Neda Vahabzad - Faculty of electrical and computer engineering, University of Tabriz
Saeed Zeynali - Faculty of electrical and computer engineering, University of Tabriz
Behnam Mohammadi-Ivatloo - Faculty of electrical and computer engineering, University of Tabriz
Mehdi Abapour - Faculty of electrical and computer engineering, University of Tabriz

This study proposes a stochastic framework in order to increase wind power profitability by optimally deploying vehicle to grid (V2G) capability of present Plug-in Electric Vehicles (PEVs). Due to stochastic nature of the problem, conventional deterministic methods are likely to provide unrealistic results. Hence, an effective stochastic methodology with low computational burden is offered. The uncertainties issued from random nature of wind speed, electricity price, daily driving mileage of the PEVs and their arrival-departure time are handled through point estimate method, which has proven to be computationally effective with an acceptable degree of accuracy.Using k-means clustering, the PEVs are grouped according to their arrival and departure time. The proposed Mixed Integer Nonlinear Programming (MINLP) formulation is solved in GAMS and once the stochastic decision variables are optimally determined, the corresponding Probability Distribution Functions (PDFs) are calculated by maximum entropy method.

Plug-in Electric Vehicles (PEVs), Vehicle to Grid (V2G), renewable energy, uncertainty modeling, point estimate method (PEM), stochastic optimization, and maximum entropy method