基于Internet的压缩机远程监测与故障诊断技术研究
发布时间：2019-06-26 12:51
【摘要】：压缩机是工业领域的关键机械设备,一旦其出现异常或突发状况,往往会造成整套设备或机组的瘫痪,为企业和社会造成巨大的经济损失,甚至会引起重大的人员伤亡事故。所以及时的状态监测及故障诊断对安全生产有着重要意义。同时随着压缩机机组的复杂化、网络化的发展,传统的本地监测与诊断模式已很难满足诊断的需求,所以远程化与智能化成为了当前研究的热点。本文以螺杆压缩机为研究对象,建立压缩机远程监测平台以及智能故障诊专家系统。 首先对远程监测平台的总体架构进行了研究,根据压缩机监测的实际需求,确定了其网络结构,并通过对当前软件平台结构的对比,确定了基于RIA的互联网应用框架,由CBX解决方案进行开发,并选用SQL Server2000数据库管理系统进行数据库开发。另外对于系统平台的网络安全策略,分别在硬件与软件上进行了相关研究。 在对螺杆机基本结构及工作原理分析的基础上,总结出螺杆压缩机常见的故障。主要研究了基于振动信号的分析方法,包括时域分析、频域分析以及时频分析,并通过实测信号进行分析,以验证分析方法的可行性。 传统基于规则的专家系统存在着知识获取“瓶颈”等问题,而神经网络因其具有并行、自适应性、自学习性等优点,有效地解决了传统专家系统遇到的问题。本文综合其各自优点,提出了神经网络专家系统,分析并确定了神经网络系统的结构。针对压缩机故障种类繁多的特点,采用集成神经网络的思想,将各类故障由诊断子网络进行诊断,最后再由决策网络进行融合。在诊断子网络的处理过程中,多传感器的局部信息融合则由D-S信息融合完成。最后,对神经网络专家系统的知识库进行了设计。 最后设计开发出基于internet的压缩机远程监测及故障诊断系统,其功能包括数据采集与通信、实时监测、数据分析、信息查询与管理、权限管理、报表管理以及专家系统等等。
[Abstract]:Compressor is the key mechanical equipment in the industrial field. Once it appears abnormal or unexpected situation, it will often cause the paralysis of the whole set of equipment or units, cause huge economic losses for enterprises and society, and even cause heavy casualties. Therefore, timely condition monitoring and fault diagnosis is of great significance to safety in production. At the same time, with the complexity of compressor units and the development of network, the traditional local monitoring and diagnosis model has been difficult to meet the needs of diagnosis, so remote and intelligent has become the focus of current research. In this paper, the remote monitoring platform and intelligent fault diagnosis expert system of screw compressor are established. Firstly, the overall architecture of remote monitoring platform is studied, and its network structure is determined according to the actual requirements of compressor monitoring. Through the comparison of the current software platform structure, the Internet application framework based on RIA is determined, which is developed by CBX solution, and the SQL Server2000 database management system is selected for database development. In addition, the network security strategy of the system platform is studied in hardware and software. Based on the analysis of the basic structure and working principle of screw compressor, the common faults of screw compressor are summarized. The analysis methods based on vibration signal, including time domain analysis, frequency domain analysis and time frequency analysis, are mainly studied, and the feasibility of the analysis method is verified by the measured signal analysis. There are some problems in the traditional rule-based expert system, such as the bottleneck of knowledge acquisition, and the neural network has the advantages of parallelism, adaptability and self-learning, which effectively solves the problems encountered by the traditional expert system. In this paper, based on their respective advantages, a neural network expert system is proposed, and the structure of the neural network system is analyzed and determined. According to the characteristics of many kinds of compressor faults, the idea of integrated neural network is adopted to diagnose all kinds of faults by diagnosis subnetwork, and finally to merge them by decision network. In the process of diagnosis subnetwork processing, multi-sensor local information fusion is completed by D / S information fusion. Finally, the knowledge base of neural network expert system is designed. Finally, a compressor remote monitoring and fault diagnosis system based on internet is designed and developed, which includes data acquisition and communication, real-time monitoring, data analysis, information query and management, authority management, report management and expert system.
【学位授予单位】：江南大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP274;TH45
本文编号：2506192
[Abstract]:Compressor is the key mechanical equipment in the industrial field. Once it appears abnormal or unexpected situation, it will often cause the paralysis of the whole set of equipment or units, cause huge economic losses for enterprises and society, and even cause heavy casualties. Therefore, timely condition monitoring and fault diagnosis is of great significance to safety in production. At the same time, with the complexity of compressor units and the development of network, the traditional local monitoring and diagnosis model has been difficult to meet the needs of diagnosis, so remote and intelligent has become the focus of current research. In this paper, the remote monitoring platform and intelligent fault diagnosis expert system of screw compressor are established. Firstly, the overall architecture of remote monitoring platform is studied, and its network structure is determined according to the actual requirements of compressor monitoring. Through the comparison of the current software platform structure, the Internet application framework based on RIA is determined, which is developed by CBX solution, and the SQL Server2000 database management system is selected for database development. In addition, the network security strategy of the system platform is studied in hardware and software. Based on the analysis of the basic structure and working principle of screw compressor, the common faults of screw compressor are summarized. The analysis methods based on vibration signal, including time domain analysis, frequency domain analysis and time frequency analysis, are mainly studied, and the feasibility of the analysis method is verified by the measured signal analysis. There are some problems in the traditional rule-based expert system, such as the bottleneck of knowledge acquisition, and the neural network has the advantages of parallelism, adaptability and self-learning, which effectively solves the problems encountered by the traditional expert system. In this paper, based on their respective advantages, a neural network expert system is proposed, and the structure of the neural network system is analyzed and determined. According to the characteristics of many kinds of compressor faults, the idea of integrated neural network is adopted to diagnose all kinds of faults by diagnosis subnetwork, and finally to merge them by decision network. In the process of diagnosis subnetwork processing, multi-sensor local information fusion is completed by D / S information fusion. Finally, the knowledge base of neural network expert system is designed. Finally, a compressor remote monitoring and fault diagnosis system based on internet is designed and developed, which includes data acquisition and communication, real-time monitoring, data analysis, information query and management, authority management, report management and expert system.
【学位授予单位】：江南大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP274;TH45
【引证文献】
相关硕士学位论文 前1条
1 王芹芹;基于Internet的简易智能小车监控系统设计[D];华中师范大学;2013年
，本文编号：2506192
本文链接：https://www.wllwen.com/kejilunwen/jixiegongcheng/2506192.html
