Transformer differential protection algorithm based on accelerated convolutional neural network
Aiming at the problem of false tripping due to excitation surge current in transformer differential protection,an al-gorithm based on accelerated convolutional neural network is proposed.A neural network is used to distinguish between inter-nal fault current and surge current,the compression of fully connected and convolutional layers is applied,and the modified linear cell activation function and batch normalization technique are integrated.The 220 kV transformer differential protection model is established with PSCAD/EMTDC software and the algorithm is applied,which verifies that the algorithm is faster and more reliable.
万方数据
维普
