Network Reconfiguration With Distributed Generantion Based On GA-ACO Algorithm

Distribution network reconfiguration is an effective approach to improve the power system security and stability. The accession of DG makes the original single power supply network into a multiple power network. And then the installation position and capacity of DG make a lot of impacts on the node voltage, network loss and line flow distribution, which make the reconfiguration problem of distribution network become more complex. In this situation, the objective function, rules and constraints of distribution network reconfiguration, are no longer viable. In order to make sure the system securely and reliably, we need to make research on Distribution network reconfiguration with DG.In this paper, we firstly simplify the DG and use different power flow model to different DG. In simple distribution network, we demonstrate strictly and derivate the network influenced by different capacity of DG and the location of DG. Based on the different features of PV nodes, PI nodes and PQ (V) nodes, we construct the flow calculation models respectively and propose the Improved Layered Forward-backward Sweep Method to increases the calculation speed. We build three kinds of load model to simulate the output change of the load and DG. We make a network reconfiguration for each model and obtain three corresponding scheme to make the scheme closer to the reality.In this paper, chromosome constitution is based on the open switch in the circuit. The number of the chromosome is the same as the circuit. This method can greatly reduce the extent of the search space and improve search efficiency. Through the comparison of genetic algorithm and ant colony optimization, we propose a new algorithm for the combination of genetic algorithm and ant colony optimization. Firstly, we generate the distribution of pheromone by using fast global search ability of genetic algorithm and then we use the ant colony optimization to find the shortest path and the exact solution. According to the simulation of the IEEE-16nodes distribution power system with DG and IEEE-33nodes distribution power system with DG, we can prove the accuracy of the proposed algorithm.