Research On Distribution Network Dynamic Reconfiguration Considering Distributed Generation

As a form of new energy power generation,distributed generation(DG)is widely used because of its advantages of environmental protection,flexibility and efficiency,which brings a lot of convenience to the grid.At the same time,DG integration inevitably brings the complexity and uncertainty to the grid.In this thesis,the dynamic reconfiguration model and method of distribution networks with DGs are studied.The main tasks are as follows:(1)The dynamic reconfiguration model of the distribution network that can restrict the continuous multiple actions of the switches is proposed.Considering the fact rule that distribution network switches can't be continuously operated in a short period of time,the operation cost function of the switches is processed referring to the bathtub curve model.Then the operation cost function is introduced into the dynamic reconfiguration model so that our model can limit the short-time continuous actions of the switches.(2)An improved binary particle swarm optimization algorithm is proposed.The evolution mechanism of differential evolution algorithm is introduced into the speed updating mechanism of binary particle swarm.At the same time,the crossover and mutation operation of genetic algorithm is introduced to enhance the algorithm global optimization ability and the ability to get out of local optimal solutions.Based on the broken coil method,a new mechanism of location updating is established to guarantee the feasible solution in the iterative process and improve the search efficiency.The effectiveness of the proposed intelligent algorithm is verified by simulation.(3)The uncertainty of distribution generation is dealt based on scene method.Using the Latin hypercube sampling mechanism and error normal distribution function,the prediction error of wind energy is sampled and random scenes are generated to transform the uncertainty problem into multi-scene deterministic problems.(4)The time division method is propose based on voltage fluctuation deviation.According to the overall load predictive value,the power flow simulation of the distribution network is carried out,and the voltage fluctuation deviation index is calculated according to the voltage value of each node in the period of simulation.Based on the magnitude and timing distribution of the voltage fluctuation deviation indexes,the result of the time division is presented intuitively.The feasibility and validity of proposed method are verified by the case analysis of IEEE33 node distribution system with distributed generations.