Distributed Power Flow Calculation And Optimization Methods For Hybrid AC/DC Microgrid

The microgrid is an effective method to realize the safe and reliable connection of distributed energy resources that are massive and widely distributed to the grid.The hybrid AC/DC microgrid which combines the characteristics of AC and DC grids,and thus possesses excellent advantages.Effective energy management for the microgrid is required to fully exert the advantages of the microgrid.Power flow calculation and optimal operation,as two important aspects of energy management,have attracted extensive attention and research.However,the three-phase imbalance of the AC microgrid,the coupling between the power of AC and DC microgrids,the volatility and intermittence of distributed renewable energy resources,and the difference of different operation modes all increase the difficulty of analysis,calculation and optimization for the microgrid.The existing research fails to fully consider the above factors,and also fails to properly solve the contradiction between the increasing demand for analysis,calculation,optimization and regulation under the scenario that large-scale distributed energy resources are connected to the grid,and model complexity and calculation amount.Therefore,it is significant to study efficient and reliable power flow calculation and optimal operation methods for the microgrid to ensure its safe and economic operation.To resolve the above issues,from the aspects of modelling and algorithm,this paper proposes a distributed power flow calculation method for the AC/DC hybrid microgrid and an optimal operation method for the microgrid under different operation modes.First,considering the primary control of distributed power generators,three-phase imbalance of feeders,and the coupling between AC and DC sides,this paper establishes the sequence-component mathematical model of common elements in the microgrid.Based on the models,a sequence Newton-Raphson-based algorithm for solving the power flow of the hybrid AC/DC microgrid is designed.The algorithm is applicable to both grid-connected and islanded microgrid,and can be used for power flow analysis and security analysis.Then,the linearization method for the AC power flow model that leads to an NP-hard optimization problem is studied to improve the computational efficiency of the subsequent optimization problem.Based on the accumulated historical data of the dispatching organization,a data-model-driven linear power flow model is proposed,which also considers the impact of bad data and ZIP load.Then,since the renewable energy resources lead to the uncertainty of system operation states,a linear power flow model that considers the constraints of the confidence of linearization errors is proposed to improve the "out-of-sample accuracy".Finally,an online application method for online optimization is proposed,which improves the construction speed while guarantees the accuracy of the linear power flow model.The study of linear power flow model is the basis for the subsequent research of the optimization operation of microgrid.Regarding the optimal operation methods,two kinds of research are studied based on different operation modes of the microgrid.First,for the grid-connected microgrid,this paper focuses on its ability to interact with the upper grid,that is,the microgrid receives and tracks the dispatching commands of the upper grid.This paper proposes distributed command tracking framework and algorithm,which solves the contradiction between the command tracking accuracy and the calculation amount of the optimization problem based on the mode of"measurement feedback and trot",and also improves the rapidity and accuracy command tracking of the microgrid.For the islanded microgrid,ensuring its frequency and voltage security is one of the primary issues due to its low-inertia characteristics and lack of power support of the main grid.The primary voltage and frequency regulations are automatically accomplished by the droop control.To improve the droop control effect,"look-ahead and modification" multi-stage optimization method for droop control parameters is proposed.The uncertainty of load or other power fluctuations is considered in the look-ahead stage which improves the robustness of droop control parameters while ensures economy to some extent.When the power fluctuation exceeds the regulation capacity of the droop control based on the parameters obtained from the look-ahead stage,the droop control parameters are modified in two adaptive modification stages based on the real-time operating states of the microgrid,which promotes the restoration of the violated frequency and voltage.In all,the above multi-stage cooperation method can improve the operating safety and economy of the islanded microgrid simultaneously.Finally,all the methods proposed in the above researches are analyzed under different test systems and scenarios which all validates the effectiveness of the methods.