Intelligent Fault Classification Solution for Networks Connected with Unconventional Sources

The increasing penetration of inverter-based renewable energy sources (unconventional sources) in transmission networks introduces significant challenges for fault classification methods due to their unique fault response characterized by limited and control-modulated fault currents. Fault response could also vary from grid to grid depending on prevailing grid codes related to integration of unconventional sources. To address these challenges, an intelligent fault classification solution has been developed that ensures reliable fault classification under diverse grid conditions. Based on grid short circuit strength and source type, the proposed solution intelligently selects the correct fault classification method such as modified voltage/current angle or only magnitude-based method. Fault is classified by comparing the ratios of negative to zero-sequence and negative to pure fault positive-sequence voltage/current angles with predefined fault zones. Additionally, changes in apparent power are monitored alongside angle ratios to enhance reliability. To mitigate frequency deviations due to inverter controls, a method based solely on apparent power magnitude is also incorporated. As a final solution we combined all these methods which exploit the combined advantages of the methods to classify the faults reliably and the solution is resilient against variations in network configurations, grounding types and inverter-based resources control strategies. The solution has been extensively validated on high-voltage (HV) and medium-voltage (MV) networks connected to different types of sources including renewables. Simulation studies and real-time testing confirm high accuracy in fault classification under varying operating conditions.