In view of such problems as complex network topology structure due to access to the distribution network of multiple types of distributed power supply

the traditional genetic algorithm has the problem of big optimization search range and easy convergence by local optimization. In this paper

a kind of distribution network reconfiguration algorithm based on improved bat algorithm is proposed. The algorithm uses the back branch correlation dynamic matrix to cut and merge the network loop

which simplifies the topology search while ensuring the radial topology of the distribution network. The K-means clustering is used to initialize the population data to solve the local optimum. In addition

the Powell local search is used to accelerate the convergence rate of the algorithm. It is shown by the simulation experiment that the improved algorithm

compared to other standard bat algorithm and particle swarm optimization algorithm

has faster convergence speed and higher searching accuracy. In the reconstruction of distribution network

the improved algorithm proposed in this paper

with comparison to other algorithms

not only reduces effectively the active power loss of original distribution network

but also improves the amplitude of node voltage and

to a certain extent

improves the voltage quality of the system.