Study On Distributed Optimazation Methods For Economic Dispatch Of Multi-Microgrid Systems Considering Electricity Trading

With the large-scale exploitation and consumption of fossil fuels,the global energy crisis is getting worse.The renewable energy has become the major use to replace the traditional fossil energy power generation.However,renewable energy is highly random and intermittent,and its direct access to the grid has a strong impact on the grid.Therefore,the use of microgrid technology can effectively reduce the adverse impact of renewable energy fluctuations on the grid.However,the capacity of the microgrid is limited.In order to improve the reliability of power supply,interconnect many geographically adjacent microgrids to form multi-microgrids is an effective method.However,compared with a single microgrid,the economic dispatch and power transaction issues of multi-microgrids is more complicated.In order to solve the economic dispatch problem of multi-microgrids,this paper studies the constrained distributed optimization method with weighted gradients(DOWG),and then the multi-microgrid distributed optimization economic dispatch method with weighted gradients is studied.And further considers the power transaction mechanism to study the economic dispatch method of multi-microgrids.Firstly,in order to improve the convergence speed and accuracy of traditional distributed optimization methods,a constrained distributed optimization method with weighted gradients is proposed.Secondly,construct a multi-microgrid economic dispatch model based on the multi-microgrid information-energy interdependent network,and propose a multi-microgrid distributed optimization economic dispatch method with weighted gradients.Then,considering the microgrid as an independent stakeholder to participate in the economic dispatching of multi-microgrids,a multi-microgrids hierarchical economic dispatch model considering bilateral bidding transactions is proposed,and the distributed optimization economic dispatch method for multi-microgrid considering bilateral bidding transactions is also proposed.Finally,a multi-microgrid simulation platform is built in MATLAB/Simulink,and related examples are designed to test the method proposed in this paper.The main research contents of this article include:(1)Constrained distributed optimization method with weighted gradients.Aiming at the problems of slow convergence speed and low convergence accuracy of distributed optimization methods,the idea of Nesterov acceleration gradient and virtual information network is introduced,and a constrained distributed optimization method with weighted gradients is proposed.This method can improve the convergence speed while ensuring the convergence accuracy.Furthermore,the convergence of the distributed optimization method is theoretically analyzed.Finally,the standard test function is used to construct an optimization problem with equality constraints and inequality constraints,and a design example is used to compare the convergence speed and accuracy of different distributed optimization methods.The test results show that the distributed optimization method proposed in this paper has faster convergence speed and higher convergence accuracy,and the convergence accuracy is not affected by changes in the communication topology.(2)Distributed optimization economic dispatching method for multi-microgrid with weighted gradients.In order to realize the distributed economic dispatch of multi-microgrid,this paper proposes a distributed optimization economic dispatching method for multi-microgrid with weighted gradients.First,build a multi-microgrid information-energy interdependence network model,where the bottom network is a multi-microgrid system,and the upper network is an information network composed of Agents;then,based on the multi-microgrid information-energy interdependence network model,build multi-microgrid economic dispatching model within and between networks;after that,the multi-microgrid economic dispatching model is solved by a constrained distributed optimization method with weighted gradients,and the optimal economic dispatching plan of the multi-microgrid is obtained.Finally,a multi-microgrid simulation platform was built in MATLAB/Simulink,and the example is designed to test whether the proposed distributed optimization method can realize the economic dispatch of multi-microgrid.The simulation results show that when the load and environment fluctuate drastically,or the communication network topology changes,the method proposed in this paper can realize the economic dispatch of multi-microgrid.In addition,by comparing with the centralized economic dispatching method,the method proposed in this paper have higher fault tolerance in the case of Agent failure.(3)Distributed optimization economic dispatch method for multi-microgrid considering bilateral bidding transactions.Introduce the electricity market trading strategy,and consider the microgrid as an independent stakeholder,and participate in the economic dispatch of the multi-microgrid,so as to propose a multi-microgrid distributed optimization economic dispatching method considering bilateral bidding transactions.First,a multi-microgrid hierarchical economic dispatch model with bilateral bidding transactions is established.Then,according to the multi-microgrid hierarchical economic dispatch model,a multi-microgrid intra-network distributed dispatch model and an inter-network transaction model are constructed.Next,the strategy of quotation and bilateral bidding trading for microgrid are proposed,and the constrained distributed optimization method with weighted gradients is used to solve the distributed dispatching model in the multi-microgrid network,and the bilateral bidding transaction strategy is used to solve the multi-microgrid transaction model.Finally,a multi-microgrid simulation platform was built in MATLAB/Simulink,and the example is designed to compare the impact of different DG power generation cost parameters and different trading strategies on the economic dispatch of multi-microgrid.The simulation results show that the method proposed in this paper can greatly reduce the comprehensive transaction cost of multi-microgrid and has better economic efficiency compared with the distribution network transaction.