| Renewable energy has greater randomness and intermittency,which makes the structure of the power grid more complicated.Microgrid can coordinate the conflict between distributed power generation utilization and large power grid,and has been widely used.With the increasing number of distributed power generation units incorporated into the microgrid,the scale of the grid has gradually increased,and the regulation pressure and operating cost of the microgrid have also increased.In order to better ensure the frequency synchronization of the microgrid and reduce the operating cost of the microgrid,this paper combines the Kuramoto model and the consensus algorithm to study the frequency synchronization and economic operation of the island microgrid.In the model construction and research of the microgrid,the models of producers and consumers(renewable energy generation with local load)such as rooftop solar energy are considered,making the research of microgrids more comprehensive.The research content of this paper is as follows:The first part is to study the frequency synchronization of island microgrid.Firstly,using graph theory knowledge,considering the characteristics of distributed power generation,micro-gas turbines,energy storage,and load,combined with droop control to derive the corresponding Kuramoto-like models,and on this basis,construct the Kuramoto-like model of the overall system,thereby integrating the micro The stability problem of the power grid is transformed into the frequency synchronization problem of the Kuramoto-like model.Based on the overall model,the sufficient and necessary conditions for system frequency synchronization are derived.At the same time,the Lojasiewicz exponent is used to prove that the rate at which the system converges to frequency synchronization is exponential on the basis of satisfying the sufficient conditions for frequency synchronization.Then,simulations are carried out in three large-scale networks of star,ring and tree to judge whether the system achieves frequency synchronization under different parameter conditions,so as to prove the validity of sufficient and necessary conditions.Finally,in the non-uniform Kuramoto model,the validity of the sufficient and necessary conditions is verified again,and the coupling strength between nodes obeys the degree distribution and power law distribution,and the speed of the network convergence is evaluated according to the order parameters of the network.Therefore,in actual situations,theoretical results can be used to judge the frequency synchronization of the system,adjust the system parameters,and construct the network topology.The second part is to study the economic operation of island microgrid.First,establish a corresponding cost model based on the functional relationship between the output power and cost of each unit in the microgrid,including micro-gas turbines,renewable energy units,energy storage,loads,and renewable energy units with local loads.Second,consider the power constraints to obtain the objective function with the lowest total operating cost of the system,combine the optimal conditions to extract the cost increment of each node,and use the distributed consensus algorithm to solve the incremental cost of the node,thereby obtaining the increment Consistent cost and optimal power.Finally,a five-node model is constructed,and simulations are performed under the six conditions of not considering generation power constraints,considering generation power constraints,plug and play,generation power changes,load power changes,and large-scale networks,and compare them with centralized algorithms to verify the effectiveness and robustness of the control strategy.In the third part,the optimal power obtained in the second part is used as the target value,and the rated power in the first part of the Kuramoto model is iterated to the optimal power through the consensus algorithm.Combined with the analysis of sufficient and necessary conditions for system frequency synchronization,when frequency synchronization can be achieved,the frequency of the system node will converge to the natural frequency,and the actual output power will be consistent with the optimal power,thereby achieving frequency synchronization and economic operation of the system. |