Optimal Scheduling Strategy for Electric Vehicles Based on User Response Willingness Three Dimensional Sigmoid Cloud Model

There is a considerable uncertainty about the willingness of electric vehicle (EV) owners to respond to scheduling, which brings great challenges to the optimal scheduling of electric vehicles. Therefore, a three-dimensional Sigmoid cloud model of user response willingness is proposed to obtain the time-phased EV optimal scheduling strategy. First, the Copula entropy between the probability density functions of EV charging load for each historical day is analyzed in accordance with Hampel's criterion to identify the historical correlation days. The probability distribution of vehicle parking time and charging state is predicted by BP neural network on the basis of the travel probability distribution of strongly correlated days. Second, for the goal of describing the response willingness of EV users under the influence of multiple factors, a three-dimensional Sigmoid cloud model is developed to depict the uncertain mapping relationship of EV users' price gain and time margin on response willingness. Finally, based on the sensitivity of response willingness, different response willingness is stripped down to correspond charging scheduling intervals, and EVs are optimally scheduled in time with the aim of minimizing the load fluctuation and schedule cost of the distribution network. Simulation result shows that the proposed three-dimensional Sigmoid cloud model quantifies the uncertainty of EV response behavior and reduces the root-mean-square error by about 35% compared with the traditional method. Moreover, the load fluctuation rate of the distribution network is decreases by 11.28% while satisfying the vehicle needs of users and their response willingness.