Research On Bi-level Optimization Dispatch Of Active Distribution Network With Multi-Microgrids

With the wide application of distributed resources in electric power system,the operation and control of distribution network have changed dramatically.The traditional passive single-source and one-way power flow distribution network is developing towards the active multi-sources and two-way power flow distribution network.The safety,reliability,economy of distribution network have been greatly improved and the system becomes more environmental-friendly.However,the joint of large number of distributed generators,energy storage devices,controllable loads and microgrids have greatly increased the complexity of the distribution network,especially the joint of wind,solar energy and other renewable energy with great volatility and uncertain characteristics,which has brought greater challenges to the operation stability and control management of the distribution network.In order to make full use of distributed energy and ensure the coordinated operation of active distribution network and multi-microgrids,the research on optimal dispatch of active distribution network with multi-microgrids is proposed.In this paper,the main work and results are as follows:(1)This paper describes the structure,characteristics and development of microgrids and active distribution network,analyses the control and management methods of active distribution network,and the corresponding schematic diagrams are designed.Firstly,the problems of coordinated optimal dispatch of distribution network and microgrids are summarized.Secondly,MPC model predictive control,adaptive genetic algorithm and bi-level programming theory are introduced.The typical distributed generators,energy storage and loads in active distribution network are analyzed.Finally,the mathematical models of wind turbine,photovoltaic power generation system,micro gas turbine,fuel cell and storage battery are established,and various types of diagrams are designed.(2)Considering the relationship of competition and cooperation between the active distribution network and the microgrids,a bi-level optimal dispatch mode considering the participation of multi-microgrids in the operation of active distribution network is proposed in this paper,of which the upper-level model takes the distribution network as research object and its objective is to improve power quality and reduce power loss;and the lower-level models takes the microgrids as research object and aims at minimizing the operational cost.The genetic algorithm is applied in this paper to solve the upper-level model and mixed integer linear programming algorithm is introduced to solve the lower-level model.The modified IEEE 33 distribution network is presented as a case,and the simulation results demonstrate the effectiveness of the proposed bi-level optimal dispatch model.(3)Considering the significance of economic benefit of distribution network,a two-stage optimal dispatch model of distributed energy system is established with the objective to minimize the operation cost of distribution network.MPC rolling optimization and feedback are used in the day-to-day phase to track the changes of operation parameters in real time and make corrections.Active power optimal dispatch of multi-time scale active distribution network based on MPC model predictive control is realized.Results obtained from the case of modified IEEE 33 and comparative analysis demonstrate the feasibility and effectiveness of the proposed dynamic optimization strategy.(4)A GUI experimental platform for joint optimization dispatch of active distribution network with multi-microgrids is designed.Based on the analysis and research of the optimal dispatching situation of distribution micro-coordination,an experimental platform of active distribution network dispatching optimization with multi-microgrids is designed by using MATLAB GUI platform.This platform have the characteristics of good sensibility,convenience and rapidity in simulating optimization dispatching of active distribution network.It provides convenient conditions for further study and research.