Impact Of Large-Scale Electric Vehicles Charging On Power Grid And Its Optimal Control Strategy

With the continuous deterioration of the energy shortage and environmental crisis, the electric vehicle(EV) is raising worldwide concern due to its low carbon and energy saving properties. Since EV technology is getting matured, large-scale application of EVs means not only an opportunity but also an unprecedented challenge to grid. Therefore, research on establishment of EV charging load model based on users’ driving pattern, impact of large-scale EVs connected to grid, and coordinated charging schedule, is of significant importance to promote the reliability of power supply and the spread of EVs.From the perspective of the factors influencing EV’s charging demand and the uncertainty of EV’s spatial and temporal distribution, the charging load model of large-scale EVs connected to grid without control is built based on the calculation of a single EV’s charging load. On the basis of the charging load model, a loss-optimal charging strategy and a time-of-use(TOU) charging strategy which aims at the comprehensive optimization of users’ charging costs and load variance are studied respectively. Mainly focused on the aspects of load curves, voltage deviation, and total losses, the impact of large-scale EVs charging on grid is analyzed and compared under three different charging strategies: uncontrolled, loss-optimal and TOU strategy. Since charging behavior involves both grid and EV users, single-objective optimal charging can hardly balance the interests of both sides. A multi-objective optimal model with three objective functions, the equivalent load peak-valley difference, total network losses, and users’ charging costs, is built, also considering the constraints of charging load, flow, and voltage amplitude, et al. The improved multi-objective particle swarm optimization algorithm and analytic hierarchy process are applied to solve the model and get the best scheme of EV numbers at each gridable time, thus maximizing the overall interests for both grid and users. In research on the hierarchical coordinated scheduling of EVs, considering distributed generation and controllable load, upper scheduling takes minimizing load variance as object to establish a regional-level control system, and lower scheduling takes minimizing the error between the command and EV’s real-time charging as object to built a user-level control system for guiding EV’s charging behavior. The hierarchical coordinated scheduling model provides a reference for actual dispatch management of EV access to grid.