Study On Steam Turbine Operation Monitoring And Health Management System

Steam turbine is a large rotating machinery with complicated structure, poor working environment, so,there are multiple faults in its operation process.In order to guarantee the safety operation of the steam turbine, to develop a system with condition monitoring,fault diagnosis,health management is of great significance to Steam turbine. Recent years, the Internet and mobile communication technology developing rapidly, promoted by these technologies, smart phone,tablet PC,smart watches and other mobile terminal devices is widely used.Along with the rise of mobile devices,mobile application APP is deep into all aspects of people’s life and make a great impact on people’s production and life.The enrichment and diversity of APP’s content make people’s daily life more convenient,besides,Big Data era puts forward a concept that it should be the movement of data rather than people.The purpose of this subject is to develop a monitoring and health management system of steam turbine,it has the function of data acquisition, data analysis, fault diagnosis, data storage, trend analysis and remote control on PC.Also, designed and developed a APP in the mobile terminal which can remote control the PC software.The paper studies several typical faults of the turbine unit, according to the fault mechanism of the signal, made a detailed research in time domain and frequency domain.In time domain analysis, monitoring the signal original waveform, measuring the wave peak, transient characteristics, average, maximum, minmum,etc.In addition,in frequency domain,using the wavelet transform, power spectrum and phase spectrum, make the signal characteristics can be more clearly shown in the monitoring software.Wavelet packet frequency analysis technology can extract the fault signal energy of each band, after the energy normalization process it can get the fault feature vector,the BP neural network has association,speculation and memory function,accurate recognition of complex system,so it is very suitable for fault diagnosis and trend analysis of mechanical equipment.Based on a large number of steam turbine fault database. Combining wavelet packet with BP neural network to their advantages and function can establish the network model of fault diagnosis.Combining wavelet packet with BP neural network to their respective advantages and function, can establish the network model of fault diagnosis.Industrial computer can be used as a server after some configuration on Windows’ IIS.According to the network communication characteristics of Lab VIEW,using Data Socket technology as a means of remote monitoring in LAN,while choosing Lab VIEW Web publishing technology in WAN,published the monitoring and health management system on the web page,then the wan browser has the access to realize the remot control.The design and develop of the APP application software is programming on the Eclipse platform, the Android running environment need to be established on Windows System.The paper designed the layout of the APP, UI interface and communicate with the PC server, completed the development of the APP.when the mobile terminal is connected to the Internet,users can monitor the scene.Finally, experiments have been carried out to test on PC software and mobile terminal, the results show that the steam turbine monitoring and health management system runs stably on PC and mobile terminal, meets the design requirement of manufacturer.