Low-current fault detection in high impedance grounded distribution networks, using residual variations of asymmetries

This paper presents a new method of detecting low-current faults in high impedance grounded distribution networks, in both compensated and ungrounded medium voltage networks. Based on monitoring the residual variations of line and phase asymmetries, the proposed algorithm can be used in combination with other steady-state data-based methodologies, in order to detect the existence of a single-phase ground fault, as well as identify the faulty feeder and phase. Although it could also be used in compensated networks by means of other techniques (such as the injection of zero-sequence currents in the neutral), the algorithm described in this paper has been developed by using superposed voltages, so as to make its application possible in compensated and ungrounded distribution networks. Furthermore, the reliability of this algorithm has been extensively tested in Matlab/Simulink, considering a wide range of resistive faults, and its practical applicability has also been described.