Intelligent Identification of Transmission Line Fault Types Based on Wavelet Energy Spectrum Entropy

In order to solve the problems of poor accuracy of traditional transmission line fault identification, which is easily affected by the performance of phase-selecting components and fault parameters, a fault type identification method based on wavelet energy spectrum entropy and particle swarm optimization support vector machine (PSO-SVM) is proposed. Considering that the related research is mostly based on simulation data, the multi-resolution analysis of fast wavelet transform and information entropy is used to analyze and process the fault recording data of a power grid company in recent years, extract the entropy value of the wavelet spectrum of faulty three-phase currents, and introduce the zero-sequence currents to form the eigenvectors, which are inputted into the SVM model, and at the same time, the PSO algorithm is utilized to obtain the optimal SVM parameters. For the imbalance of fault types in the actual data, the actual line parameter model is simulated to obtain a large number of fault samples, and finally, the model is trained and tested with the actual and simulated data, respectively. The results show that the method can accurately identify fault types and is not easily affected by fault randomness and parameter variability.