The document analyzes the performance of a method for fault detection and localization in a distribution network using measurements from PMUs. Through simulations conducted on the Cigré medium-voltage benchmark network, the effectiveness of the algorithm was evaluated in different scenarios, considering variations in fault types (three-phase, two-phase, single-phase) and the state of neutral grounding (solid or compensated). Additionally, the effectiveness of the algorithm was assessed in scenarios with a limited number of measurement devices compared to the network nodes and considering an approximate modeling of noise introduced by the measurement chain.

PMU devices, originally developed for high-voltage networks, have been proposed by numerous studies and projects in recent years, including low-cost devices, as a viable solution for monitoring medium-voltage networks to improve observability in response to the spread of distributed generation. Therefore, the possibility of using these devices for network protection functions was investigated to help amortize installation and management costs.

The analyzed algorithm, considering PMUs located at every node of the network and assuming ideal measurements, effectively detects and localizes faults in each of the analyzed scenarios. However, when considering the modeling of measurement noise, some issues arose in detecting events in single-phase faults, with both solid and compensated neutral grounding. To evaluate the proposal under conditions closer to a possible real application, where not all network nodes might be monitored, scenarios were considered where PMUs were not placed at every node. The necessary PMU placement conditions for the correct functioning of the algorithm were analyzed. In this configuration, the algorithm has a lower resolution in fault localization but still allows for identifying the occurrence within a connected portion of the network (cluster). Challenges remain in detecting single-phase faults.

Overall, the results highlight the need to further develop and modify the fault detection and localization algorithm to improve its precision and reliability.