Energy and power management in Smart Commercial Buildings using V2G applying a Robust Affine Policy

As electric vehicles (EV) become increasingly widespread, smart commercial buildings, among others, can exploit the related opportunities for their battery usage to improve global sustainability. This paper presents a comprehensive solution for an existing standard office building to optimize the energy purchased from the grid, considering energy and power bills and different demand scenarios. The goal is to obtain a day-ahead charging policy for the EV to apply in out-of-sample scenarios. The building includes four EV charging stations, with similar battery characteristics for all vehicles, differing only in the initial state-of-charge and the efficiency of the battery. All four vehicles are available for both grid-to-vehicle and vehicle-to-grid connections. The proposed mathematical model is formulated to minimize the total cost of electricity among all scenarios the building perceives, considering a robust approach with the demand as the uncertain parameter. Experiments for the in-sample test are conducted considering 15 working days of three consecutive months of the same year, for which we can obtain an affine policy for each month. The following working days of each are used for the out-of-sample test to evaluate the monthly affine policy’s performance. The results show a reduction of about 2.5% in the monthly billing value compared to the deterministic model available in the literature.