|GIUDICI FEDERICO||Cycle: XXXII |
Section: Systems and Control
Tutor: BOLZERN PAOLO GIUSEPPE EMILIO Major Research topic
:Developing a Decision-Analytic Framework to explore the water-energy nexus in the remote communities of small Mediterranean islands.
Advisor: CASTELLETTI ANDREA FRANCESCOAbstract:
Water and energy supply in small Mediterranean islands are strictly interrelated and face a large number of pressing issues, mainly caused by the distance from the mainland and the lack of accessible and safe potable water sources. The majority of these remote and isolated communities are not connected to the mainland electricity grid and energy security is mainly assured by stand-alone, carbon intensive systems, which use diesel generators to meet the energy demand. The high dependence from the remote supply of fuel and the need to store adequate reserves to cover possible refuelling delays make the operations of these systems very costly and inefficient. Moreover, ensuring the supply-demand balance of the energy services is, sometimes, extremely difficult, due to the impossibility to rely on surrounding areas to modulate the offer. At last, the high seasonal variability of the energy demand, typical of small touristic islands, brings to the oversizing of diesel generators, which need to be dimensioned on the summer peak demand, even if demand average value is considerably lower. The energy system is also significantly stressed by water supply operations. In several islands the potable water, which is typically transported with tank vessels from the mainland, is nowadays produced by alternative technologies, which are in some cases able to meet the entire water demand, yet usually consuming large amount of energy. For example, the energy consumption for desalination, which is considered an advanced alternative water supply source, ranges between 7-14 and 2-6 kWh/m3, for thermal and membrane based technologies, respectively. High energy consumption brings to high cost of production, especially when the energy is produced by a costly and inefficient power system. The water cost in these islands varies from 7 to 10 €/m3, about ten times the cost of production on the mainland. Even though energy and water systems provide essential services for these communities, they are strongly unsustainable due to high operational costs and high greenhouse gas (GHG) emissions. In the recent years several studies, mainly focused on the introduction of renewable energy sources (RES), have been developed to improve the sustainability and the efficiency of these systems. The high RES potential, especially wind and solar, is in fact not adequately exploited: about 70% of the Mediterranean islands produce just between 0.7% and 25% of their electricity from RES. The main goal of our research is to develop a Decision-Analytic Framework to identify optimal structural and control solutions to improve the overall sustainability of the integrated water-energy system of small Mediterranean islands. More precisely, on the one hand we focus on the introduction of renewable energy sources (RES) as carbon free technologies for power generation, whereas, on the other hand, we analyse different water supply solutions to improve the efficiency of the whole water supply system, from abstraction to distribution. To achieve this goal, we develop a robust multi-objective optimization approach for the planning and the control of the integrated water-energy system, identifying the most efficient solutions with respect to different sustainability objectives and over a range of possible future scenarios, which consider changes in both climate and socio-economic external drivers. Moreover, the optimization process has to take into account some specific issues related to the high seasonal variability of the energy and water demand, the intermittent nature of RES and the high environmental protection of the majority of the small Mediterranean islands. We apply this approach on a real case study in Italy, namely Ustica. Ustica is a small island of 8 km2 in Sicily. It has a population of 1559 inhabitants, which grows to more than 3000 people during the summer touristic months. The energy is nowadays produced by 5 diesel generators, with a total installed capacity of 4.6 MW, and few private photovoltaic (PV) arrays (RES penetration: 0.5%). From 2015, potable water is obtained from a desalination plant, which is able to cover the entire water demand in every period of the year. Potable water is stored in 5 reservoirs (for a total capacity of 11000 m3) from which the public distribution network originates.We develop a model able to simulate the coupled energy and water supply systems, as well as the links and the feedbacks between them. We use a mass-balance approach to describe the electricity grid dynamics and an equivalent reservoir to represent the storage capacity of the water distribution network. The desalination plant, which constitutes the connection between the energy and the water systems, is implemented defining its maximum capacity in m3/h and a conversion coefficient, which specifies the energy consumed for the production of 1 m3 of potable water. This coefficient is estimated from data and takes into account the effective desalination process and the pumping of the water both from sea to the plant and from the plant to the reservoirs.First, we adopt this model to assess current system operations against different indicators (i.e., cost of energy, GHG emission and water deficit) and then we optimize the system over a medium-long horizon to identify the most efficient energy mix and the optimal management of the water supply system under baseline and different future scenarios.Optimizing the water supply operations could in fact considerably improve the sustainability of the whole water-energy system, especially when we are able to exploit the high RES potential. Desalination is in fact considered as a not conventional way of storing energy in the form of potable water, which can be produced during periods when it is more convenient for the power system (e.g., high RES production), like any other Demand Side Management (DSM) measure. We optimize the system using an Evolutionary Multi-Objective Direct Policy Search (EMODPS) algorithm and we explore different future scenarios, characterized by changes in the climate (e.g., temperature, solar radiation, wind profiles) and socio economic (e.g., water/energy demand, fuel prices) variables. The uncertainty of these variables is therefore considered in the optimization problem in order to generate robust solutions, characterized by satisfactory performances over a wide range of possible future conditions.