Introduction

Over the last few years, energy demand has been on the increase on a global scale. This phenomenon is more intensely observed in remote areas. On Greek islands, in particular, energy is produced in autonomous power stations (APS) through fossil-fuel consumption, which is not an environmentally friendly process. In 2019, 61 Greek islands were proven to be dependent on APS for energy production due to the economic and technical particularities of the region (Groppi et al. 2021). Some other islands are connected to the national grid of the mainland using 150-kW and medium-voltage connections (Karystianos et al. 2021). The main disadvantage of APS is the occurrence of frequent blackouts during high-demand seasons. The solution lies in the exploitation of the potential that renewable energy sources (RES) present and in the use of energy storage methods like pumped storage and hydrogen for better energy management. Energy from RES is generated through natural processes; it is also limitless, economical, and clean but is characterized by stochastic behavior, as it is dependent upon the weather and climatic conditions.

A hybrid renewable energy system (HRES) is defined as a blend of different RES and various energy storage methods (Esmaeilion 2020). An island can achieve energy autonomy through the installation of an HRES on it (Al Katsaprakakis et al. 2019). The use of RES is associated with reductions in CO2 emissions, greenhouse gases, and climate-change factors (Nikolaou et al. 2020) and the prevention of global warming (Ammari et al. 2021). The storage technology of an HRES stores the excess RES energy and uses it to cover the energy demand during periods with low wind or low solar energy potential (Marocco et al. 2021). The goal is to minimize the amount of unexploited RES and to maximize the covered energy demand of the area. HRESs are efficient because, compared to the current situation in Greek islands, which involves transferring water from the mainland, they reduce costs. Additionally, they offer a solution that combines RES with storage, leading to a transition to RES, which is one of the targets for Europe for 2030, while simultaneously meeting both water and energy needs.

HRESs have been studied extensively over the last decades, with the research community focusing on hybrid storage in the last few years. A hybrid-storage micro-grid with the best design, based on renewable energy and hydrogen, is presented by Ancona et al. (2022). The prospects for 100% RES penetration in the Faroe Islands are investigated by Al Katsaprakakis et al. (2019). That research includes energy consumption for heating and transportation purposes. Kaldellis (2020) examines the characteristics of wind-based electricity production by autonomous energy consumers. Attention is drawn in particular to communities residing on remote islands. Roy et al. (2022) investigate power architectures and the topologies of power electronic converters, mathematical modeling, and design techniques for optimization while investigating a wind–solar HRES. An HRES (wind park, photovoltaic (PV) park, and desalination plant) on the island of Karpathos, which is intended to meet the demand for water and electricity on site, is examined by Skroufouta and Baltas (2021). The methodological framework for the minimization of the rejected energy of an HRES is presented by Bertsiou and Baltas (2022b). Al Katsaprakakis and Voumvoulakis (2018) present the technical and financial aspects of an HRES that generates 100% of the power used by the independent island of Sifnos, thereby allowing the island to achieve energy independence.

Kumar et al. (2022) examined the optimum configuration for electrifying five villages—all of which were unelectrified and disconnected from the grid—in Odisha, India, using RES. With the purpose of maintaining an uninterrupted power supply, three different types of battery technologies along with a diesel generator were investigated. A solar PV, diesel generator, wind turbine (WT), and battery-storage off-grid hybrid microgrid have been built for a potential location in Kuakata, Bangladesh, by Hossen et al. (2022). Real-time field data on solar radiation, wind speed, and biomass were used to run the simulation. Rana et al. (2022) present a comparison analysis of the categorical variable decision-tree algorithm, a peak load-shaving approach. This study suggests a suitable peak-saving technique. The optimized design of a standalone HRES is examined by Shezan et al. (2022a). The system consists of solar PV, a backup diesel generator, a WT, and battery storage for two divisional locations in the areas of Dhaka and Khulna in Bangladesh. By using the HOMER software platform, Shezan et al. (2022b) designed and optimized an islanded microgrid for Melville Island. The system consists of a diesel generator, solar PV, battery storage, and WT. Two hybrid power systems are currently in use in Greece (Al Katsaprakakis and Voumvoulakis 2018). The first one is installed in Ikaria, and it consists of a wind park, a storage unit, and thermal generators (backup). The second one, in Tilos, consists of a WT, a PV station, and advanced NaNiCl2 batteries.

Hydroelectric energy is founded on the conversion of the kinetic energy of water into electricity. It is the most common and mature grid-scale energy storage technology (Walker and Duquette 2022). In the case of a reservoir, the term “exploitation” refers to the available volume of water that is stored. The potential energy of the water due to the height of the upper reservoir is changed into kinetic energy and then into electricity using a hydro turbine (Blakers et al. 2021). In 2020, hydropower plants produced around 20% of the electricity consumed globally (Tzoraki 2020). The best size of pumped storage hydropower stations coupled with grid-connected solar and wind energy to minimize the levelized cost of energy (LCOE) is proposed by Bhimaraju et al. (2022).

Another energy storage method is hydrogen tanks. In a period of low energy demand, the extra energy can be used to create green hydrogen through the process of water electrolysis (Yue et al. 2021). Wind and solar energy are considered the best-suited energy sources for hydrogen production (Kovač et al. 2021). The hydrogen is separated through a chemical solution and stored in a tank (Guo et al. 2019). In a high-energy-demand period, this energy can be released by using hydrogen as fuel in a combustion engine or a fuel cell (FC) (Maggio et al. 2019). Energy production through hydrogen FCs leads to the emission of only liquid water or water vapor or a combination of these two and warm air into the atmosphere, producing no harmful tailpipe emissions (Kovač et al. 2021). Izadi et al. (2022) examine a transient simulation of a HRES consisting of PV panels, WTs, an electrolyzer, a hydrogen tank, a fuel cell, and controller devices.

Desalination industry research has recently concentrated on the use of RES in conjunction with desalination facilities to lower energy costs as well as the cost of the desalinated water produced (Kourtis et al. 2019). This method gives no emissions as a by-product. Reverse osmosis (RO) is the most suitable desalination technique due to its low energy consumption and associated costs (Esmaeilion 2020), its suitability for connection with RES, and the large-scale production of high quantities of desalinated water (Al-Karaghouli and Kazmerski 2013). Okampo and Nwulu (2021) suggest using optimization techniques to combine the system's operation, size, and thermodynamic impacts in the analysis of RES-RO systems. Sarathe et al. (2022) analyze the cost involved in RES-powered desalination technologies used all over the world to produce freshwater. Agapitidou et al. (2022) analyze an HRES on non-interconnected Lemnos Island, comparing pumped and hydrogen storage to meet water and energy needs.

The novelty of this study in the field of HRESs is the combination of two different energy storage technologies, namely pumped-storage hydropower and hydrogen storage. In hybrid energy storage, wind energy can be stored both as hydraulic energy and as hydrogen. Data on the population and weather are used to create a methodological framework. An energy management methodology is proposed for the energy produced by the WTs and the energy stored by different storage methods to cover the potable water and electricity demand. Also, a desalination unit is used to produce potable water. These data can be used to modify this framework when studying other remote islands. The outcomes provide important insight into the water and energy management of HRESs in remote areas.

Materials and methods

Study area: Skyros

A case study for the implementation of the suggested technology was conducted on the island of Skyros in the Aegean Sea. This is the southernmost and the largest island of the Sporades archipelago and is located along the east coast of Evia at a distance of 30 km. The island has a total area of 220.45 km2, a maximum length of 29 km, a minimum width of 3 km, and a coastline length of 134 km. The maximum altitude (792 m) of the island is located on Mount Kochilas (Triantis et al. 2005). Based on data from the census of 2011, there are 2994 inhabitants, a number that reaches 12,000 during the summer months due to tourism according to the Hellenic Statistical Authority. Most of the population resides in the island's capital, Chora.

The island enjoys a Mediterranean climate, with moderate, rainy winters and warm, sunny summers, as a result of its location and the impact of the sea. The average annual temperature during the years 2011–2020 was 18.5 °C. The monthly average minimum is 10.5 °C (January) and the mean monthly average maximum is 27.4 °C (July). The total annual rainfall is 491 mm, and the prevailing winds are mainly northerly and around 3–4 Beaufort. This information is gathered from the Skyros weather station at the National Observatory of Athens (Lagouvardos et al. 2017). According to the Database for the Natural Environment of Greece—FILOTIS (2022), there are nature reserves and Corine areas in Skyros.

Water in Skyros is not suitable for human consumption, as it comes from brackish water sources (mainly from the Anavallousa source), and residents consume bottled water. Additionally, transporting water by ship has a high cost, up to 12 €/m3 (Bardis et al. 2020). A potential solution to this issue is the desalination of seawater (Kyriakarakos et al. 2022). The island has a local APS for its electricity needs (Petrakopoulou et al. 2016). The mean monthly demand for water supply and the mean monthly electricity demand are shown in Fig. 1. Data refer to the years 2011–2020. The assessment of the water demand of the inhabitants of Skyros is carried out by considering individual water supply needs. Specifically, the daily water consumption per person is calculated at 200 L/day/capita for permanent residents and 300 L/day/capita for tourists (Koumoura et al. 2018). The energy consumption data are provided by the Hellenic Electricity Distribution Network Operator (HEDNO). The load profile of the island is characterized by a quite variable distribution. During the summer months (July and August), due to tourism, the demand for water and electricity is increased (Trikalitis et al. 2021).

Fig. 1
figure 1

Mean monthly demand for water supply (blue) and mean monthly electricity demand (black)

Full size image

Methodology

System

The installation of an HRES on the island of Skyros aims to satisfy the water and electricity demand. This system provides a solution to the lack of local potable water and to the instability and insecurity of the local power grid. The system consists of an onshore wind park with three 2350-kW WTs (ENERCON E-82), a desalination unit that provides 3500 m3/day, a 4-MW pumping station (four pumps: 2 × 0.5 MW, 1 × 2 MW, and 1 × 3 MW) for pumping seawater, a 4.7-MW hydroelectric station, a water reservoir with a capacity of 200,000 m3 at a height of 300 m above sea level, a hydrogen production station that generates 1000 kg/day, and two tanks to store the hydrogen (capacities: 2000 kg and 3000 kg). The sizes of the water reservoir and the hydrogen tanks ensure that the island can be autonomous for 3 days; the ability to provide 2–3 days of autonomy is common for storage systems in HRESs (Bhandari et al. 2015), as this avoids over-dimensioning.

Wind energy (Ew) is generated by three WTs. Of the energy produced, 30% is sent to the electrical grid to satisfy electricity needs (Eed) and to minimize the rejected energy (Bertsiou and Baltas 2022a). The remaining energy is then used for desalination (Ed) to fulfill the water requirements (Vwd). This desalination process involves using seawater (Vsw), which is stored in a tank (Vd). If there is surplus energy after desalination (Ew − Ed), it is primarily used to meet any outstanding electricity demands and secondarily used for powering water pumps (Ep) and producing hydrogen (Ehp). This energy is stored in a water reservoir (Vwr) and a hydrogen tank (Mht), respectively. When there is no stored energy available, a hydro turbine and/or a fuel cell generates the required energy to meet any unmet demand.

The amount of wind energy produced depends on the wind speed and the WT swept area (Cupples et al. 2021). To calculate the energy, the power curve that was provided by the company for this model was used. The WTs and the weather station are placed at different heights. The equation that helps to calculate the wind speed at the hub height using the speed at the weather station is as follows (Yang et al. 2018):

$$V/V_{0} = \left( {y/y_{0} } \right)^{n} ,$$
(1)

where V (m/s) is the wind speed at height y (808 m), V0 (m/s) is the wind speed at height y0 (27 m), and n is the Manning coefficient, which takes different values depending on the natural surface of the installation. In the case of Mount Kochilas, the value of n is 0.20. The equation that calculates the power of a hydro turbine is as follows (Acharya et al. 2015):

$$P = \rho \cdot g \cdot Q \cdot H \cdot n,$$
(2)

where ρ is the density of water (1000 kg/m3), g is the acceleration of gravity (9.81 m/s2), Q (m3/s) is the water supply, H (m) is the total height, n is the output coefficient of the hydro turbine (0.85), and P is expressed in W (watts). The reservoir's altitude and the amount of water that is accessible determine how much electricity is produced. The type of hydro turbine selected is the Pelton (Hatata et al. 2019).

The energy requirements for RO desalination vary between 2.5 and 7.0 kWh/m3 (Fornarelli et al. 2018) and depend on the quality of the water and the size of the RO system (Bertsiou and Baltas 2022a). In this case, it is considered to be 5.0 kWh/m3 when utilizing the seawater that surrounds the island.

Proton Exchange Membrane (PEM) electrolysis is used to produce green hydrogen (Zeng and Zhang 2010) utilizing the excess wind energy from the WTs. The production of 1 kg of hydrogen requires 9 kg of clean water and 0.06 MWh of electrical energy (Rievaj et al. 2019). The heating value of hydrogen is 33.4 kWh/kg (at a temperature of 25 °C and pressure of 1 atm). The efficiency of a PEM fuel cell is 47.6% (Taner 2018). Finally, 1 kg of hydrogen produces 15.9 kWh of energy. The hydrogen is stored as gas in tanks.

Scenarios

Three scenarios have been analyzed to investigate the response of the HRES. The model simulation uses 88,081 hourly time steps (10 years). The input data inserted into the simulation are the wind speeds at the time steps. Each wind speed is converted to the wind energy by applying Enercon's power curve for the WT (ENERCON 2015). The equation that converts wind speed into wind energy arises as a numerical approximation of the power curve and is as follows:

$$E = ( - 0.001018 \cdot V^{6} + 0.074914 \cdot V^{5} - 1.971453 \cdot V^{4} + 21.102244 \cdot V^{3} - 66.803471 \cdot V^{2} + 49.99868 \cdot V + 11.189772)/1000,$$
(3)

where V (m/s) is the wind speed at the height of the WTs and E is expressed in MWh. The three scenarios correspond to different usages of the produced wind energy.

Three different scenarios were selected, all of which focused on managing the water and energy demand after analyzing the data for the specific area. The energy demand concerns the entire island and includes energy for homes, energy for desalination, water pumping, and electrolysis. Each is analyzed through the prism of optimum coverage. In all three scenarios, WTs electrify the system: 30% of the energy is immediately directed to the power grid to cover the electrical demand. The remaining 70% of the generated wind power may be used for desalination, pumping, and hydrogen production. In all three scenarios, the primary purpose of the WTs is to desalinate the abundant seawater that surrounds the island of Skyros, aiming to meet the water demand. When the wind energy production exceeds the immediate need, the surplus is allocated for additional desalination. This extra desalinated water is stored in tanks for future use. Furthermore, the stored desalinated seawater can also be utilized for hydrogen production. The excess amount of energy after the desalination is used to cover the electricity demand. The difference between the three scenarios is based on the storage method for the excess energy. This excess energy can be used during peaks in water and electricity demand. The output of the simulation refers to the water and electricity coverage percentages. The three storage methods are examined to investigate which one is the most efficient.

The first scenario only depends on pumped-storage hydroelectricity technology. If there is a lack of energy produced by the WTs, water is released from the upper reservoir into the sea, and the hydro turbine produces hydroelectric energy to cover the electricity demand. In the case of an excess in energy, this is used to pump water into the reservoir. The equation that calculates the volume of the water that is pumped into the reservoir is as follows:

$$V = \left( {E \cdot n \cdot 10^{6} \cdot 3600} \right)/\left( {\rho \cdot g \cdot H} \right),$$
(4)

where E (MWh) is the energy for pumping, n is the output coefficient of the pumps (0.80), ρ is the density of water (1000 kg/m3), g is the acceleration due to gravity (9.81 m/s2), H (m) is the total height, and V is expressed in m3.

The second scenario uses hydrogen storage technology. If there is a lack of energy, electrical energy is produced from hydrogen through the fuel cells. If there is an excess, it is used to produce hydrogen in a gaseous form through the process of PEM electrolysis, which is then stored in hydrogen tanks.

The third scenario is a combination of the previous two. If there is a lack of energy, firstly, the hydro turbine produces hydroelectric energy to cover the electricity demand. If more energy is required, electrical energy is produced from the hydrogen. In the case of an excess in energy, this is used to pump water into the reservoir; if there is an even greater excess, it is used to produce hydrogen. The configuration of the system and a flow chart for the three scenarios are shown in Fig. 2a and b, respectively.

Fig. 2
figure 2

Configuration of the system (a) and flow chart for the three scenarios (b)

Full size image

Table 1 depicts all the different technical characteristics of the systems employed in each scenario.

Table 1 Technical characteristics of the systems employed in each scenario
Full size table

The output data from the simulation are the amount of potable water per hour and the system coverage percentage in each scenario for the island’s energy demand per hour.

Results and discussion

Wind energy produced

The usage of the wind energy produced by the WTs in the three scenarios throughout the year is shown in Fig. 3. In all scenarios, 30% of the overall wind energy production is directed to the power grid to cover the electricity demand, 12% is used for desalination, and 11% is used directly to meet the electricity demand. The remaining 57% of the wind energy produced is treated differently in each of the three scenarios. In the first scenario, 25% is used for pumping and 22% is not usable because either the reservoir is full or there is an insufficient amount of energy for the water pumps to function. In the second scenario, 37% is used for hydrogen production and 10% is not usable due to the limited capacity of the hydrogen tank. In the third scenario, 25% is used for pumping, 13% is used for hydrogen production, and 9% is not usable.

Fig. 3
figure 3

Usage of the wind energy produced by the WTs in the three scenarios throughout the year

Full size image

These percentages of the wind energy produced arise from the simulation. Only the 30% provided to the power grid is set before the simulation. The percentage of desalination (12%) arises from the water demand. This percentage corresponds to the ratio of the energy for desalination to the total wind energy produced (Ed/Ew). The energy for desalination is equivalent to the volume of the water demand. The percentage of direct cover (11%) arises from the electricity demand and corresponds to the quantity of wind energy required to cover the electricity demand in the specific time step ((Ew − Ed)/Eed). The rest (57%) of the produced energy needs to be stored. Because of the energy storage limitations, there are different quantities of energy that cannot be stored in the three scenarios. For the first scenario, 25% is used for pumping, which corresponds to the ratio Ep/Ew. For the second scenario, 37% is used for hydrogen production, which corresponds to the ratio Ehp/Ew.

Desalination

As mentioned above, in all three scenarios, the first process is desalination. In all scenarios, there is total coverage of the water demand due to the desalinated water tank. The total quantity of desalinated water produced every year is 504,260 m3. August is the month with the highest desalination production (102,472 m3), due to tourism. The minimum desalination production is observed in December (19,212 m3). The total amount of electrical energy required for desalination per year is 2521 MWh. The maximum quantity is observed in August (512 MWh) and the minimum in December (96 MWh).

Scenarios

First scenario

The first scenario (pumped-storage hydropower) covers 88% of the entire island’s electricity demand throughout the year; 64% comes directly from the WTs and 22% is hydroelectric energy. The mean monthly coverage of the electricity demand is shown in Fig. 4a. Every year, a total amount of 9485 MWh is covered by the WTs, 3337 MWh is covered by hydroelectric energy, and 2114 MWh is not covered. The maximum non-covered amount of electrical energy is observed in June (250 MWh) and the minimum in January (124 MWh).

Fig. 4
figure 4

Mean monthly coverage of the electricity demand (a) and the monthly release of water from the reservoir as well as the water that must be released to cover the electricity demand (b) for the first scenario

Full size image

Every year, a total amount of 3,368,505 m3 of water is released from the reservoir to produce hydroelectric energy from the hydro turbine. The total volume required to cover all of the electricity demand every year is 4,476,309 m3. Figure 4b depicts the amount of water released from the reservoir per month and the total amount of water that must be released to cover the electricity demand. The maximum quantity of released water is observed in August (439,755 m3) and the minimum in May (129,112 m3).

Second scenario

The second scenario (hydrogen storage) covers 83% of the entire island’s electricity demand throughout the year; 64% comes directly from the WTs and 17% from the hydrogen. The mean monthly coverage of the electricity demand is shown in Fig. 5a. Every year, a total amount of 9485 MWh is covered by the WTs, 2603 MWh is covered by the hydrogen, and 2848 MWh is not covered. The maximum amount of non-covered electrical energy is observed in June (385 MWh) and the minimum in February (136 MWh).

Fig. 5
figure 5

Mean monthly coverage of electricity demand (a) and the hydrogen produced monthly as well as the hydrogen needed to cover the electricity demand (b) for the second scenario

Full size image

Every year, 164,036 kg of hydrogen is produced. To cover the electricity demand, a total amount of 342,860 kg of hydrogen is needed. In Fig. 5b, the amount of hydrogen produced per month and the total amount of hydrogen required to cover the electricity demand are shown. The maximum amount of hydrogen produced is observed in March (16,086 kg) and the minimum in June (11,208 kg).

Third scenario

The third scenario (hybrid storage) covers 95% of the entire island’s electricity demand throughout the year; 64% comes directly from the WTs, 22% is hydroelectric energy, and 7% is from hydrogen. The mean monthly coverage of the electricity demand is shown in Fig. 6a. Every year, a total amount of 9485 MWh is covered by the WTs, 3337 MWh is covered by hydroelectric energy, 1060 MWh is from hydrogen, and 1055 MWh is not covered. The maximum amount of electrical energy that is not covered is observed in June (385 MWh) and the minimum in March (36 MWh).

Fig. 6
figure 6

Mean monthly coverage of the electricity demand (a) and the hydrogen produced monthly as well as the hydrogen needed to cover the electricity demand (b) for the third scenario

Full size image

In the third scenario, the second method for the storage of the excess energy is via hydrogen. Every year, 168,657 kg of hydrogen is produced. To cover the electricity demand, a total amount of 132,974 kg of hydrogen is needed. In Fig. 6b, the amount of hydrogen produced per month and the amount of extra hydrogen that is needed to cover the electricity demand are presented. The maximum amount of hydrogen produced is observed in February (18,973 kg) and the minimum in June (6857 kg).

Comparison of the three scenarios

The island's energy mix includes three main energy sources: WTs, pumped hydro-storage, and hydrogen technology. Figure 7a presents a comparison of the three scenarios, depicting the mean monthly demand for electrical energy, the coverage from the three scenarios, and the coverage from the WTs. The coverage from the third scenario is higher compared to those from the other two scenarios and closer to the demand. The coverage from the first scenario is higher compared to the coverage from the second scenario during the summer and lower during the winter. The coverage from WTs is the lowest in all three scenarios.

Fig. 7
figure 7

Mean monthly demand for electricity energy, the coverage from the three scenarios, and the coverage from the WTs (a) and the hourly coverage from the three scenarios (b) throughout the year

Full size image

Figure 7b presents a comparison of the three scenarios, depicting the hourly electrical energy coverage of the three scenarios throughout a mean year. In the first scenario, total coverage of electrical demand is achieved in 76% of the hours on an annual basis, whereas zero coverage occurs in 8% of the hours. The corresponding percentages for the second scenario are 77% and 12%, respectively, and those for the third one are 89% and 4%, respectively. These percentages explain the different behaviors of pumped-storage hydroelectricity, hydrogen storage technologies, and a combination of them.

Cost–benefit analysis

The total amount of the investment is equal to 11,904,000 € for the first scenario, 8,804,000 € for the second scenario, and 13,206,000 € for the third scenario. The repayment period is 10 years and the interest rate is 3%. Table 2 depicts the estimated supply and equipment costs for the three scenarios, and Table 3 depicts the estimated annual operating and maintenance costs for the three scenarios. It must be mentioned that the supply and equipment costs and annual operating and maintenance costs were selected by following the guidelines provided by previous authors (Xu et al. 2021; Hu et al. 2021).

Table 2 Estimated supply and equipment costs for the three scenarios
Full size table
Table 3 Estimated annual operating and maintenance costs for the three scenarios
Full size table

The NPV of the investment is set to zero for a specific desalinated water selling price and a specific selling price for the electrical energy for each scenario. The selling price of desalinated water is 1.00 €/m3, as the same quantity of desalinated water is produced in scenarios 1, 2, and 3 (503,385 m3 per year). The selling price of electrical energy varies in each scenario, as different quantities of electrical energy are produced. For the first scenario, 12,822,188 kWh is produced per year, and the price corresponds to 0.113 €/kWh; for the second scenario, 12,088,107 kWh is produced, and the corresponding price is equal to 0.097 €/kWh; for the third scenario, 13,881,484 kWh is produced, and the selling price is calculated is equal to 0.134 €/kWh.

In 2019, the selling price of water for the households of Skyros was 0.20 €/m3 for the first 20 m3 of use and 0.45 €/m3 for the rest m3 of the consumption. Skyros is electrified through a local APS (isolated grid), so the price of electricity is heavily subsidized. Based on the latest statistics (April 2022) provided by HEDNO, the average full cost of fossil-fuel-based power stations was 0.52757 €/kWh, which is around 5 times higher than the cost of the HRES calculated in this model. HRESs are a solution to reduce current subsidies, as the cost is more fixed.

Conclusions

This paper presents the development of an HRES on Skyros Island for water and energy independence. Three different scenarios are investigated, based on different energy storage methods. The use of pumped-storage hydroelectricity is the only factor that acts as a storage technology in the first scenario. The second scenario uses hydrogen storage technology. Finally, the third scenario is a hybrid storage technology based on pumped-storage hydroelectricity and hydrogen storage.

In all scenarios, there is total coverage of the water demand, due to the existence of a reservoir for desalinated water. Regarding the need for electricity, the third scenario offers the highest coverage of the demand: 95% throughout the year. The first scenario appears to be the second most effective, covering 88% of the electricity demand, while the lowest coverage is observed in the second scenario, in which 83% of the demand is covered. Hourly results show the different behaviors of the different storage methods in each scenario. In the third scenario, there is coverage of the energy demand in 89% of the 8766 hours throughout the year, followed by the second scenario (77%) and finally by the first scenario (76%).

According to the cost–benefit analysis, in all scenarios, the selling price of desalinated water is 1.00 €/m3, as the same quantity of desalinated water is produced. Regarding the selling price of electrical energy, the lowest-priced scenario is the second one, at 0.095 €/kWh, followed by the first scenario, at 0.113 €/kWh, and the third scenario, at 0.134 €/kWh.

Comparing the two first scenarios, it appears that during the summer months, there is higher energy coverage when the method of pumped storage is used. During the winter months, the opposite behavior is observed. Based on the results, in the first scenario, the ratio of the hydroelectric energy produced to the wind energy available for pumping is 68%, while in the second scenario, the ratio of the energy produced from hydrogen to the energy consumed to produce hydrogen is 27%. For the hybrid storage (third scenario), the energy efficiency is 49%. Consequently, in terms of energy losses, the method of pumped storage (first scenario) is the approach that performs the best.

Based on the results obtained and conclusions drawn after the completion of this research paper, some topics that deserve future research have emerged. This methodology could be applied in other regions of Greece, especially islands with no access to the national electrical grid. The approach used in this research work can be transferred to any other part of the world, taking into account the specific conditions of the area for water and energy autonomy. Moreover, the use of different forms of stored hydrogen, such as liquid hydrogen, could be examined.