Абстрактный

Improved Transformer Fault Classification Using ANN and Differential Method

Geeva Varghese, Amey George

During differential protection in transformers if false tripping occurs without existence of any fault then continuity of supply is affected. So false tripping conditions has to be avoided. This requires accurate fault classification. This paper presents improved fault classification for differential protection in transformers. This fault classification is done by using Artificial Neural Network (ANN).Here we use Back Propagation Neural Network (BPNN) and Radial Basis Neural Network (RBNN).The Neural Network is trained for classifying fault for differential protection in transformers. Different fault conditions are taken. The output of neural network will be a tripping signal in case of fault condition. The proposed method is simulated using MATLAB neural network tool and simulink package