Mathematical Morphology-Based Fault Identification for Distribution Network with DFIG

With the large-scale access to the distribution grid by new energy power sources, the short-circuit currents provided by doubly-fed induction generator (DFIG) increase the transient complexity of the grid. This is due to their low-voltage ride-through (LVRT) requirements in case of faults on the distribution network. Consequently, traditional power system relay protection schemes are impacted. Therefore, a multi-resolution morphological gradient-based fault identification method for distribution network is proposed. This method focuses on the short-circuit currents characteristics of DFIG. It aims to achieve accurate fault identification when DFIG is connected to the distribution network. A appropriate data window is selected and then the multi-resolution morphological gradient (MMG) is processed to obtain the morphological gradient characteristics of the currents signal. After analyzing the peak-to-peak ratio of gradient characteristics, a criterion is constructed which is used to correctly classify different types of faults and select the fault phase. A distribution network model with DFIG is constructed in MATLAB/Simulink, and the effectiveness of the proposed fault identification method is verified.