Dynamic reconfiguration of distribution network considering scheduling of DG active power outputs

Considering the impacts on distribution network with large-scale distributed generation (DG) integration, a model is established to serve as a method of dynamic reconfiguration of distribution network considering scheduling of controllable DG active power outputs. Day-ahead outputs of uncontrollable DG are predicted based on Markov chain model with historical data. The load-DG equivalent load curve is built in terms of outputs prediction results and then the equivalent load curve is divided by fuzzy clustering analysis. A dynamic reconfiguration model considering scheduling of controllable DG active power outputs is established based on bi-level programming model. The minimum daily running costs and switch states are taken as objective function and optimization variables respectively in the upper level planning. The minimum running costs of each time interval and controllable DG outputs of each hour are taken as objective function and optimization variables respectively in the lower level planning. Based on the strategy of spanning tree and elitist, the developed model is solved by the hybrid algorithm combined by ant colony algorithm and genetic algorithm. The simulation results of the example show the method mentioned in this paper can effectively gain profitability of electric power companies.