Real-Time Optimal Dispatch Of Large-Scale Electric Vehicles Charging Considering Characteristics Of Distribution Network

With the development of technology, the electric vehicles (EVs) are now gradually from the product concept to large-scale industrial production. They are not only clean, low-carbon, efficient, and can be regarded as mobile energy storage devices, the participants of the electricity market, transmission of the information carriers etc. All that would be an Epoch-making significance to the development of human society. For the power systems, to take the advantages of electric vehicles as well as to reduce their effects on the power grid, has become the key subject for the related research.EVs involved in power grid operation by means of charging which is the most important reason of effects arise. Studies have shown that the uncontrolled charging can cause charging power quality decline, net loss increases, system failures and other problems. On the contrary, the optimal scheduling real-time charging (also known as intelligent charging) can take grid security into account, and also generating considerable economic benefits. However, grid complexity, charging behavior uncertainty increases the difficulty of the optimal scheduling real-time charging, the optimal scheduling real-time charging must include at least the following two technological breakthroughs:An optimization model which can accurately track changes in load profiles and a control method which can revise error in real-time. The main contributions of this paper are as follows:① Present an optimal charging model which has two characteristics:a. The improved balance equation can accurately describe the complex three-phase distribution network structure, but also improve the speed of calculation model. b. The new objective function can fully present the EV charging service framework by integrating the interests of all parts (customers, the power grid and EV aggregators).② Introduce receding horizon control (RHC) to scheduling process. Basing on its fundamental theory, an efficient real-time charging strategy is proposed in this thesis. By integrated the features of electric vehicle charging, the proposal method can achieve a balance between the optimization objective and the computing efficiency using the variable horizon receding, selection of partial control targets, slack of constraints and event-driven.Simulation results on Irland-11, IEEE-33 and IEEE-119 distribution networks demonstrate that the proposal optimal scheduling real-time charging method can manage the security constraint effectively. The improved strategy shows efficient calculation and great optimization effects, while the load and electricity price profile can be tracked as well.