Research On Reconfiguration Of Distribution Network Containing Renewable Distributed Generation

The application of distributed generation(DG)in power grids is more and more extensive.The fluctuation and randomness of DG output make the operation scene of distribution networks more complex and changeable,and increase the technical and economic risks of the system operation.Distribution network reconfiguration can optimize the operation of distribution networks by changing the topology of distribution networks through switching operation without a lot of hardware investment.DG accessing to the grid makes the existing distribution network reconfiguration technology faced with some challenges.This thesis studies the distribution network reconfiguration from two perspectives of planning stage and operation stage,aiming at three problems of DG: multi-scene,fluctuation and uncertainty.In the planning stage,an improved adaptive fuzzy C-means clustering algorithm is adopted to extract typical operation scenarios of distribution networks,effectively reducing the number of scenarios.A distribution network reconfiguration model based on typical operation scenarios is built,which takes the minimum network loss as the objective function,includes the operation constraints of distribution networks,and optimizes the topology of distribution networks.The reconfiguration scheme can ensure the optimal operation of distribution networks under complex scenarios in a long period.In the operation stage,in view of the voltage fluctuation caused by the DG output fluctuation,the voltage fluctuation index is defined from the perspective of the influence of the distribution network topology on the voltage fluctuation.The multi-objective reconstruction model is established with the active power loss and voltage fluctuation indicators as the objective function,and the optimal compromise solution is obtained from the Pareto solution set using the technique for order preference by similarity to ideal solution(TOPSIS).The reconstruction scheme obtained by the solution of the model can effectively reduce the network loss and suppress the voltage fluctuation,taking into account the safety and economy of distribution network operation.In view of the challenges brought by the uncertainty of DG output to the traditional distribution network reconfiguration in the operation stage,the intelligent soft open point(SOP)is introduced in the distribution network reconfiguration to carry out real-time refined control of the distribution network,and a two-layer reconfiguration model of the distribution network with SOP was established considering the risk of prediction errors.The upper-level model jointly optimizes traditional switches and SOPs based on dayahead forecast data,and uses Conditional Value at Risk(CVa R)to model uncertainty;the lower-level model redesigns the operating strategy of SOP based on more accurate intraday forecast data to further optimize the operation of the distribution network by the fast and real-time response capability of SOP.The effectiveness of the proposed model is verified through numerical examples,the superiority of the uncertainty reconstruction model compared with the deterministic reconstruction model is compared and analyzed,and the optimization effect of SOP under different optimization time scales is analyzed.