Improved Genetic Algorithm and Its Application in Electric Distribution Network Reconfiguration

The electric distribution network can effectively reduce the network loss by network reconfiguration. In this paper, an improved genetic algorithm is proposed for the structure of distribution networks. By linking the parameters of the genetic algorithm to the overall fitness of the population, a nonlinear crossover (p
c
) and variation (p
m
) probability adaptive scheme is proposed, which can not only ensure the convergence speed of the algorithm in the early stage, but also improve the efficiency of the later stage of the search for the best solution by different designs of pc and pm. To reduce the generation of infeasible solutions, a lion group crossover strategy is proposed to accelerate the convergence speed. A loop-based decimal coding strategy is also used in the paper to simplify the distribution network structure to reduce the chromosome length. Finally, the algorithm is verified to be feasible and effective by IEEE 33-node.