A stochastic simulation framework for planning and management of combined hydropower and wind energy systems

Pumped storage within hydroelectric reservoir systems is a proven technology with very high efficiency, as well as the unique large-scale energy buffer. The storage of energy is implemented by pumping water upstream, for taking advantage of the excess of energy (e.g. during night hours), and next retrieving this water to generate hydropower during demand peaks. Interestingly, this excess can be offered by other renewable energy sources, particularly wind turbines, which can be integrated within hydroelectric systems with pumped storage facilities, to formulate autonomous hybrid renewable energy schemes. The optimal planning and management of such systems is a challenging task, which requires a holistic viewpoint and a consistent representation of the multiple sources of uncertainty. In this respect, a novel framework is proposed, which is tested in an existing hydrosystem of Greece (i.e. the reservoir system of Aliakmon, which also serves other water uses), considering a combined operation with a hypothetical wind power system. The two components, which are linked through a single pumping storage plant, are modelled in different time resolutions. In particular, for the representation of the water resource system we adopt, as typically, a monthly time step, while for the wind power system we use hourly steps. For both systems, the input variables (i.e. hydrological inflows and wind velocity, respectively) are generated via appropriate stochastic simulation models, by means of synthetic time series of 1000 years length. In order to ensure the most beneficial performance of the integrated system, we investigate different design parameters of the wind turbines, for which we optimize the operation policy of the hydroelectric reservoirs.