Fast Heuristic Based Distribution Network Reconfiguration with Distributed Generations

This paper introduces a two-stage heuristic-based approach for distribution network reconfiguration (DNR), aiming to optimize the system's performance with the incorporation of distributed generation (DG) with much reduced computational time. In the initial reconfiguration phase, it is presumed that every switch is closed. A sequential switch-opening technique is employed to achieve a nearly ideal radial arrangement. In the second step, two switches are chosen at random from the radial network and their on/off status is exchanged. In every exchange transaction, a pair of switches is chosen to minimize power loss in the network using a mathematical model that links bus injections to branch currents. After feeder reconfiguration the Genetic Algorithm (GA) is implemented to optimize distributed generation (DG) to minimize power losses in the network. The IEEE 33-bus radial distribution network (RDN) is used to test the proposed technique.