Algorithm for fault detection and localisation in a mesh-type bipolar DC microgrid network

This study proposes a new method to detect and locate the fault in a DC distribution system in the presence of distributed generators. As the mesh-type network provides good continuity of service during faults and the bipolar system transmits high power at low-voltage level to the end user, the proposed scheme has utilised 1200 V mesh-type bipolar DC microgrid network. The proposed scheme detects a fault by comparing the calculated value of fault current and pole-to-pole voltage with the pre-defined threshold values and then distinguishes internal or external fault based on an estimation of line parameters. In case of an internal fault, the estimated parameters yield the exact location of the fault. The modelling of the aforementioned DC microgrid network has been carried out in PSCAD/EMTDC environment and various faults involving pole and ground have been performed with variable fault resistance and location. The simulation results clearly indicate that the proposed technique is not only capable to detect the fault rapidly but also able to disconnect the faulty section speedily (around 1.25 ms). As the proposed method detects fault by utilising local voltages and currents, its reliability is better than the previous algorithms.