•  
  •  
 

Keywords

transient voltage; fault diagnosis; time-frequency analysis; continuous wavelet transform; BP neuralnetwork

Abstract

To address the misoperation and failure issues of existing transient protection schemes for high-voltage transmission lines, this paper proposes a transmission line protection method that combines a back propagation (BP) neural network with the time-frequency matrix of transient voltage signals. Transient protection of transmission lines is implemented based on the time-frequency analysis matrix obtained through a one-dimensional continuous wavelet transform of fault-induced voltage traveling waves. According to the existing fault voltage traveling wave data or simulated fault voltage traveling wave data, the time-frequency matrix is obtained by time-frequency analysis. The part with obvious time-frequency characteristics is taken as the input of the BP neural network, and the fault condition is taken as the output. Through neural network learning, reliable discrimination of internal and external faults within the high-voltage transmission line protection zone is realized, and rapid protection of high-voltage transmission lines is achieved. Simulation results demonstrate that the proposed method comprehensively utilizes fault characteristics of transient waveforms in both the time and frequency domains while maintaining a high computational efficiency. It is expected to improve the reliability of transient protection of high-voltage transmission lines.

DOI

10.19781/j.issn.1673-9140.2026.01.005

First Page

46

Last Page

52

Download

Share

COinS