Following deregulation in the electricity grids, power systems has faced new challenges in terms of diversification of generation units and demands types, which requires a more comprehensive management framework. For this purpose, several new players were introduced to resolve the challenges between generation and demand side. Among others, retailers are the one that play a crucial role by creating a link between electricity market operators and the consumers, seeking maximize profits and reduction of the costs of their customers. Electric vehicles (EVs), meanwhile, are among the bilateral consumers which retailers are able to both provide energy for as well as see as an energy sources for sales in Day-Ahead (DA) energy and reserve markets. Nevertheless, Retailers face several uncertainties regarding the physical characteristics of electric vehicles, the behavior of their owners, in addition to the uncertainties inherent in energy and reserve markets faced by any player. In order to optimal participation of retailers in those markets as well as to meeting the needs of electric vehicles, a two-stage optimization framework is presented in this paper. Vehicle clustering is also utilized to model all uncertainties simultaneously.
Thus the main contributions of this paper can be summarized as follows:
A new method for classifying electric vehicles based on battery characteristics (such as battery capacity, charge and discharge rate, etc.) and owners' behaviors (availability at parking stations, arrival and departure times, initial charge state, etc.) is proposed. This clustering helps reduce the computational load by avoiding duplicate calculations.
A novel model is presented for retailers to the participate in the reserve market using the capabilities of electric vehicles. Therefore, in this paper, retailers participate in the energy and reserve markets simultaneously using the potential of electric vehicles.
A two-stage stochastic linear model has been introduced to consider most of the uncertainties with respect to the aggregation of the potential of EVs by the retailer to plan their participation in different electricity markets.
Using the proposed optimization framework and vehicle classification, all uncertainties related to the initial charge of the EVs’ batteries, type and capacity of batteries, the expected final state of charge, the times of arrivals and departures of vehicles to / from parking lots, EVs’ battery charging and discharging rates, EVs’ battery efficiency, reserve market call status, as well as uncertainties related to DA energy and spinning reserve prices, and the number of EVs in parking lots are modeled simultaneously.
Finally, the model has been implemented in GAMS considering the option of the retailer participation as a seller in the energy and spinning reserve markets. It has been shown that if the retailer has the mentioned choice, (s)he can benefit from selling in both markets even if sell energy at a low price to electric vehicles in parking lots.