Fault Location of Hybrid Three-terminal HVDC Transmission Line Based on Neural Network

The fault transient process of HVDC transmission line has stronger natural frequency signal. Therefore, this paper extracts the amplitude and frequency of natural frequency as sample attributes, and proposes a neural network based hybrid multi-terminal HVDC transmission line fault location algorithm. For the traditional-flexible hybrid three-terminal transmission line, the fault branches are different, and the components, amplitude and phase of the natural frequency are also different. Firstly, the branch line where the fault occurs is determined according to the magnitude of the dominant frequency in the natural frequency spectrum, and then the faulted ranging is performed by using the hierarchical distributed neural network. The particle swarm optimization algorithm is used to optimize the weight and threshold of BP neural network, and the network training efficiency is improved, and the convergence speed is accelerated. The simulation results of PSCAD and MATLAB show that the fault location algorithm has high reliability and accuracy.