Research on distribution network reconfiguration

This article proposed the affect of distribution network reconfiguration in the power grid, summaries a variety of methods to solve distribution network reconfiguration problem and analyzed features of these methods applied to distribution network reconfiguration. The solution worked out using the algorithm based on optimal flow pattern may not be optimal or near optimal. But the algorithm combined with heuristic rules can quickly obtain satisfactory results. Mathematical optimal approaches are not suitable for solving this problem. Algorithms based on simulated annealing can work out optimal solution, however, they are very time-consuming and their performances depend on initial parameters. The algorithms based on artificial neural network can find out optimal distribution configuration quickly, however, their results depend on the training sets and the total training sets are very difficulty to obtain, further the training set is very time-consuming. Genetic algorithms are suitable for solving distribution network reconfiguration problem, and its wide applications to solve this problem will be expected if its convergence is improved.