Optimal V2G Method Considering Uncertainty of Distributed Generation and Load

Abstract

The penetration of electric vehicles (EV) has brought many benefits to the active distribution network due to their unique advantages. But, EV also has negative impacts on the distribution network for their uncertainty. Moreover, the integration of distributed generation (DG) and the uncertainty of load demand also increase the negative impacts of EV on the distribution network. In this paper, optimal vehicle-to-gird (V2G) method considering uncertainty of distributed generation (DG) and load was proposed to reduce the negative impacts by minimizing the load peak-valley difference and maximizing the user economic benefits. To reduce the uncertainty of DG, we combined particle swarm optimization (PSO) and backpropagation neural network to create a prediction model to predict the output of DG. And Time-of-use (TOU) is used to optimize the load curve. Finally, the optimal charging and discharging time in a day can be obtained by improved particle swarm optimization algorithm (IPSO). The proposed method is verified by numerical analysis, which highlights that the proposed V2G scheduling can significantly improve distribution network stability and user economic benefits.