A Bi-level Topology Similarity-based Optimal Reconfiguration Method for Distribution Network With Distributed Wind Power
Under the background of high penetration of wind power,optimal reconfiguration of the distribution network is an efficient way to address the issues of wind curtailment and loss of electricity.Multi-objective optimal reconfiguration is a non-convex nonlinear optimization problem containing many discrete variables,which is difficult to solve.Therefore,this paper establishes a bi-level topology similarity-based interactive reconfiguration method for a distribution network with high penetration of distributed wind power.A branch complex betweenness similarity evaluation index is designed to measure the similarity of all radial topologies for clustering similar networks and forming a tree-shaped solution space.The reconfiguration level optimizes the network topology with the states of the sectionalizing/tie switches as the decision variables which are transmitted to the scheduling level.The scheduling level carries out the coordinated active and reactive power optimal dispatch of the distribution network using the distributed wind power output,operation status of energy storage,power of reactive power compensation device,and demand response as decision variables.The optimized results of the scheduling level are fed back to the reconfiguration level,and the two levels continuously interact and iterate to come up with the final strategy.Furthermore,a pruning branch-harmony search algorithm is proposed to enhance the solution efficiency of the reconfiguration level network reconfiguration,in which the solution space can be switched from the branch to the intra-branch.The proposed method is benchmarked on the modified IEEE 33-node system,and the comparative results demonstrated its superiority in improving the system voltage security and reducing the system operation cost.
distribution networksdistributed wind powertopological similaritybi-level optimal reconfigurationsecond-order cone optimization
万方数据
维普
