Modeling On Fault Diagnosis Of Generator Rotor Winding Inter-turn Short-circuit And Its Application Research

Fault diagnosis for turbo-generators plays a very important role in the safe, stable and economic operation of the whole power system, and has great importance to national economy and the people's livelihood and national defense construction. Power grid is a huge system, once a fault happened, it would cause huge economic losses and disastrous consequences. Along with the rapid development of power system, fault diagnosis of the generator is continuously put forward new requirements. Technology of the generator fault diagnosis has become one of the important methods to ensure the power system operating safely and reliably. The advanced technology of generator fault diagnosis, especially the online monitoring technology has become a powerful guarantee for the safe and economic operation of power system. Therefore, the development of fault diagnosis system for generator and research on the fault diagnostic methods are getting more and more important. The fault diagnosis technique of the generator would be developing towards the computerization, networking and intelligent trend.The rotor winding inter-turn short circuit is a sort of usual electrical fault of generator, and it will affect the safety operation of electric power system very seriously. Early diagnosis and online monitoring of rotor winding inter-turn short circuit is a topic gained attention all around the world. In this paper, firstly it analyzed current situation of the research on rotor winding inter-turn fault diagnosis at home and abroad. Then it deeply analyzed the cause and the fault mechanism of the rotor winding inter-turn short circuit, and studied the change characteristic of electric parameters when the fault happened. Furthermore, it compared the advantage and disadvantage of various fault diagnosis technology. Finally, based on the study of particle swarm optimization algorithm and RBF neural network, it proposed a new fault diagnosis method of the rotor winding inter-turn short circuit by means of RBF neural network, and established a model of the generator rotor winding inter-turn short circuit fault diagnosis. In general, this paper has completed the following research works:â‘ Studied the fault mechanism of generator rotor winding inter-turn short circuit, and investigated the aberration of electromagnetic fields and the change characteristic of Electric parameters when the fault happened, and briefly compared the advantage and disadvantage of various traditional methods. It provided an important theoretical foundation for the selection of fault characteristicsâ‘¡Studied the application of RBF neural network in the rotor winding inter-turn short circuit fault diagnosis, and it provided a technological basis for the establishment of fault diagnosis model.â‘¢Introduced PSO-RBF neural network into the rotor winding inter-turn short circuit fault diagnosis, and established a fault diagnosis model based on PSO-RBF neural network.â‘£By simulation experiment and comparative study, the results showed that the fault diagnosis model based on PSO-RBF neural network can not only identify the rotor winding short circuit fault effectively, but also get a higher accuracy. Compared with traditional fault diagnosis models, it improves the diagnosis effect greatly.