基于多Agent技术的往复式压缩机在线智能诊断系统研究

发布时间：2018-04-07 16:06

本文选题：往复式压缩机　切入点：多Agent　出处：《大连理工大学》2012年硕士论文

【摘要】：往复式压缩机是石油化工行业中使用最广泛的机械设备之一,作为工业生产的关键设备,它的运转情况关系到整个生产线安全可靠,研究往复式压缩机的状态监测与故障诊断具有现实意义和经济价值。传统智能方法在面对大型复杂系统故障诊断时具有一定的局限性,而多Agent的反应性、社会性和主动性等优点能够弥补当前智能诊断方法在工程应用中的缺陷,较好地解决大型复杂系统故障诊断问题。本文主要研究了多Agent技术在往复式压缩机故障诊断中的应用。在分析了往复式压缩机基本结构和工作原理的基础上,结合往复式压缩机以往的故障诊断经验,总结了往复式压缩机的典型故障及其故障特征。提出了一种基于多Agent的往复式压缩机故障诊断系统总体框架。设计了该框架中各Agent的结构和功能以及它们之间的协作关系,该框架具有较好的扩展性。提出一种基于模糊信息融合的往复式压缩机故障诊断方法。在多Agent往复式压缩机故障诊断系统中采用混合分解方法对诊断任务进行分解,即在高层采用结构分解,底层采用故障分解；采用改进的合同网方法对诊断任务进行分配；介绍了专家系统、神经网络和模糊逻辑三种智能诊断方法与模糊信息融合方法,并通过仿真验证了模糊信息融合方法的有效性和准确性。开发了基于多Agent技术的往复式压缩机故障诊断系统。设计了往复式压缩机在线状态监测与故障诊断系统的整体方案,并详细阐述了该诊断系统的诊断对象、硬件组成和数据库的结构；然后利用LabVIEW和MATLAB混合编程开发出了基于多Agent技术的往复式压缩机故障诊断系统。该诊断系统已经应用于某石化车间往复式压缩机上,经过三个月的试运行,诊断系统稳定而有效。
[Abstract]:Reciprocating compressor is one of the most widely used mechanical equipment in petrochemical industry. As the key equipment in industrial production, its operation is related to the safety and reliability of the whole production line.It is of practical significance and economic value to study the condition monitoring and fault diagnosis of reciprocating compressor.The traditional intelligent method has some limitations when it comes to the fault diagnosis of large and complex systems, and the reactivity, sociality and initiative of multiple Agent can make up for the shortcomings of the current intelligent diagnosis method in engineering application.It can solve the problem of fault diagnosis of large complex system.This paper mainly studies the application of multi-Agent technology in fault diagnosis of reciprocating compressor.On the basis of analyzing the basic structure and working principle of reciprocating compressor, combined with the past fault diagnosis experience of reciprocating compressor, the typical faults and their fault characteristics of reciprocating compressor are summarized.A general framework of reciprocating compressor fault diagnosis system based on multi-Agent is presented.The structure and function of each Agent in the framework and the cooperative relationship between them are designed. The framework has good extensibility.A fault diagnosis method for reciprocating compressor based on fuzzy information fusion is proposed.In the fault diagnosis system of multiple Agent reciprocating compressors, the hybrid decomposition method is used to decompose the diagnosis task, that is, the structure decomposition is adopted in the high level, the fault decomposition is used in the bottom layer, the diagnosis task is assigned by the improved contract net method.Three intelligent diagnosis methods, expert system, neural network and fuzzy logic, and fuzzy information fusion method are introduced. The validity and accuracy of fuzzy information fusion method are verified by simulation.A reciprocating compressor fault diagnosis system based on multi-Agent technology is developed.Then a reciprocating compressor fault diagnosis system based on multiple Agent technology is developed by using LabVIEW and MATLAB.The diagnostic system has been applied to reciprocating compressor in a petrochemical workshop. After three months of trial operation, the diagnostic system is stable and effective.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH165.3

【相似文献】