This paper reports the soft sifting stopping criterion-empirical mode decomposition(SSSC-EMD)and error feedback neural network(BPNN)based on soft sifting stopping criterion-empirical mode decomposition(SSSC-EMD)and error feedback transmission type neural network(BPNN)for transmission line ground fault methods for identification,By building a transmission line model in PSCAD/EM'TDC,setting up ground faults,and importing the fault signals into MATLAB after phase-mode decoupling,single-phase ground faults and two-phase ground faults are identified and compared using the $ SSC-EMD+BPNN and EEMD+BPNN methods respectively.Simulation results show that 3 out of 10 groups of data identified by the EEMD+BPNN method for single-phase ground faults are identified as two-phase ground faults,and the identification of 10 groups of two-phase ground faults is completely correct,and the accuracy rate of fault identification is 85％:And the SSSC-EMD+BPNN method identifies both single-phase ground faults and two-phase ground faults correctly,and the fault identification accuracy rate reaches 100％.
万方数据
维普
