Fast identification of internal and external faults of HVDC transmission line based on frequency spectrum correlation

Abstract

This paper presents a new method for rapid detection of internal and external faults on HVDC transmission line based on frequency spectrum (FS) correlation. Due to different physical boundaries consisted of DC filters, smoothing reactors and DC line distributed capacitance, the transient DC current signals of internal and external faults have different amplitude frequency characteristics, which constitute the intrinsic feature of its inherent. The FS characteristics of the DC current signals can be obtained via discrete Fourier analysis method. First, a set of eigenvector defined as eigen frequency spectrum eigenvector (EFSE) is formed including four types of intrinsic FS scenarios which are: normal operation, internal fault, external fault local side (EFLS) and external fault opposite side (EFOS). Second, the same method is utilized to obtain the FS of DC current of any fault, and the similarity coefficients between the FS and eigen frequency spectrum (EFS) are calculated. Thus, the fault type can be identified by comparing the magnitude of similarity coefficients of the FS and EFS. The fault type corresponding to the maximum similarity coefficient is the product of the judgment. Finally, a large number of fault simulations are performed on the PSCAD simulation platform and the effectiveness and robustness of the proposed method are verified.