Condition Monitoring And Fault Diagnosis Of Turbine Governor Based On BP Network

Hydro generator group turbine governor is an important part of the load and frequency regulation, turbine governor generators have seriously affected the normal operation of the network security system as a whole and the quality of economic operation. Therefore, the states of repair of the turbine governor achieve full play Hydropower Plant efficiency is an important factor. For the realization of the "state of repair" this goal, we must assess the state of the system and fault diagnosis. Therefore, the state of maintenance condition monitoring and fault diagnosis is the core issue. It is essentially a major overhaul to achieve a state of technical support, advanced monitoring and diagnosis. Technical analysis is a necessary means of implementing state overhaul. Therefore, in order to improve the state of repair work Repair work of the state must be strengthened condition monitoring and fault diagnosis technology, and improvement of the study.This paper introduces the content of the equipment condition monitoring and fault diagnosis. It analyzes the condition monitoring and fault diagnosis of the situation. The development of artificial intelligence and computer-based condition monitoring and fault diagnosis of the parameters provided an important prerequisite. ANN has become one of the primary means of intelligent fault diagnosis. One of the most useful applications is to the multi-network, in the learning process using a back-propagation algorithm fuzzy reasoning diagnosis.In this paper, the author focuses on how to achieve BP neural network theory in the state of recognition of turbine governor. There is an in-depth study of the basic principles of BP neural network learning algorithms and conducted in various parts of the turbine governor based on the analysis of the structure and functions. The author has designed the model based on the parameters of the turbine governor condition monitoring and fault diagnosis. In particular, the research gives an identification of the relay response time of turbine governor, which solved an important problem for the testing device of turbine governor.This paper discusses the author's main work. The concrete work is, based on the design of the turbine governor BP artificial neural network model to achieve a state of literacy, which is a program design; solve condition monitoring and fault diagnosis of a series of data-processing techniques, which based on the actual design of the easy task of convergence and the BP neural network training speed. Finally, the author provides the specific method and simulation results using MATLAB neural network toolbox function method and the concrete results.