A Deep Learning Based Method for Verifying Relay Protection Settings of Transmission Lines

In order to optimize the relay protection strategy and improve the standardization of power protection, the design and research of the relay protection setting verification method for transmission lines are carried out based on the application of deep learning algorithms. Design a multi-source data fusion processing framework that integrates power grid topology, real-time operation mode, and historical fault waveforms to construct feature inputs for network models. Construct a bidirectional long short-term memory network model with attention mechanism to deeply explore the complex nonlinear mapping relationship between electrical quantity temporal characteristics and protection settings. Based on deep learning models, design a fixed value intelligent verification system to achieve rapid batch verification of existing fixed values. The comparative experimental results show that the design method can ensure the accuracy of transmission line relay protection and avoid the occurrence of misoperation.