LAUSR

Abstract The distribution system service restoration (SR) process implies to change the network topology configuration via a real-time maneuver plan, which in turn results in increased load current and changed short-circuit levels in some adjacent feeders. As a consequence, some protective devices (PD) can operate undesirably. In addition, some network branches may itself become totally unprotected. Therefore, it is essential to consider phase and neutral overcurrent protection in the SR problem, in order to maintain service continuity during emergency situations. A very efficiently approach, named MEAN (Multi-Objective Evolutionary Algorithm with NDE), for solving the SR problem is based on Multi-objective Evolutionary Algorithm (MOEA) that combines subpopulation tables with the Node-Depth Encoding (NDE) technique. This paper proposes to incorporate the phase and neutral overcurrent protection constraints into the MEAN approach for service restoration in radial distribution systems. The approach proposed in this paper, named MEAN-PC (MEAN with Protection Constraints), uses additional subpopulation tables to store the solutions that consider protection constraints. The objective is to show that the overcurrent protection constraints on service restoration problems can be considered in an effective way, correctly indicating feasible solutions of maneuver plan. The proposed overcurrent protection constraints in service restoration are tested and verified on the 135-bus test distribution system.