A Method Based on Time-Domain Convolutional Network for Detecting and Classifying Traveling Wave with Hazard in Transmission Line

With the rapid development of society and economy, the power grid has put forward higher requirements for the reliability and stability of transmission lines. The existing transmission line fault location monitoring system can identify and locate lightning strike faults and non-lightning strike faults by traveling wave detection technology and signal extraction technology. However, in terms of hazard prediction, there is still a lack of algorithms with high maturity and accuracy, which is difficult to effectively prevent the occurrence of transmission line faults. Therefore, this paper proposes a hazard detection and classification method based on time-domain convolutional network to realize hazard detection and classification in transmission line. The experiments show that the method has high accuracy, can effectively detect and classify the waveforms with hazard, and improve the online safe operation level of transmission line.