基于图论的压缩机故障诊断算法实现

发布时间：2019-01-30 14:25

【摘要】：多级往复式压缩机通常由电动机通过皮带轮传动使曲轴产生旋转，经连杆带动活塞产生往复运动，从而使气缸、活塞顶面、阀片所组成的空间容积发生周期变化以到达压缩气体的目的。多级压缩机作为一类复杂的机电系统，各级汽缸状态相互关联、相互耦合，因此对其进行故障诊断有较大的难度。基于图论的故障珍断方法是故障诊断领域的一个重要分支，具有很大的应用和发展前景。本文以图论为理论基础，以三级往复式压缩机为研究对象，对这一方法若干重要方面进行了系统、深入地研究。图论应用于故障诊断时，其方法有两种：故障传播有向图DG和符号有向图SDG(Signed Directed Graph)。指出了基于传播有向图DG诊断方法的不足之处，系统阐述了基于符号有向图SDG的故障诊断方法。SDG的故障诊断方法，有着固有的缺陷，诊断分辨率和诊断速度有待提高。因此，本文将采用SDG和专家系统相结合的方法。本文的主要研究工作有两方面：一建立压缩机的SDG模型，二构建出人机交互界面。在SDG建模方面，针对往复式压缩机，，提出了一个综合性SDG故障诊断架构。该综合性故障诊断架构采用基于经验知识的建模方法，并引入了分级SDG建模的思想。在软件方面，使用C++语言，以Visual Studio2005、SQL Server2000为平台，来实现压缩机故障诊断系统的开发。基于SDG的故障诊断方法与专家系统相结合，具有很大的优势，是一种极具发展前途的故障诊断方法。
[Abstract]:The multi-stage reciprocating compressor is usually driven by the motor through the belt pulley to cause the crankshaft to rotate, and the connecting rod to drive the piston to produce reciprocating motion, thus making the cylinder and the top surface of the piston. The space volume of the valve sheet changes periodically to achieve the purpose of compressed gas. As a kind of complex electromechanical system, multistage compressors are interrelated and coupled with each other, so it is difficult to diagnose their faults. Graph theory based fault diagnosis method is an important branch of fault diagnosis field and has great application and development prospect. Based on graph theory and three-stage reciprocating compressor, this paper makes a systematic and in-depth study on several important aspects of this method. When graph theory is applied to fault diagnosis, there are two methods: fault propagation directed graph DG and symbolic directed graph SDG (Signed Directed Graph). This paper points out the shortcomings of the DG diagnosis method based on propagating directed graph, and expounds systematically the fault diagnosis method based on symbolic directed graph SDG. The fault diagnosis method of SDG has inherent defects, and the diagnosis resolution and speed need to be improved. Therefore, this paper will adopt the method of combining SDG and expert system. The main research work of this paper is as follows: first, establish the SDG model of compressor; second, construct the human-computer interface. In the aspect of SDG modeling, a comprehensive SDG fault diagnosis framework is proposed for reciprocating compressor. The integrated fault diagnosis architecture adopts empirical knowledge based modeling method and introduces the idea of hierarchical SDG modeling. In the aspect of software, C language and Visual Studio2005,SQL Server2000 are used to realize the development of compressor fault diagnosis system. The combination of fault diagnosis method based on SDG and expert system has great advantages and is a very promising fault diagnosis method.
【学位授予单位】：西安工程大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH45;TH165.3

【参考文献】