Fault Diagnosis For Trun To Turn Short Circuit In Rotor Windings Of Turbo-generator

With the rapid development of China’s economy, there is increasing demands for electricity in the society. As an important equipment of generating, turbo-generators play a very important role in safety and stability of the whole power system. Developing fault diagnosis technology of generator to find fault timely and effectively has gradually become one of the important methods to ensure the equipment operating safely and reliably. Due to the high-speed rotation of the generator, the rotor endures extreme mechanical and thermal stresses, which can easily cause a short circuit fault between the rotor windings. Turn-to-Turn short circuit of rotor windings is a usual electric fault, and it can cause point or two ground of the rotor easily or further accident. Therefore, it is significant to surpvise and diagnosis for turn-to-turn short circuit fault in rotor windings of generator.In this paper summarizes the current detection methods at home and abroad on inter-turn short circuit fault of the rotor, and analyzes their advantages and disadvantages.Then, introduces the structure of generator rotor and causes of inter-turn short circuit fault, pointing out that poor manufacturing processes and insulation damage are the main reason causing short circuit fault between rotor windings. Then, analyse the electromagnetic and vibration characteristics of the rotor when inter-turn short circuit occurs, establish a two-dimensional model of the magnetic field. When the inter-turn short circuit occurs, the rotor excitation current increases and the reactive power is reduced. In addition, generator vibration increases under inter-turn short circuit, primarily caused by electromagnetic tension imbalances and asymmetries heat.BP neural network as a new type of intelligent fault diagnosis method, you can not need to know the precise mathematical correspondence between failure and data. Use Matlab to establish a BP neural network model, put some electrical parameters of the generator as the input of the network, and success in fault diagnosis on short-circuit between the generator rotor turns, and can predict the severity of the fault.