Research On Bi-level Optimal Dispatching Of Active Distribution Network With Multi-microgrid Connetion

The distributed generation(DG)is connected to the distribution network(DN)at a high ratio to form an active distribution network(ADN)system with source-network integration control.However,the uncertainty of DG output such as wind / light has a certain impact on the reliability of the ADN system,so multiple DGs in the form of microgrid(MG)are connected to ADN.MG operation can maintain the voltage and power stability at the grid connection point,meanwhile reducing the impact of DG uncertainty on ADN system.Therefore,it is advantageous to connect multiple DGs with uncertain output in the form of MG to the ADN system.With the increasing demand for electricity,the depletion of non-renewable energy and the intensification of environmental pollution,researchers began to study the integration of multi-MG into the ADN system.On the one hand,increase the use of renewable energy,on the other hand,improve the flexibility and stability of the ADN system.After the multi-MG is integrated into the ADN system,how to solve the problems of coordination and stable operation between each microgrid and ADN system,improve the anti-interference ability of the ADN system,and realize the optimal economic dispatch of multi-MG and ADN system still need further study.For the optimal scheduling of ADN systems with multi-MG,when the entire network is used to optimize scheduling,the different operating conditions of each MG will make it difficult to solve the optimal power flow of the ADN system,and the actual fluctuations of MG will has an impact on the stability of the ADN system.In order to realize the optimal operation of the MG and ADN,solve series of problems caused by the uncertaint y of its operating state and output after being incorporated into multi-MG,as well as the problems of low accuracy and poor convergence during optimization and solution.This paper puts forward a bi-level optimization method is proposed to study the optimal scheduling problem of ADN system with multi-MG.The main work and results are as follows:(1)This thesis inductes and summarizes the research situation of optimal scheduling of MG and MG access to the ADN.The network structure and operation mode of in the form of MG and ADN are summarized,the relationship between MG and ADN is discussed,and the basic method of power flow calculation in ADN system and the concept and solution method of optimal power flow are described.(2)It based on the optimal power flow of a b i-level optimal dispatching model for an ADN with multi-MG is constructed.The ADN system with multiMG is divided into upper and lower layers for research.The upper layer is AD N and the lower layer is MG.The upper layer regards the connected MG as power source and MG predicted output value is used as input data.The power of the PCC between the upper and lower layers is the optimization variable,with the balance of power flow as the constraint,and the minimum network loss of the system is the goal,and set up optimal power flow model on ADN layer.The second-order cone relaxation technique is used to convert the non-convex nonlinear optimal power flow model into a convex feasible domain second-order cone programming model,and the Gurobi solver is used to obtain the optimal power flow distribution at the ADN layer and PCC power on the MG.The power of the PCC optimized by the upper layer of lower MG is constrained,and the optimal scheduling model of the controllable unit is established with the minimum operating cost of the MG as the target.The improved genetic algorithm(GA)combining Tent mapping chaos technology and NDX cross over technology to solve.Finally,an improved IEEE-33 node DN including multi-MG is analyzed as example.The results show that the proposed model and algorithm are feas ible,which reduces the network loss of the ADN system to a certain extent and decreases the economic cost of the ADN system.Meanwhile,when the wind / sun output fluctuates,the lower model can still be locally adjusted and optimized,thereby reducing the impact of MG fluctuations on the ADN system and improving the reliability and robustness of the system.By optimizing the control of the power of PCC,the DN layer and multi-MG layer are coordinated with each other to achieve their optimal operation.(3)It based on the time-of-use(TOU)power price mechanism of a bi-level optimal dispatching model of multi-MG and ADN is established.Considering the impact of multi-MG electricity sales revenue at different time periods and electricity prices on the economics and reliability of the ADN system,the amount of power transmitted when buying and selling electricity from the MG to ADN is used as a decision amount to establish a bi-level optimal scheduling model of multi-MG and ADN system based on TOU power price mechanism.According to the optimization scheduling model of the lower grid-connected MG,the MG trading power is obtained and transmitted to the upper layer through the PCC.The upper is constrained by the multi-MG trading power and power flow balance,and a new optimal power flow model is established,and the second-order cone relaxation technique and Gurobi solver are used to solve the optimal power flow of system.Still taking the improved IEEE-33 node DN as an example,the results show that the revenue from buy and sell of electric ity on each MG under the TOU power price reduces the operating costs of multi-MG and operating cost of ADN system.Simultaneously the power flow solution is not affected by the buy and sell of electrical power on each MG,and the system can still be operated safely and reliably.