A Fault Location Scheme for Active Untransposed Distribution Systems Using a Limited Number of Synchronized Measurements

Abstract

This paper presents a fault location scheme for unbalanced and untransposed distribution systems which contain different types of distributed generation (DG). A general formula has been derived for any fault type using a limited number of synchronized measurement points. To avoid having to synchronize all of the measurement points, the voltage and current measured locally at the DGs are processed locally to calculate the equivalent impedance of the DG at the non-fundamental frequencies. This is then used in the fault location process. The IEEE 34-bus feeder is simulated using the distributed parameter model for the lines and is used to validate the proposed scheme. Uncertainties associated with fault type, fault location, fault resistance, inception angle, noise in measurements and load profile are considered in the evaluation. The simulation studies demonstrate that the scheme can have a high accuracy and is robust.

Deep Learning-Based Fault Classification and Location for Underground Power Cable of Nuclear Facilities

“…An advanced signal processing technique called the Stockwell Transform (ST) that combines the advantages of wavelet transform and short-time Fourier Transform (FT) was employed recently to detect and classify faults in distribution grids in [22], but not employed in locating faults and identifying faulty sections in distribution grids. A wideband fault location scheme for distribution systems in distributed generation of different types has been presented and evaluated in [23]. The proposed scheme analyses the system using the high frequency non-fundamental components calculated from the available synchronized and nonsynchronized measurements.…”