The ever-increasing penetration level of renewable energy and electric vehicles may threaten power grid operation. Dealing with uncertainty in smart grids is critical in order to mitigate possible issues. This research work proposes a two-stage stochastic model for large-scale energy resources scheduling for aggregators. The proposed model is designed for aggregators managing a smart grid. The idea is to address the challenge brought by the variability of demand, renewable energy, electric vehicles, and market price variations while pursuing cost minimization. Benders' decomposition approach is implemented to improve the tractability of the original model and its' computational burden. A realistic case study is presented using a real distribution network in Portugal with high penetration of renewable energy and electric vehicles. The results show the effectiveness and efficiency of the proposed approach when compared with a deterministic formulation and suggest that demand response and storage systems can mitigate the uncertainty.

Electric Vehicles (EVs) are an important source of uncertainty, due to their variable demand, departure time and location. In smart grids, the electricity demand can be controlled via Demand Response (DR) programs. Smart charging and vehicle-to-grid seem highly promising methods for EVs control. However, high capital costs remain a barrier to implementation. Meanwhile, incentive and price-based schemes that do not require high level of control can be implemented to influence the EVs' demand. Having effective tools to deal with the increasing level of uncertainty is increasingly important for players, such as energy aggregators. This paper formulates a stochastic model for day-ahead energy resource scheduling, integrated with the dynamic electricity pricing for EVs, to address the challenges brought by the demand and renewable sources uncertainty. The two-stage stochastic programming approach is used to obtain the optimal electricity pricing for EVs. A realistic case study projected for 2030 is presented based on Zaragoza network. The results demonstrate that it is more effective than the deterministic model and that the optimal pricing is preferable. This study indicates that adequate DR schemes like the proposed one are promising to increase the customers' satisfaction in addition to improve the profitability of the energy aggregation business.

No abstract

With the high penetration of renewable generation in Smart Grids (SG), the uncertainty behavior associated with the forecast of weather conditions possesses a new degree of complexity in the Energy Resource Management (ERM) problem. In this paper, a Multi-Objective Particle Swarm Optimization (MOPSO) methodology is proposed to solve ERM problem in buildings with penetration of Distributed Generation (DG) and Electric Vehicles (EVs) and considering the uncertainty of photovoltaic (PV) generation. The proposed methodology aims to maximize profits while minimizing CO2 emissions. The uncertainty of PV generation is modeled with the use of Monte Carlo simulation in the evaluation process of the MOPSO core. Also, a robust optimization approach is adopted to select the best solution for the worst-case scenario of PV generation. A case study is presented using a real building facility from Brazil, to verify the effectiveness of the implemented robust MOPSO.

We apply a hierarchical SDN controller to demonstrate multi-layer orchestration of a commercial Optical Network Control platform. We show dynamic allocation of transport resources for bandwidth on demand and congestion control of packet services.