Distributed Economic Dispatch Method Of Smart Grid And Its Convergence Optimization

Using distributed algorithms can enjoy the advantages of local information requirement,more robustness,and more flexible scalability in wide area protection and control applications.The consensus-based distributed algorithms can eventually drive state valuables of each controllable object to a same value rely on the information came from its own sensors and its neighbor controllers in the local area network.Smart grid is a typical cyber-physical system(CPS)combine the electric power system with the information and communication system.It provides a new method to solve the economic dispatch problem(EDP)with the rapid development of power CPS.In the first part of this paper,the CPS model of smart grid is established and the closed-form solution of the EDP is deeply researched.Secondly,the closed-form solution of the EDP is rewritten into two forms,one of which can easily apply average-consensus algorithm and the other can apply asymptotically consensus algorithm.Meanwhile,the distributed polynomial filtering skill is introduced to make the proposed algorithms converge fast,which can be utilized in a fixed communication network by using the historical datum recorded in each controller.Simulation results are provided to demonstrate that our distributed economic dispatch methods is effective and the convergence speed is accelerated by the polynomial filtering skill.Thirdly,a distributed consensus-based collaborative control strategy was proposed to deal with the load dispatch problem in active distribution network with a variety of power sources,which can deal with the cases of constant tie-line power and optimal economy.Simulation results indicated that the proposed strategy was effect in real-time economic dispatch in cases of peak-valley price of the main grid,intermittency of the renewable energy and power fluctuation of the load.