Research On Coordinated Optimization Strategy Of Fault Recovery And Emergency Repair For Distribution Network With Distributed Generations

In recent years,with the depletion of traditional fossil fuels and the increasingly serious environmental problems,distributed generation(DG),represented by photovoltaic power generation and wind power generation,has developed rapidly.Due to the large-scale access of DG to distribution network,the operation mode of traditional distribution network has changed tremendously.How to quickly recover the power loss area when the distribution network with DG fails and formulate the emergency repair strategy after failure have become a research hotspot.This paper makes an in-depth study on the power flow calculation of distribution network with DG,fault recovery considering the demand of user side,and the coordinated optimization strategy of fault recovery and emergency repair.The main research results are as follows:After a large number of DGs are connected to the distribution network,the topology of distribution network is more complex,and the traditional power flow calculation method is no longer suitable for distribution network with DG.Therefore,the topology search method of distribution network with DG is studied.Based on the basic principle of graph theory,the breadth first search method and depth first search method are introduced.Then,the typical DGs are introduced,including wind power generation,photovoltaic power generation,micro gas turbines,and fuel cells in power generation devices.At the same time,the energy storage devices are also introduced.Finally,the power flow calculation method of distribution network with DG is studied.According to the different interface modes between DG and power grid,the nodes are classified into PQ nodes,PV nodes,PI nodes and P-Q(V)nodes.The power flow calculation method with different types of nodes is researched and the traditional backward/forward sweep algorithm is improved.Considering the improvement of users’ requirements for the reliability of the distribution network,the fault recovery strategy of the distribution network that takes into account the demand of user-side is studied.In order to improve the power grid reliability,the time-varying demand of user-side is considered in the process of distribution network fault recovery,the typical load time-varying demand model is established,and the controllable model load is introduced.In the process of fault recovery,the load with high time-varying demand of user-side is given priority to restore power supply.In the study of the island partition scheme,the DG is used to form the island to restore power to the power-loss load.In order to further improve the efficiency of distribution network fault recovery,the island formed by DG and the main network of distribution network jointly recover the fault.And the overall objective function is to take the minimum total power-loss load,the minimum power loss and the minimum number of switching actions as the comprehensive objective function.In order to improve the efficiency of the algorithm,the traditional particle swarm optimization(PSO)algorithm is improved to form the mutation particle swarm optimization(MPSO)algorithm,which is applied to the optimization process of fault recovery strategy.Finally,the IEEE 33 example is analyzed to verify the superiority of the research strategy.In view of the problem that the DG cannot support the stable operation of the load for a long time,it is necessary to complete the emergency repair of the fault area in time after the fault occurs.The fault repair model is established with the shortest fault repair time and the minimum social and economic loss as the comprehensive objective function,and the improved PSO is used to obtain the optimal repair sequence.In the research of fault recovery and emergency repair strategy,considering the time-varying demand changes of the load,the optimal plan for fault recovery and emergency repair is dynamically adjusted to ensure the reliability and rapid recovery of the distribution network to normal operation.Finally,an example is given to verify the effectiveness of the proposed method.