| With the development of the power system and the improvement requirements for the safety operation,although the conventional relay protection principle and fault diagnosis technology has been already very perfect, but has not fully adapted to the specific requirements of the continuous development of the power system. At present, the adaptive relay protection’s importance is constantly increasing, the related professional of school and the related industry has made unremitting efforts in the study of power system and the adaptive neural network in the past 10 years.The neural network has good adaptability and learning ability, with which the neural network model can be built.With the method of putting the data that are collected in the daily work from the practical fault lines into the neural network model and training it, we can effectively solve the problem that the adaptive distance protection’s protection space changes with the actual situation of the operation mode of the system and the type of short circuit. We can also make accurate judgments of the boundary angle for distance protection quickly and efficiently. It provide a very good basis for the protection devices to remove the faults responsively and quickly. The capabilities of the protections are also improved significantly.We can predict the transition resistance of distance protection in the non grounding system effectively with building the platform by using the support vector machine algorithm, which can effectively decrease the protection maloperations and failure behaviors caused by the transition resistance. And the method of neural network can inspire and also make good reference for the zero sequence overcurrent protection. The Adaptive distance protection is obviously superior to the traditional distance protection system in selectivity and sensitivity, with witch the existing electric network can not only transmit more power, but also improve the flexibility and stability of the power system operation. |
PDF Full Text Request