Research On Network Reconfiguration Of Active Distribution Network With Distributed Generations And Electric Vehicle

In recent years,with the rapid growth of energy demand,Distributed Generation(DG)and Electric Vehicle(EV)have attracted people’s attention due to their characteristics of renewable,economic and environmental protection.The randomness and dispersion of DG and EV loads,after large-scale access to the distribution network,pose a challenge to the safe,reliable and stable operation of the distribution network.The distribution network reconfiguration changes the topology of the distribution network by operating the closed state of the switches in the network to improve the economic reliability of the distribution network and to make the distribution network in an optimal operating state.In order to improve the economics and stability of the distribution network operation after DG and EV access to the network,this paper mainly combines the characteristics of DG and EV access to the distribution network to carry out the research on active distribution network reconfiguration based on artificial fish swarm algorithm.(1)Establish the power output model of wind power,photovoltaic and other DGs,and transform the node types of DG in the distribution network into PQ nodes,and propose improved utilization for the deterministic and uncertain factors of the active power flow calculation process.The pre-pushing back method and the two-point estimation method are used to calculate the deterministic and probabilistic power flow of the distribution network to improve the accuracy and efficiency of the power flow calculation.(2)Aiming at the low precision of traditional artificial fish swarm algorithm,it is easy to fall into local optimal value in the later stage,and can not adapt well to the defects of active distribution networkreconstruction.An improved artificial fish swarm algorithm based on differential evolution(DE-IAFSA)is proposed.By comparing the test results,the DE-IAFSA is compared with the traditional fish swarm algorithm.The results show that DE-IAFSA has better resolution and convergence speed.(3)Study the static reconstruction of active distribution network with DG,such as wind and light.The overall optimization is aimed at reducing network loss and load balance,and a static reconstruction model is established.In order to reduce the infeasible solution,a "no-repetition" spanning tree strategy is proposed.The IEEE33 node system is used as the simulation test system,and the DE-IAFSA algorithm is used to statically reconstruct the active distribution network.The reconstruction results show that DE-IAFSA The algorithm can achieve the expected effect in the optimization of distribution network reconstruction.The access of DG can also reduce the network loss of the system to a certain extent,improve the node voltage level and improve the power quality.(4)The study considers DG output and EV charging load uncertainty,and dynamic reconstruction of active distribution network based on DE-IAFSA.Firstly,the mathematical model of EV charging load is established.The Monte Carlo method is used to simulate the EV daily charging load,and the DG output such as wind and light and the original load of the distribution network are superimposed to form an equivalent load.An equivalent load period division strategy based on information entropy subtraction is proposed.The equivalent load is divided into multiple time periods.Finally,the DE-IAFSA algorithm is used to simulate the simulation to verify the reasonable validity of the time division strategy based on information entropy subtraction.