Distribution Network Reconfiguration With Distributed Generation

With the development of smart grid, as an important part of grid, the distribution generation has attracted more and more attention. The distribution generation is used as a supplement for traditional power supply mode. In recent years, DG’s proportion in power system increases gradually. After connected with grid, the distribution generation will influence the planning, operation, control and protection of power system. Among them, the influence on distribution network reconfiguration is an important aspect.Distribution network reconfiguration is a control method of realizing the distribution network optimal operation. To optimize the objective function, with the change of switches state, it changes the network topology to affect the power flow on the premise of satisfying the constraints. When distribution generation is connected to distribution network, the magnitude and direction of distribution network system is changed, which also changes the steady-state voltage of distribution network. Besides, the start-stop and output change of distribution generation can cause the distribution network voltage fluctuate and even exceed the limits, which makes distribution network reconfiguration become more complex.Firstly, the current situation, various algorithms and models of distribution network is introduced in the paper. Then, this paper classifies distribution generation and builds models to make it applicable to power flow calculation. Subsequently, the ant colony algorithm for distribution network reconfiguration is described. However, the traditional ant colony algorithm is easy to fall into local optimum and converges slowly. To solve the problems above, this paper improves the pheromone update link of ant colony algorithm. Besides, to better combine the inspiration factor which is distribution network certain factor with the actual planning, this paper also improves the inspiration factor. For ensuring the feasibility of distribution network reconfiguration results and the radiation structure, the minimum spanning tree is applied in the search process of ant colony in this paper.In the end of the paper, with the IEEE33nodes distribution network as test system, the improved ant colony algorithm is testified. The results show that the improved algorithm can avoid local optimum and get the global optimal solution quickly. Then, distribution generation is connected with distribution network, and the network is reconfigured. It can be concluded that the active power loss of improved network deceases and voltage quality of all nodes increase.