Performance of the simulated annealing-based algorithms for the optimal reconfiguration of distribution systems

The optimal reconfiguration of a distribution network is a constrained combinatorial problem where the open branches and the state of the switches are identified in order to maintain the radial configuration of the network and to minimise a specified objective function. Simulated annealing-based algorithms have been proposed as a useful technique for computing near-optimal solutions. The performance of these algorithms depends on the choice of several parameters. This paper deals with the effects of the parameter choice on the computation time and on the effectiveness of the solutions. Results of extensive tests performed on real urban distribution systems are presented and discussed