Short-term Load Forecasting Study For Electric Vehicle Charging Stations In Smart Grids

With the deepening of the global energy crisis in recent years, oil and other resources are dried up, the global warming trend is increasing, electric vehicles is becoming one of the main development direction of new energy vehicles because of its energy saving and emission function. The large-scale use of electric vehicles can not only peak load shifting,improve the economic benefits of power grid, but also help us protect the environment.As an important part of power load, we study on the short term load forecasting of electric vehicle charging station. On one hand, it is signality to the economic dispatch of power system; on the other hand, the economic operation of charging stations is also need the support of corresponding charging power data.On the basis of reading a large number of references, in this paper, smart grid and electric vehicle charging station are introduced, and the practical application of the electric vehicles in smart grid is described.We study the load characteristic of the electric vehicle charging station, a short term load forecasting model for electrical vehicle charging stations based on spike neural network is raised in this paper. Spike neural network encode information in the timing of single spike, making it with strong calculating ability, good real time capability and large information capacity. Verifies with simulation example, spike neural network forecasting model based on SNN has a better prediction accuracy than traditional BP-NN forecasting model, which demonstrates the effectiveness of this model.As the electric vehicle charging load has the characteristic of volatile, the error of using spike neural network forecasting model is still large. In order to further improve the accuracy of short term load forecasting of electric vehicle charging station, the Elman feedback spike neural network is proposed in this paper, this new model add a Elman feedback on the original forecasting model, making the network has the ability of storing previous information. Verifies with simulation example, the prediction accuracy of Elman feedback spike neural network has been further improved, which provides a strong theoretical basis for application practice of short term load forecasting of electric vehicle charging station.