Study Of Identification Of Shielding Failure And Back Striking By Deep Learning Method

This paper proposes a strategy based on deep learning method for classifying two kinds of lightning: shielding failure and back striking. Firstly, the simulation model of complex lightning is built based on ATP-EMTP to collect the sample data for deep learning. And then the sample data is used to train a Deep Learning Network. Finally, the simulation and the actual observation of Jiangsu Lightning observation records are combined as a cross-validation set to test the Deep Learning Network. The results show that the proposed Deep Learning Network can achieve a much more higher recognition rate compared to the conventional methods such as BP Neural Networks and wavelet transform.