Research On Vibration Online Monitoring And Diagnosis Of Steam Turbine Generator Unit Based On Fuzzy Neural Network

Steam turbine generator unit is one of the core equipment in the power system, which is easy for malfunction and causes serious financial losses in the long-term continuous operation of high-speed. According to the status of steam turbine generator unit, in this paper, its importance of fault diagnosis was analyzed; the research situation quo of relevant fields at home and abroad was discussed; the advantages and diasadvantages of the diagnosis methods were analyzed. Thus, good effect was achieved by choosing the method of based on fuzzy neural network fault diagnosis to solve the problem.The main research content is shown as following.(1) The OPC technology was adopted in system, the data from present control equipment was accessed by the interfaces which OPC server provided. Thus, the problem of developing communication DRIVER constantly was avoided, the system of openness and interoperability was improved.(2) The main fault mechanism and characteristics of steam turbine generator unit was researched. Due to the fuzzy neural network not only has the advantages of numerical calculation of the neural network, but also has the ability to handle expert knowledge for fuzzy system, based on neural network and fuzzy system combining ways of study, a method of rotating machinery fault diagnosis on the basis of BP fuzzy neural network was adopted, and which was applied to fault diagnosis for steam turbine generator unit. The simulative experiment shows that the method was feasible and effective.(3) On the basis of in-depth analysis the process of fault diagnosis on steam turbine generator unit, using powerful development environment, using VC+ + 6.0, Pro Essentials chart controls and SQL Server 2000, a set of vibration online monitoring and diagnosis system of steam turbine generator unit was developed. The requirement of online monitoring on large-scale steam turbine generator unit and the functions of displaying the real-time data, pre-alarm and so on were realized. Its powerful software functions were certain to help for professional personnel to state analysis and fault diagnosis for unit.