基于案例推理的往复压缩机故障诊断专家系统研究

发布时间：2019-03-26 20:03

【摘要】：往复压缩机作为流程工业应用广泛的一类关键设备，一旦发生故障将对生产造成严重影响，因而对其故障诊断就显得尤为重要。现如今对设备的故障诊断逐步发展为基于人工智能的专家系统诊断阶段，促使对往复压缩机故障诊断专家系统的研究也都有了深入的发展。基于案例推理是人工智能领域中新兴的一种高效的问题求解方法，在故障诊断领域是采用近似于专家诊断故障的思维模型的而建立故障诊断专家系统的方法，将此方法应用于往复压缩机故障诊断专家系统，对提高往复压缩机的使用率、减少事故发生、降低维修费用有着非常重要的意义。本文结合往复压缩机的结构特点，，工作原理以及各种故障机理，研究了基于案例推理技术应用于往复压缩机故障诊断专家系统的若干关键技术问题：应用面向对象知识表示案例知识，建立具有层次结构的案例库；对案例分层组织及索引，采用了改进的层次分析法—三标度层次分析法进行权重计算方法；采用优化的K-近邻加权相似度计算方法进行推理，为了将基于规则推理与基于案例推理的专家系统相互融合，研究了两者融合机制及案例转化为规则的方法。最后基于知识表示和推理技术的研究成果，设计了基于案例推理的往复压缩机故障诊断专家系统的功能结构和诊断流程，开发实现了专家系统。
[Abstract]:Reciprocating compressor, as a kind of key equipment widely used in process industry, will have a serious impact on production in case of failure, so it is very important to diagnose the fault of reciprocating compressor. Nowadays, the fault diagnosis of equipment is gradually developed into the diagnosis stage of expert system based on artificial intelligence, which makes the research on fault diagnosis expert system of reciprocating compressor also have in-depth development. Case-based reasoning (CBR) is a new and efficient problem solving method in the field of artificial intelligence. In the field of fault diagnosis, a fault diagnosis expert system is established by adopting a thinking model similar to that of expert diagnosis. The application of this method to the fault diagnosis expert system of reciprocating compressors is of great significance for increasing the utilization rate of reciprocating compressors, reducing the occurrence of accidents and reducing the maintenance costs. In this paper, combined with the structure characteristics, working principle and various fault mechanism of reciprocating compressor, Several key technical problems of case-based reasoning (CBR) applied to fault diagnosis expert system of reciprocating compressor are studied in this paper: applying object-oriented knowledge to represent case knowledge and establishing case base with hierarchical structure; For the hierarchical organization and index of the case, the improved analytic hierarchy process-three-scale analytic hierarchy process is used to calculate the weight. In order to integrate the rule-based reasoning and the case-based reasoning expert system, the fusion mechanism and the method of case-based transformation into rules are studied in order to integrate the rule-based reasoning and case-based expert system by using the optimized K-nearest neighbor weighted similarity calculation method. Finally, based on the research results of knowledge representation and reasoning technology, the functional structure and diagnosis flow of the fault diagnosis expert system for reciprocating compressor based on case-based reasoning are designed, and the expert system is developed and implemented.
【学位授予单位】：北京化工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH452;TH165.3

【参考文献】