Dedicated to the memory of Elias Kyriakides.

Peak shaving applications provided by energy storage systems enhance the utilization of existing grid infrastructure to accommodate the increased penetration of renewable energy sources. This work investigates the provision of peak shaving services from a flywheel energy storage system installed in a transformer substation. A lexicographic optimization scheme is formulated to define the flywheel power set-points by minimizing the transformer power limit violations and the flywheel energy losses. Convex functions that represent the flywheel power losses and its maximum power are derived and integrated in the proposed scheme. A two-level hierarchical control framework is introduced to operate the transformer-flywheel-system in a way that handles prediction errors and modelling inaccuracies. At the higher level, a model predictive controller is developed that solves the lexicographic optimization scheme using linear programming. At the lower-level, a secondary controller corrects the power set-points of the model predictive controller using realtime measurements. A software platform has been developed for integrating the proposed controllers in an experimental setup to test their effectiveness in a realistic testbed setting, and the flywheel system characteristics are experimentally identified. Simulation and experimental results validate and verify the modelling, identification, control and operation of a real flywheel system for peak shaving services.

Increased level of flexibility is essential in power systems with high penetration of renewable energy sources in order to maintain the balance between the demand and generation. Actually, the flexibility provided by energy storage systems and flexible conventional resources (i.e., generating units) can play a vital role in the compensation of the renewable energy sources variability. In this paper, the flexibility of the conventional generating units is quantified and incorporated in a unit commitment model in order to evaluate the impact of different system flexibility levels on the optimal generation dispatch and on the operational cost of the power system. An emerging flexible option such as the battery storage is included in the unit commitment formulation, evaluating the flexibility contribution of the storage and its effect on the system operational cost. In this paper, the flexibility of a real power system is assessed while the unit commitment problem is formulated as a mixed-integer linear program. The results show that the integration of a storage unit in the power generation portfolio provides a significant amount of flexibility and reduces the system operational cost due to the peak shaving and valley filling.

Microgrids are becoming one of the main components of future smart grids. Ensuring their optimal and stable operation is of crucial importance and can be a challenging task. In this paper, two optimization algorithms are implemented for scheduling the microgrid operation in grid-connected and islanded modes, according to the priorities and objectives in each mode. For achieving an optimal operation at each mode, the proposed scheme is able to shed loads, define the generation level of the photovoltaics and regulate the charging/ discharging level of the Energy Storage System (ESS). The effectiveness of the proposed scheduling is demonstrated through an analytical realtime simulation, where various transitions between the gridconnected and islanded modes are considered. The results indicate that the proposed scheme is able to regulate successfully the energy flows of the microgrid even under various transitions.

No abstract