Electric Vehicle Load Forecast Based on Higher Order Markov Chain

Electric vehicles (EVs) with mobile energy storage characteristics are a class of flexible and high-quality demand-side resources. In order to solve the problem that the load of EV charging stations is difficult to be accurately predicted, this paper proposes a high-order Markov chain-based EV aggregation model. Firstly, the Poisson distribution is used to predict the charging start time of EVs to solve the external influencing factors in the subsequent modeling; then, the State-of-charge (SOC) state of EVs is discretized in two layers, the first layer can clearly define the charging and discharging state of each EV in the charging station by using fuzzy partition, and the second layer continues to subdivide each interval on the basis of fuzzy partition to realize the double layer discretization, which reduces the dimensionality of the state space and Finally, the results show that the proposed model can accurately predict the load of EVs in charging stations.