Research On Architectural Reorganization Of AC/DC Hybrid Microgrid Based On NSGA-?

Microgrid includes distributed generation units such as photovoltaic and wind turbines,which can make full use of renewable energy.At present,the topology of microgrid structure will not change during stable operation.However,the distributed generation units are vulnerable to environmental changes,resulting in large changes in output power.Meanwhile,the load of the microgrid 1s also prone to fluctuations,which makes it difficult for the system to achieve optimal operation status.When the microgrid fails,the traditional fault recovery method takes too long to recover.In order to solve this problem,according to the output of distributed generation units,the architecture restructuring of AC/DC hybrid microgrid is studied in this thesis.By dynamically adjusting the structure of microgrid,not only the optimal mode of energy management can be achieved,but also the problem of fault recovery in the operation of microgrid can be well solved.In this thesis,the AC/DC hybrid microgrid is taken as the research object.The BP neural network algorithm 1s used to predict the ultra-short-term load and the output power of PV and wind turbine of the microgrid.According to the prediction results,Non-dominated Sorting Genetic Algorithms-?(NSGA-?)with elite strategy 1s selected to reconstruct the structure of AC/DC hybrid microgrid.The optimization and reorganization of the microgrid structure in the normal operation of the microgrid and the failure reorganization of the microgrid are analyzed respectively.In the aspect of optimization and reorganization of licrogrid,firstly,the simulation model of microgrid 1s built on the platform of MATLAB/Simulink.In order to ensure the normal operation of the microgrid model,experiments are carried out to verify the operation of the microgrid under two modes of interconnection and disconnection.Secondly,the objective function and constraints of microgrid optimization and restructuring are determined,and the optimal results are solved by NSGA-? algorithm.Finally,with the help of Simulink software platform,the microgrid structure optimization and restructuring experiments are carried out to verify its correctness.In microgrid fault reconfiguration,firstly,the objective function and constraints are determined.Secondly?load shedding is determined according to the output of the distributed generation unit.If the total output meets the supply of all loads,no load shedding is needed.If the total output can not meet the supply of all loads,load shedding processing is needed.For these two cases,the optimal results of microgrid fault reconfiguration are solved by NSGA-II algorithm.Finally,an example is given to verify the correctness of microgrid fault recovery.The simulation results show that the optimal reconfiguration of microgrid greatly reduces the energy flow between AC and DC buses during the operation of microgrid.In detail,the switching power of bi-directional converter is reduced by 9.4 kW in grid-connected mode and 10.8 kW in off-grid mode.When microgrid failures occur,the fault reconfiguration of microgrid can quickly restore the power supply in the fault area within 0.2 s.Distributed generation unit can ensure that all the load in the blackout area can be restored to power supply under normal conditions.In the case of individual distributed generation unit failures,the total output of distributed generation unit is less than the total load,so as to ensure the continuous power supply of important loads.