High-voltage transmission line fault identification method and device for realizing method

The invention discloses a high-voltage transmission line fault identification method, and the method comprises the steps: firstly, carrying out the learning of two-dimensional time series data through employing a CNN model, and employing the powerful feature extraction capability of the CNN model, thereby solving a problem that a feature extraction method is too complex; secondly, a segmentation identification method is provided, that is to say, aiming at the problems that the high-voltage line phase-to-phase fault identification accuracy is low and the high-voltage line phase-to-phase fault identification is difficult to distinguish, the high-voltage transmission line fault is effectively identified by judging whether the phase-to-phase fault is grounded through the phase-to-phase current difference and the negative sequence and zero sequence components of the non-fault phase and by using the SVM on the basis of adopting the CNN to preliminarily judge the fault type. Compared with the ...