Transmission Line Fittings Defect Recognition Method Based on Knowledge Enhancement

In this paper, aiming at the problem that it is difficult to extract robust features in the existing technology and the low accuracy of defect recognition of connecting fittings caused by the lack of correlation between the features of connecting fittings, a defect recognition method of connecting fittings of transmission lines based on knowledge enhancement is proposed in combination with knowledge graph. On the one hand, GCN network is used to guide the model to identify defects by using the correlation information between nodes in the knowledge graph of connecting fittings. On the other hand, for the feature vectors of the input GCN network, in the training phase, the contrast learning method is used to compare the similarity of the feature vectors of the same label, so that the model can learn more robust feature knowledge. Through two knowledge enhancements, the accuracy of connection fittings defect recognition is effectively improved.