Fault-Tolerant Control of Power Distribution Networks Using Adaptive Protection Schemes

Abstract

Power flow reversals, bi-directional energy exchange, and a broader range of fault conditions are manifestations of unprecedented operational complexities brought by the rapid transformation of the modern power distribution networks, which was fuelled by the spread of distributed energy resources (DERs), renewable power production, and increased popularization of electric vehicles. Types of conventional protection systems that are often designed using fixed relay setting and centralized coordination and thus show slower fault isolation, low flexibility and increased vulnerability to cascading failures are becoming inefficient in these changing environments. This paper proposes a new Fault-Tolerant Control (FTC) algorithm that integrates Adaptive Protection Schemes (APS) to ensure these challenges can be overcome and offer high resiliency and reliability of active distribution networks. The three main capabilities proposed are the functions (i) of real-time fault classification and detection through advanced signal processing algorithms and machine learning techniques to identify exactly where a fault occurs and/or what type of fault it is in a precise and fast manner; (ii) optimal relay coordination in which the given relay settings adapt to achieve selectivity, sensitivity, and speed under different topology and penetrations of DERs; and (iii) self-healing network reconfiguration, which exploits graph algorithms to restore supply to non-affected areas, limit interruption of services, and equilibrate network loads after Because of the detailed simulations with complete IEEE 33-bus and IEEE 69-bus radial distribution test systems under several operating conditions with different DER penetration levels and fault conditions, the FTC-APS model has been tested. As results show, the proposed system provides up to 40 55% faster fault isolation, up to 25 faster outage impact, as well as an exceptionally high value of the System Resilience Index as compared to a conventional, static protection scheme. The results validate the promise of adaptive intelligence-based protection and control systems to enable the shift to smarter, resilient distribution grids that can withstand and recover to fault events in seconds.