Impact And Optimal Dispatch Of Electric Vehicles Access To Power Grid

With the further deterioration of the fossil energy depletion and environmental crisis, the electric vehicle(EV), which uses electricity as the energy carrier, is considered as a promising option to reduce the CO2-emissions and oil dependency. It has been promoted extensively by governments, thus the EV market may come to an explosive growth. However, EVs would exert inevitable impact on power system operation due to the spatial and temporal stochastic nature of the charging load. Therefore, research on establishment of EV charging load model considering spatial and temporal distribution, coordinated charging control of large-scale EVs and multi-objective optimization scheduling of renewable energy and EVs, is the basis and premise to improve the ability for the power grid to accept EVs, to smooth the natural variability of the renewable energy, and to promote the EV popularization.In this paper, the travel survey data of the traditional fuel vehicles, combined with the diversity of the users' travel habits and the charging locations, is used to analyze the spatial and temporal distribution rules of the EV charging behaviors. The EV charging demand model is established by the energy consumption model based on the actual operating conditions and the charging load probability model. To reduce the negative effect of large-scale EVs randomly access to the grid, two EV coordinated charging strategies, i.e., variance-optimal and loss-optimal, are studied in this paper. The variance-optimal charge strategy is a convex quadratic programming problem containing EV number constraints. The loss-optimal charge strategy contains charging load constraint, battery charge capacity constraint, power flow constraint, voltage amplitude constraint, line capacity constraint and etc., and is solved by PSO algorithm and dynamic penalty function method. In research on multi-objective optimization dispatch of the renewable energy and EVs, wind power generation model and solar power generation model are established. A multi-objective optimal function containing the equivalent load variance minimization, the total distribution network losses minimization, and the EV users' charging cost minimization is built. The energy storage system of the EV battery is used as the buffer between the power grid and the renewable energy. The multi-objective search algorithm based on quantum-behavior particle swarm optimization which has stronger global search capability, is applied to get the reasonable EV numbers at each gridable time, thus maximizing the bilateral interests for both the grid side and the user side.