Research On The Remote Monitoring And Fault Diagnosis For Compressor Based On The Internet

Compressor is the key mechanical equipment of industrial field. If compressors appear abnormal condition or sudden fault in the process of running, it will bring on complete paralysis of equipment or unit and produce great economic loss, even cause major personnel casualties. Therefore, timely condition monitoring and fault diagnosis has important significance for safety production. with the compressor unit towards complication and networking, the traditional model of local monitoring and diagnosis has been difficult to meet the needs of the diagnosis, remote and intelligent diagnosis become a hot research. This dissertation researches the screw compressor and establishes the compressor remote monitoring and intelligent fault diagnosis expert system.Firstly, the overall architecture of remote monitoring platform is researched. According to the actual needs of the compressor monitoring, the network structure is confirmed. By comparing the structure of the current software platform, the RIA-based framework for internet applications which is developed by CBX solutions is determined, and use SQL Server2000 database management system for database development. In addition, for the network security strategy of the system platform, relevant research on hardware and software is carried out respectively.Based on the analysis of the basic structure and working principle of screw compressor, this dissertation sums up the common fault of screw compressor and analysis methods of vibration signals are researched mainly, Including the time domain analysis, frequency domain analysis and time-frequency analysis, and the measured actual signal is analyzed to verify the feasibility of the analysis methods.There are some problems such as knowledge acquisition "bottleneck" in the traditional expert system based on the rules, and neural network can solve the problems of traditional expert system effectively because of its parallel, adaptive, self-learning and other benefits. Based on the synthesis of their respective advantages, neural network expert system is put forward and its structure is analyzed and determined. According to the characteristics of the compressor fault which is various, Integrated neural network is established, which makes local diagnosis by individual neural networks and final information integration by decision network. In the processing of local diagnosis, multi-sensor information fusion is completed by the D-S evidence theory. Finally, the neural network expert system knowledge base is designed.At last, the compressor remote monitoring and fault diagnosis system based on the internet is developed. Its major functions include data acquisition and communication, real-time monitoring, data analysis, information search and management, rights management, report management, and expert system, etc.