【作者】 牛强；

【导师】 夏士雄；

【作者基本信息】 中国矿业大学 ， 电力电子与电力传动， 2010， 博士

【摘要】 矿井提升机是矿井安全生产过程中关键设备之一,是集机械、电气、液压于一体的矿山大型复杂机电系统,其系统单元之间以及系统单元内部都存在很多错综复杂、关联耦合的相互关系,不确定因素和不确定信息充斥其间,其故障可能是多故障、关联故障等多种复杂形式。因此,故障诊断问题是目前提升机系统维护工作中的首要问题。针对这些问题,本文以数字化提升机故障诊断技术为核心开展相关的研究,其主要工作包括:1.从提升机系统物理结构和逻辑组成出发,深入分析提升机故障模式和故障原因的传播机理,融合语义知识,对提升机系统单元及其故障领域知识进行分析,构建了语义环境下的矿井提升机故障诊断模型,提出了基于本体的提升机故障语义知识系统框架。2.通过对提升机系统及其故障现象、故障原因的逻辑关系进行分析,提出基于描述逻辑的提升机故障知识表示方法,对提升机系统单元及其故障领域知识进行语义描述,并对所构建的故障知识库进行了逻辑检错推理。3.为增强提升机故障语义知识库中故障规则更新完善的能力,应用粗集理论及属性约简方法,提出一种改进的故障规则更新算法,提高了提升机故障语义知识库中诊断规则的获取和更新的效率。4.通过对故障诊断推理方法的分析,借鉴专家系统中模糊规则推理的思想,根据提升机故障诊断知识相似性强的特点,提出了一种基于语义相似度的提升机故障知识推理方法,增强了提升机故障语义知识模型的推理能力,并兼顾了规则匹配的效率和准确性要求。5.通过构建提升机故障语义知识推理原型系统,从而实现了语义环境下提升机故障知识推理中故障知识的分析、转换和处理等关键问题,印证了提升机故障语义知识表示与推理方法的可行性和有效性,为提升机故障诊断方法提出了新的思路和新的探索。更多还原

【Abstract】 Mine hoist is a large and complex system of machine, electric and hydraulic pressure, which is the key equipment in mine safety production. Because of uncertain factors the faults may be multiple failures, associated faults or other complex forms. Therefore, how to resolve the problems of fault diagnosis becomes the most important issue to system maintenance. Then, related application research is carried out and corresponding system is developed in the background of the digital hoist system, the main contents are as following:1. By analyzing the fault pattern and fault transmitting mechanism and fusing semantic knowledge and ontology technology, fault diagnosis model is proposed in semantic environment, which mine hoist fault semantic knowledge system frame is put forward.2. The knowledge describing method of mine hoist fault knowledge is proposed based on Description Logic by analyzing the logic relation between fault symptom and fault causes, and design the logic reasoning.3. An improved fault rules update algorithm is put out in order to enhance the ability of updating and perfecting diagnosis rules in the mine hoist fault semantic knowledge base.4. By analyzing fault knowledge reasoning methods, a hoist fault knowledge reasoning method is proposed based on semantic similarity, which enhances the reasoning ability of the hoist fault semantic knowledge model and takes into account the efficiency and accuracy requirements of the rule matches.5. Mine hoist fault semantic knowledge system is building to solve the critical issues of analyzing, converting and processing the hoist fault knowledge in semantic environment, and it proves that the mine hoist fault semantic knowledge representation and reasoning methods are feasible and effective, and puts forward new ideas and new exploration for hoist fault diagnosis methods.更多还原