In order to solve the problems of long classification time and low accuracy of traditional voltage sag source identification methods, on the basis of CNNs, the author conducts an in-depth study on the identification of voltage sag sources in the electrified railway power supply systems. Firstly, study and research from three methods of line fault analysis, simulation verification method, and power quality analysis method are conducted. In order to improve the classification and identification of compound-voltage sag sources when the distribution network contains harmonics, the author proposes a new method for the classification and identification of compound-voltage sag sources based on the eigenvalue synthesis method. Among them, according to the different voltage sag waveform characteristics caused by different composite voltage sag sources, the three-phase voltage unbalance is defined, and the result is obtained by combining the compound-voltage sag source including single-phase grounding with the voltage sag source compounded by induction motor startup and transformer input. After experiments and research, the proposed method is verified by simulation experiments, the results show that the method can classify and identify the types and fault sequences of the compound-voltage sag sources well, and the identification accuracy rate is higher than 96%.