Efficient Scheduling And Charging Of Electric Vehicles With Load Forecasting Models

Electric vehicles are the main technological means for green transportation and environmental control.It is necessary to develop an optimized electric vehicle charging scheduling strategy to minimize charging costs for electric vehicles.Firstly,the optimization of charging scheduling strategies for electric vehicles is closely related to the basic load of electricity.Therefore,the paper utilized historical data and temperature information of a certain area to predict the basic power load of the region based on artificial neural networks(ANN),bagged trees,and linear regression models,and compared the prediction accuracy of the three models.Subsequently,the paper established an optimized charging scheduling model for electric vehicles.The main objectives of this model include reducing the energy demand of vehicles,eliminating waiting time at charging stations(CS),and reducing charging costs.Assuming that the number of electric vehicles and CS available in a specific limited area is limited,and considering the traffic conditions,driver behavior,vehicle density and other constraints on the road,the optimization algorithm based on linear programming can be used to optimize the model.Finally,combining the load forecasting model and the optimized charging scheduling model,the paper conducted actual simulation of the charging scheduling scheme for electric vehicles.The simulation results show that the average absolute percentage error of the ANN model for monthly load forecasting is 0.89%,which is significantly better than the other two models.The electric vehicle charging scheduling model can allocate electric vehicles to the preferred CS and ensure that the number of electric vehicles in the CS does not exceed the limit,while all vehicles are allocated to the CS with the best charging state.The results of the paper can provide reference for the charging scheduling of electric vehicles.