Multi-Terminal Transmission Line Fault Localization Based on 1dCNN-Transformer

Aiming at the problem of difficult extraction of fault traveling-wave front of transmission line faults, this paper proposes a fault traveling-wave front extraction method based on the combination of Variational Mode Decomposition (VMD) and Teager Energy Operator (TEO). First, the fault current traveling wave signal is decoupled by the Karenbauer transform to obtain the α-mode component, and then the VMD decomposition is implemented on this component, and the decomposed IMF3 component is selected, and the TEO energy operator is applied to determine the arrival time of the head. Aiming at the problem of missing detection point data in multi-terminal transmission line fault localization, a method combining 1dCNN-Transformer neural network and double-ended traveling wave ranging formula is proposed. The fault feature information is input into the 1dCNN-Transformer to identify the fault section, and then the distance of the fault point is calculated according to the arrival time of the wave head and using the double-ended traveling wave ranging formula. A 750 kV multi-terminal transmission line fault simulation model is constructed using the PSCAD/EMTDC platform, and the simulation results prove that the proposed method is still able to identify fault zones in the case of missing data at detection points, and the error of the fault ranging results is within 100 m.