Faulty feeder selection based on improved Hough transform in resonant grounded distribution networks

Abstract

Due to the weak fault characteristics and inaccurate fault line selection during high impedance faults, this paper proposes a new method for fault line selection in resonant grounded distribution networks based on improved Hough transform. Firstly, the variational modal decomposition (VMD) algorithm decomposes each feeder's zero sequence power under different operating conditions, then, using Sample entropy and Pearson correlation coefficient, selects the intrinsic mode function (IMF) with the highest correlation strength with the zero-sequence power of each feeder, called IMFK. Secondly, using the Hough transform to detect IMFK, obtaining the fitted straight line and IMFK angle during the fault's initial stage. The problem of identifying the zero-sequence power mode trend is transformed into a mathematical problem of determining the geometric shapes of straight lines and curves in the zero-sequence power mode image. Finally, a line selection criterion is established by integrating correlation coefficients to identify the faulty feeder in the distribution network system when a single-phase grounding fault occurs. The PSCAD simulation and on-site measurement results show that the proposed method can ensure the effectiveness and adaptability of the line selection.