Research On Electric Vehicle Charging Station Location And Capacity Optimization

With the increasing deterioration of the global environment and the gradual shortage of energy, electric vehicles have received extensive attention from every country all over the world as a new type of green-energy vehicles. At present, governments around the world are actively promoting the development of electric vehicles. As the electric vehicle energy supply infrastructure, the electric vehicle charging stations have a direct effect on its development. The reasonable and scientifical charging station network can not only provide the users with convenient charging service, but also reduce the construction and investment cost for the operators. As a result, carrying out research on the electric vehicle charging station planning has very important practical significance.In order to solve the electric vehicle charging station location and capacity problem further, this paper firstly analyzed four typical electric vehicle charging station location and capacity optimization models. Among them, optimization goals included maximization of annual charging station operating income, minimization of user time consumption, etc. And constraint conditions included transformer capacity constraints, node voltage amplitude constraints, system power flow constraints, etc.Then, on that basis, taking the interests of both operators and users into consideration, this paper proposed the electric vehicle charging station location and capacity optimization model considering the cost of the whole society, and combined Particle Swarm Optimization with the global random search optimization ability with the Voronoi diagram with the ability to divide areas automatically together to solve the proposed optimization model.Finally, this paper used an actual example to verify the proposed model, comparing and analyzing change tendencies and reasons of various costs with the increasing of charging station number. The results showed that the proposed optimization model and solving algorithm were reasonable and feasible.