Research On Dynamic Service Restoration Strategy For Active Distribution Network

Self-healing is an important feature of the smart grid.The rapid fault restoration in the distribution network is one of the important means to achieve self-healing control of smart distribution networks.The maturity of renewable energy generation technologies such as photovoltaic and wind power and the increase in demand response have transformed the traditional distribution network to the active distribution network,which will inevitably affect the fault restoration mode and reliability of the distribution network.The distributed generations and the energy storage systems can quickly restore the power supply of some important loads through the islands after a fault occurs in the active distribution network,However,some problems such as the unbalanced supply and demand of islands caused by the time-varying of clean energy and load power,underutilized the regulatory potential of controlled loads,coordination of island formation and tie line transfer process have not been solved perfectly.So the traditional static fault restoration scheme is obviously no longer applicable.It is necessary to study the dynamic fault restoration strategy that adapts to the active distribution network.And the reliability of strategies should be verified to quantify the level of reliability improvement.Based on this,this paper makes the following research:(1)Theoretical introduction and model building related to recovery strategy.A brief introduction to the basic knowledge of graph theory and common network traversal search methods.Analyze the topology of the distribution network on the basis of graph theory.And then this paper briefly introduces the forward/backward sweep method power flow algorithm.Finally,a time-varying model of the source-load including wind turbine,photovoltaic,energy storage and loads was established.(2)Based on the time variation of source and load,a dynamic island partitioning strategy for active distribution network is proposed.Backtracking is considered the main idea of the stratety,and the maximum value of the equivalent load recovery is taken as the objective function.The strategy can obtain the range of islands in each period under the dual role of DG and energy storage.Simulation with IEEE33 node system shows that the proposed strategy can effectively improve the normal power supply level of important loads.(3)In order to give full play to the scheduling potential of controllable load,an island-wide optimization strategy considering the controllable load regulation capability is proposed.Establish a mathematical model of controllable load and propose a controllable load contribution index.The harmonic algorithm is used to optimize the optimal adjustment of each controllable load in the original islands in each period,and then the optimal range of islanding is obtained.The simulation of IEEE33 node system shows that the proposed strategy can effectively improve the recovery of island load and reduce the overall cost loss.(4)An integrated dynamic fault restoration strategy for active distribution network combining tie line transfer and island partition is proposed.Taking the overall equivalent recovery as the objective function.The genetic algorithm is used to optimize the switch state,and then the main network recovery power supply range and the DG recovery power supply range corresponding to each time period are obtained.Finally,the IEEE33 node system simulation verifies the rationality and superiority of the proposed strategy.(5)Establish an active distribution network reliability assessment model.The fault consequence mode analysis,component state sampling and reliability index are established in turn.Make an evaluation process based on the sequential Monte Carlo method.Analyze the reliability of the recovery strategy,and compare the impact of different factors on reliability indicators.Simulation results show the reliability level of different strategies.This chapter provides a basis for the formulation of the recovery plan,and provides a reference for the resource allocation of the distribution network.