Optimal Configuration of the Radial Distribution Network Using Modified Genetic and Simulated Annealing Algorithms

Finding the optimal network configuration is one of the key problems and challenges in electric power systems. The optimal network configuration goal is to find the operating states of the switches to form a new network topology that will reduce active power losses while respecting operation constraints. To find the optimal configuration, different metaheuristic algorithms can be used. This paper presents a modified genetic algorithm and a simulated annealing algorithm adapted for the network configuration problem. Both algorithms were applied to a real 54-bus system. Different parameters of the algorithms were tested, as well as different initial network configurations. The simulated annealing algorithm adapted to the problem gave better results in terms of elapsed time. Programming and simulations were performed in the MATLAB environment using the Matpower library.