Dynamic Investigation and Optimization of a Solar-Based Unit for Power and Green Hydrogen Production: A Case Study of the Greek Island, Kythnos

Abstract

The aim of the present work is the analysis of a solar-driven unit that is located on the non-interconnected island of Kythnos, Greece, that can produce electricity and green hydrogen. More specifically, solar energy is exploited by parabolic trough collectors, and the produced heat is stored in a thermal energy storage tank. Additionally, an organic Rankine unit is incorporated to generate electricity, which contributes to covering the island’s demand in a clean and renewable way. When the power cannot be absorbed by the local grid, it can be provided to a water electrolyzer; therefore, the excess electricity is stored in the form of hydrogen. The produced hydrogen amount is compressed, afterward stored in tanks, and then finally can be utilized as a fuel to meet other important needs, such as powering vehicles or ferries. The installation is simulated parametrically and optimized on dynamic conditions, in terms of energy, exergy, and finance. According to the results, considering a base electrical load of 75 kW, the annual energy and exergy efficiencies are found at 14.52% and 15.48%, respectively, while the payback period of the system is determined at 6.73 years and the net present value is equal to EUR 1,073,384.

1. Introduction

In recent decades, the rapid increase in the world’s population, the social, economic, and industrial growth as well as urbanization, have significantly amplified the energy needs, the utilization of fossil fuels, and greenhouse gas emissions [1]. These facts have affected negatively not only the natural environment and the atmosphere, but they also can cause health problems for human beings; therefore, the international community has decided to act via international environmental policies such as the Kyoto Protocol (1997), and the Paris agreement (2015). These agreements focus on the avoidance of environmental damage and global warming, as well as the replacement of non-renewable energy sources [2]. Human needs can be covered through cleaner, more sustainable, and energy-efficient ways [1], using renewable energy sources such as biomass, wind, or solar [3].

Solar energy is an environmentally friendly source with very high potential. Many different technologies and devices have been developed to exploit solar irradiation into electricity or thermal energy. Photovoltaic (PV) cells can transform solar energy directly into electricity, while solar thermal collectors produce useful heat. Solar collectors can be utilized for small-scale applications, such as space and water heating, as well as in large-scale installations, to provide industrial heat or drive power plants [4]. The most developed and utilized solar technology is the parabolic trough collector (PTC). A PTC exploits solar beam irradiation and includes a parabolic mirror, as well as a tubular receiver, in which the heat transfer medium absorbs heat and its temperature increases. In parallel, the collectors track the Sun during the day and are incorporated with thermal energy storage (TES) units, to improve their efficiency. Moreover, PTCs operate at low- and medium-temperature levels, with values up to 400 °C; therefore, these devices can be used for industrial heating applications and can properly feed electricity generation systems [5], such as organic Rankine cycles (ORC), or Brayton cycles [6].

First of all, Yu et al. [7] investigated a solar-powered ORC unit, which consisted of a PTC field, as well as a hot, and cold TES tank. In the case of using toluene as a working medium, the most proper performance was achieved, as the overall efficiency was enhanced by 24.8%, in comparison with another study in the literature. The thermal and exergy efficiencies were also increased by 11.3% and 10.8%, respectively, for the case of using a supercritical ORC, compared with the corresponding subcritical cycle. In addition, Arslan and Kilic [8] studied an ORC plant fed by PTCs, which was supposed to be installed in a low-solar radiation zone. A hot and cold storage tank was also integrated. Six eco-friendly organic refrigerants, as well as water steam, were investigated. According to the results of dynamic analysis, the optimum and the most economically viable design was the case of the water-steam cycle, when the net present value was calculated at USD 9.012 × 10⁶. The maximum rates of energy and exergy performance were determined as 11.05% and 11.86%, respectively. Moreover, configurations based on the combination of a steam Rankine cycle and an ORC were analyzed in publications. For example, Bahari et al. [9] investigated a PTC-based power generation unit that included a two-stage steam Rankine cycle and an ORC as a bottoming cycle. Molten salt was used as a heat transfer and storage fluid in the incorporated tank. The configuration was optimized via the particle swarm optimization (PSO) algorithm considering the maximum exergy efficiency and the minimum levelized cost of electricity (LCOE). At the optimum scenario, the overall exergy efficiency, and the levelized cost of electricity were found at 63.89% and 0.1529 USD/kWh, respectively. Furthermore, Li et al. [10] studied a solar-driven electricity plant that contained a PTC field, a storage unit based on a phase change material (PCM), a steam Rankine cycle, and an ORC. The system was modeled and optimized in off-design conditions taking into account different isentropic efficiencies of the Rankine cycle screw expander when the pressure ratio was varied. The maximum achieved values of the solar thermal power efficiency were calculated as 13.74–15.45%, with an expander volume ratio that was equal to five. Finally, Wang et al. [11] analyzed a recompression supercritical CO₂ Brayton cycle, that was used to generate power and was powered by a PTC. A molten salt TES unit was also included. The system performance was investigated for different design parameters, as well as weather and irradiation conditions. The cycle energy efficiency could be found greater than 40%, in optimal conditions.

However, solar irradiation is not always available and is varied during the day and the year. Production intermittency is the main drawback of solar energy; therefore, it is important to integrate a TES system. This unit can store useful heat during the sunshine, which can be used when solar irradiation is limited or there is no irradiation. The most common and mature technology is sensible TES [12]. Most of the time, solar-based plants include one or more storage tanks, to store heat during the day and release it to continue their operation after sunset; therefore, insulated tanks are capable of short-term TES for a couple of hours [13,14]. In parallel, several other forms of energy storage have been developed to store surplus electricity during off-peak hours. Because of the further integration of renewable sources in the energy mix, it is essential to use methods for long-term or seasonal storage, such as the technologies of pumped hydro energy storage (PHES), compressed air energy storage (CAES), and thermochemical energy storage [15,16]. In addition, hydrogen has gained further attention in recent years as a way of seasonal energy storage [17]. Because of the fact that hydrogen is not directly available in nature, it can be produced from fossil fuels via steam methane reforming or renewable energy sources via water electrolysis. The water electrolyzer is fed with excess renewable electricity, to produce green hydrogen without carbon emissions. When it is reacted with oxygen, the only by-product is water; therefore, hydrogen is a sustainable energy carrier that can be used in transportation, power generation, buildings, and industry [18]. Hydrogen can be stored as a compressed gas (compressed hydrogen), as a liquid (liquified hydrogen), as well as in solid or liquid storage materials [19]. Thus, hydrogen energy storage promotes the sustainability and autonomy of the energy systems [20].

Plenty of studies based on solar-fed power plants or multigeneration units have been published by researchers that contain hydrogen as a product. These kinds of systems are more energy-efficient compared to single-production units [21]. At first, Atiz et al. [22] investigated a solar-geothermal-driven configuration that generated electricity and hydrogen. This system was made up of a PTC field, an ORC, a proton exchange membrane (PEM) electrolyzer, and a cooling tower for the ORC condenser. According to the results, the energy and exergy efficiency was found to be 5.85% and 8.27%, respectively, while the average electricity production was determined at 66.02 kW between the hours of 11:00–13:00, and the daily hydrogen production amount was 9807.1 g. Moreover, Mahmood et al. [23] analyzed a polygeneration system installed in a greenhouse that produced electricity, fresh water, and space cooling. The system contained a PTC, a TES unit, an ORC, an absorption cooling cycle, a desalination unit, and a PEM water electrolyzer. The produced hydrogen and oxygen amounts could feed a hydrogen-oxy combustor if there was no solar irradiation. The plant energy and exergy efficiencies were determined to be 41.0% and 28.4%, respectively, while the hydrogen production rate was determined at 0.01 kg/s. Furthermore, Tukenmez et al. [24] proposed a multigeneration configuration that provided electricity, heating, cooling, drying, hydrogen, hot water, and freshwater. Consequently, the unit included a PTC field, a distillation plant, a PEM electrolyzer, a hydrogen compression system, a Kalina cycle, an ORC, an ejector cooling cycle, a domestic water heater, a dryer, as well as a hot and a cold storage tank. The overall energy efficiency, the exergy efficiency, and the hydrogen production rate were calculated to be equal to 59.34%, 56.51%, and 0.0043 kg/s, respectively. In parallel, the integration of hydrogen into the energy networks of islands and remote communities has also been investigated. More specifically, Katsaprakakis et al. [25] proposed the installation of a PHES system and a wind park, which would work together with the existing PV panels and wind turbines, on the Greek island of Sifnos. The fundamental purpose of the whole project was the coverage of the entire electricity demand by the previously mentioned modules, and the energy independence of the island. The surplus electricity was used to produce potable water and hydrogen. Then, the produced hydrogen would power the passenger ships connecting Sifnos with the nearby islands.

According to the previous literature review, the majority of solar-based power plants and polygeneration systems, include storage tanks to store the excess thermal energy that comes from solar irradiation. This method can store energy only for a couple of hours. Hydrogen energy storage is a long-term method, as hydrogen can be stored and used as a fuel with numerous applications. The present study is an innovative one because two storage technologies are incorporated simultaneously, the solar TES tank and chemical energy storage in the form of compressed hydrogen. Moreover, there is a lack of studies that focus on the hydrogen storage capacity of an energy plant powered by solar energy. The surplus electricity produced by the solar ORC unit is fed to a water electrolyzer to produce hydrogen for later use after several days and even months.

The examined configuration is planned to be installed on the Greek island of Kythnos, close to Athens. This island is not interconnected to the national grid; therefore, it is important to cover the electricity demand with its own sources and to store energy for later use. The proposed plant consists of a few PTC modules that absorb the solar beam irradiation to provide useful heat as an input to the ORC module. Furthermore, the installation contains a TES tank, a PEM water electrolyzer, a hydrogen compressor, and a hydrogen storage tank. A share of the ORC power generation feeds both the electrolyzer and the hydrogen compression system, while the remaining electricity can cover the island’s electricity demand. The other useful output is compressed hydrogen fuel, which can be sold or used on-site to power fuel cell vehicles or vessels. Therefore, this small-scale solar-powered energy unit provides electricity and green hydrogen while the integration of two storage technologies increases the system’s dispatchability.

At first, a preliminary analysis is conducted, and then the system is analyzed and optimized under dynamic conditions. The whole configuration is evaluated by calculating the main energetic, exergetic, and economic indexes. Finally, the preliminary analysis is performed by an Engineering Equation Solver (EES) software code, while the dynamic simulation is carried out using Matlab.

2. Material and Methods

2.1. Installation Outline

The proposed energy plant is powered by solar irradiation, which is a totally renewable energy source with great potential. This source has not been exploited to the maximum extent possible, thus, many efforts have been directed by scientists and researchers in this field. The most common and commercially available solar thermal collectors are PTCs. They consist of a parabolic mirror that concentrates solar rays into a tubular receiver, in which a fluid is heated, as it flows. Therminol VP-1 is a widely-used heat transfer medium for temperature values up to 400 °C [26]. Taking into account the requirements of this application, PTCs are the most suitable choice, while the aforementioned thermal oil is selected as the heat transfer and storage fluid.

At first, the thermal oil enters the solar collectors’ field with temperature (T_col,in), where it is heated up and exits with temperature (T_col,out). Then, the fluid enters a TES tank, where the absorbed heat is stored to be used later after sunset when solar irradiation is not available. In parallel, the thermal oil with temperature (T_s,in) leaves the upper part of the storage tank, as the upper tank zones achieve higher temperature levels, and then enters the heat recovery system to power an ORC. The heat recovery system is properly designed aiming to transfer the heat from the thermal oil to the organic fluid, which is finally vaporized. The ORC also includes a recuperator, which is located after the expander outlet to exploit the high-temperature value of the organic medium at this point. The role of the recuperator is to exploit the waste heat after the expander device and, therefore, to increase the thermodynamic efficiency of the power cycle. Practically, the stream from the expander outlet warms up the organic liquid before entering the heat recovery system so that less heat input is needed from the solar system.

The electricity produced by the ORC module may not be able to be provided to the island’s grid because the demand is low, or covered by other power plants. In these cases, the electricity can be provided as an input to the PEM water electrolyzer. Moreover, the incoming water must be preheated. This process is achieved by the thermal oil entering a water preheater with temperature (T_s,out), after the ORC heat recovery system; Therefore, the Therminol enters the storage tank with temperature (T_s,out,2). This device can produce a pure hydrogen amount, which after compression and cooling, is stored in a hydrogen storage tank. Then, it can be used as fuel for vehicles and ships, as well as introduced to the gas network, or converted again into electricity through electrochemical reactions in a fuel cell. Thus, the hydrogen potential as a long-term storage medium is exploited in the present configuration. The whole configuration is illustrated in Figure 1.

2.2. Solar Plant Modeling

2.2.1. Solar Collector Modeling

The parabolic collectors LS-2 were selected to be used in the present installation. The module aperture was about 39 m², while the concentration ratio was around 23 suns [27]. Other geometrical and optical parameters are included in Ref. [27]. These collectors utilize direct beam irradiation ($G_{bn}$), thus, the available solar energy ($Q_{solar}$) in the installed aperture ($A_a$), is determined as [27]:

$$Q_{solar}=A_a·G_{bn}$$

(1)

The useful heat absorbed by the thermal oil is described as [27]:

$$Q_u={\dot{m}}_{col}·c_{p,oil}·\left(T_{col,out}-T_{col,in}\right)$$

(2)

The mass flow rate of the heat transfer fluid is estimated below [28]:

$${\dot{m}}_{col}=0.02·A_a$$

(3)

The PTC thermal efficiency is calculated as the fraction of useful energy to the available solar energy as below [27]:

$${\eta }_{th,col}=\frac{Q_u}{Q_{solar}}$$

(4)

The incident angle modifier (K(θ)) of these collectors is defined by the following expression [27]:

$${\rm K}\left(\theta \right)=\cos \left(\theta \right)+0.000884·\theta -0.00005369·{\theta }^2$$

(5)

In Equation (5), the incident angle (θ) is in degrees. Apart from the aforementioned basic mathematical equations, the collector performance can be extensively determined by a series of additional expressions that describe the heat convection inside the tube, as well as the thermal losses to the ambient. This procedure can be found in Ref. [29]. Thus, the thermal performance of the collector can be described by an equation based on the following structure [30]:

$${\eta }_{th,col}=a·K\left(\theta \right)+b·\frac{T_{col,in}-T_{amb}}{G_{bn}}+c·\frac{{\left(T_{col,in}-T_{amb}\right)}^2}{G_{bn}}+d·\frac{{\left(T_{col,in}-T_{amb}\right)}^3}{G_{bn}}$$

(6)

The coefficients “a”, “b”, “c”, and “d” are defined by considering the results of the detailed thermal analysis.

2.2.2. Thermal Energy Storage Tank Modeling

The TES tank is used as a short-term storage solution for the produced heat during the day. The thermal energy stored in the tank is equal to the heat input to the tank, minus the heat streams rejected from the tank, which are the heat input to the ORC module, the heat needed to preheat the incoming water to the electrolyzer, and the heat losses to the ambient. The tank can be split into thermal zones, where the temperature value is considered to be uniform. If it is supposed that the tank is separated into N zones, the upper zone, where the temperature is higher and the density is lower, is called Zone “1”, while the lower zone is called Zone “N”. Carrying out the energy balance in each mixing zone, the following expressions are made for Zone “1”, Zone “2” (intermediate zone), and Zone “N” [14]:

$$\frac{{\rho }_{oil}{V}_{tank}}{N}{c}_{p,oil}\frac{\partial T_{st,1}}{\partial t}={\dot{m}}_{col}{c}_{p,oil}\left(T_{col,out}-T_{st,1}\right)+{\dot{m}}_s{c}_{p,oil}\left(T_{st,2}-T_{st,1}\right)-U_{tank}{A}_{tank,1}\left(T_{st,1}-T_{amb}\right)$$

(7)

$$\frac{{\rho }_{oil}{V}_{tank}}{N}{c}_{p,oil}\frac{\partial T_{st,2}}{\partial t}={\dot{m}}_{col}{c}_{p,oil}\left(T_{st,1}-T_{st,2}\right)+{\dot{m}}_s{c}_{p,oil}\left(T_{st,3}-T_{st,2}\right)-U_{tank}{A}_{tank,2}\left(T_{st,2}-T_{amb}\right)$$

(8)

$$\frac{{\rho }_{oil}{V}_{tank}}{N}{c}_{p,oil}\frac{\partial T_{st,N}}{\partial t}={\dot{m}}_{col}{c}_{p,oil}\left(T_{st,N-1}-T_{st,N}\right)+{\dot{m}}_s{c}_{p,oil}\left(T_{s,out,2}-T_{st,N}\right)-U_{tank}{A}_{tank,N}\left(T_{st,N}-T_{amb}\right)$$

(9)

More specifically, (${\rho }_{oil}$) is the oil density, ($c_{p,oil}$) the specific heat of the thermal oil, ($V_{tank}$) the tank volume, (${\dot{m}}_{col}$) the collector field flow rate, (${\dot{m}}_s$) the oil loop flow rate, ($T_{st,i}$) the temperature value of the mixing zone (i), ($U_{tank}$) the tank heat loss coefficient which is assumed to be equal to 0.5 W/m²K [14], while ($A_{tank,i}$) is the outer surface of each mixing zone and is defined by the equation below [14]:

$$A_{tank,i}=\left\{\begin{array}{c}\frac{\pi ·D_{tank}·H_{tank}}{{\rm N}}+\pi ·\frac{D_{tank}^2}{4}, i=1 or i={\rm N}\\ \frac{\pi ·D_{tank}·H_{tank}}{{\rm N}}, i=2:N-1\end{array}\right.$$

(10)

The tank height ($H_{tank}$) is considered to be equal to the tank diameter ($D_{tank}$). Moreover, the temperature of the oil stream that feeds the ORC ($T_{s,in}$) is equal to the temperature of the upper tank zone ($T_{st,1}$), while the temperature of the collector inlet stream ($T_{col,in}$) is equal to the temperature of the lower tank zone ($T_{st,N}$). These assumptions are described below:

$$T_{s,in}=T_{st,1}$$

(11)

$$T_{col,in}=T_{st,{\rm N}}$$

(12)

To determine the tank temperature profile during the day, the differential Equations (7)–(9) are discretized according to the finite difference method. For Zone 1, the discretization is presented below:

$$\frac{\partial T_{st,1}}{\partial t}=\frac{T_{st,1}-{{\rm T}}_{st,1}^0}{ \mathsf{\Delta }t}$$

(13)

The exponent “0” symbolizes the temperature values at the previous time step, while ($\mathsf{\Delta }t$) is the time step. In this study, the time step is assumed at 10 min, with the tank mixing zones at 5 [30].

2.3. ORC Plant Modeling

The hot stream leaving the TES tank with the temperature ($T_{s,in}$) feeds an ORC module through a heat recovery system, which is a heat exchanger and is made up of three parts, i.e., the economizer, the evaporator, and the superheater. In the end, the organic medium exits this system at the state of superheated vapor (state 4). The heat input to the ORC ($Q_{in}$) can be defined below:

$$Q_{ORC}={\dot{m}}_{ORC}·\left(h_4-h_3\right)={\dot{m}}_s·c_{p,oil}·\left(T_{s,in}-T_{s,out}\right)$$

(14)

The most commonly used heat exchanger type in ORC systems is a shell-and-tube heat exchanger [31]. For the examined configuration, this type is selected for the ORC heat recovery system and the recuperator. The pinch point in the entrance of the evaporator is considered at 10 K [30], while the superheating is assumed at 5 K [28]. Additionally, the organic fluid high pressure ($P_{high}$) is considered as a percentage of its critical pressure ($P_{crit}$). The fraction of the high pressure to the critical pressure is called parameter (α) and is described by the following equation [30]:

$$\alpha =\frac{P_{high}}{P_{crit}}$$

(15)

The electricity output of the expander minus the electricity consumed by the pump defines the ORC electricity production, which is considered as below [30]:

$$P_{el,ORC}={\eta }_{mg}·{\dot{m}}_{ORC}·\left(h_4-h_5\right)-\frac{{\dot{m}}_{ORC}·\left(h_2-h_1\right)}{{\eta }_{motor}}$$

(16)

The isentropic efficiency of the feeding pump (${\eta }_{is,pump}$) is regarded to be equal to 0.8, which is a typical value for small-scale solar ORC applications based on Ref. [32]. Moreover, a screw expander is considered to be installed, as a proper choice for applications greater than 10 kW_el [33]. The expander isentropic efficiency is considered at 0.85, a value assumed according to the literature [28,30]. Additionally, an air-cooled condenser is incorporated, where a temperature difference of 10 K is assumed. For the condenser, the selected heat exchanger type is the fin-and-tube heat exchanger, as a widely used configuration for air-cooled condensers in the literature [34]. Finally, the ORC thermal efficiency is defined as:

$${\eta }_{ORC}=\frac{P_{el,ORC}}{Q_{ORC}}$$

(17)

Taking into account the expected temperature levels of the heat source, MDM is selected to be used as a widely-used and efficient working medium, according to the literature [14,28,30]. All the parameters used for the ORC modeling can be found in

Table 1. Parameters of the ORC.

Parameters	Values
Evaporator pinch point (PP)	10 Κ
Recuperator temperature difference ($\Delta T_{rec}$)	5 Κ
Condenser temperature difference	10 K
Superheating ($\Delta T_{sh}$)	5 K
Pump isentropic efficiency (${\eta }_{is,pump}$)	0.8
$\mathrm{Pump} \mathrm{motor} \mathrm{efficiency} ({\eta }_{motor}$)	0.8
$\mathrm{Expander} \mathrm{isentropic} \mathrm{efficiency} ({\eta }_{is,exp}$)	0.85
$\mathrm{Electromechanical} \mathrm{efficiency} ({\eta }_{mg}$)	0.97

[28,30,32].

2.4. Hydrogen Plant Modeling

Hydrogen is produced by splitting water through the PEM water electrolysis process. The electrochemical reactions are described by the following expressions [35]:

$$\mathrm{Anode}: 2H_{2}O\to O_2+4H^{+}+4e^{-}$$

(18)

$$\mathrm{Cathode}: 2H^{+}+2e^{-}\to H_2$$

(19)

The overall reaction of hydrogen production is described below [35]:

$$2H_2{O}_{\left(l\right)}\to 2H_{2\left(g\right)}+O_{2\left(g\right)}$$

(20)

The previously mentioned procedure must feed with electrical power to be achieved. The required electricity input is provided by the ORC generator; therefore, a share of the electricity production is exploited for hydrogen production via the PEM water electrolyzer ($P_{el,PEM}$). Furthermore, the incoming water is not used as received from the network but is preheated at a temperature level of about 80 °C [36]. The water preheater is located after the ORC heat recovery system. In general, the total energy (ΔH) needed for the water electrolysis is defined by the following expression [37]:

$$\mathit{\Delta }H=\mathit{\Delta }G+T·\mathit{\Delta }S$$

(21)

(ΔG) is Gibb’s free energy and ($T·\mathit{\Delta }S$) is the thermal energy demand.

The electricity input to the electrolyzer is defined below [36]:

$$P_{el,PEM}=J·V_{cell}$$

(22)

where (J) is the current density, and (V_cell) is the electrolyzer cell voltage. An analytical representation of the electrolyzer model based on the current-voltage (J–V_cell) equations is included in Ref. [36]. Then, due to the low hydrogen density at the atmospheric pressure level, the produced hydrogen amount has to be compressed up to 100 bar, to increase its density [38]. The electricity input to the hydrogen compressor, which also comes from the ORC electricity generation, is presented below:

$$P_{el,comp}=\frac{{\dot{m}}_{hydrogen}·\left(h_{d,is}-h_c\right)}{{\eta }_{is,comp}·{\eta }_{mg}}$$

(23)

The compressor isentropic efficiency (${\eta }_{is,comp}$) is considered at 85% [38], while the electromechanical efficiency (${\eta }_{mg}$) is equal to 97% [30]. Due to compression, except for the pressure increment, the temperature also increases; therefore, the hydrogen amount has to be cooled down to the temperature of 25 °C. Finally, hydrogen fuel at a pressure of 100 bar and a temperature of 25 °C, is stored in tanks, so that it can be transported and utilized.

2.5. General Thermodynamic Indexes

The governing equations of the components are described in the above sections. Both the energy and exergy balance expressions of each component are revised in

Table 2. Energy and exergy balance equations of the components.

Components	Energy Balance Equation	Exergy Balance Equation
PTC	Qsolar = Qu + Qloss,th + Qloss,opt	Exsolar = Exu + Exloss + Exdest
TES tank	ṁcol · hcol,out + ṁs · hs,out,2 = ṁcol · hcol,in + ṁs · hs,in + Qloss	ṁcol · excol,out + ṁs · exs,out,2= ṁcol · excol,in + ṁs · exs,in + Exloss + Exdest
Heat recovery system	ṁs · hs,in + ṁORC · h3 = ṁs · hs,out + ṁORC · h4 + Qloss	ṁs · exs,in + ṁORC · ex3= ṁs · exs,out + ṁORC · ex4 + Exloss + Exdest
ORC pump	ṁORC · h1 + Wpump = ṁORC · h2	ṁORC · ex1 + Expump = ṁORC · ex2 + Exdest
ORC condenser	ṁORC · h6 = ṁORC · h1 + Qout	ṁORC · ex6 = ṁORC · ex1 + Exout + Exdest
ORC expander	ṁORC · h4 = ṁORC · h5 + Wexp	ṁORC · ex4 = ṁORC · ex5 + Exexp + Exdest
ORC recuperator	ṁORC · h2 + ṁORC · h5 = ṁORC · h3 + ṁORC · h6 + Qloss	ṁORC · ex2 + ṁORC · ex5= ṁORC · ex3 + ṁORC · ex6 + Exloss + Exdest
Water preheater	ṁwater · hwater,a + ṁs · hs,out =ṁwater · hwater,b + ṁs · hs,out,2 + Qloss	ṁwater · exwater,a + ṁs · exs,out= ṁwater · exwater,b + ṁs · exs,out,2 + Exloss + Exdest
PEM water electrolyzer	ṁwater · hwater,b + WPEM= ṁhydrogen · hhydrogen,c + ṁoxygen · hoxygen	ṁwater · exwater,b + ExPEM= ṁhydrogen · exhydrogen,c + ṁoxygen · exoxygen + Exdest
Hydrogen compressor	ṁhydrogen · hhydrogen,c + Wcomp= ṁhydrogen · hhydrogen,d	ṁhydrogen · exhydrogen,c + Excomp= ṁhydrogen · exhydrogen,d + Exdest
Intercooler	ṁhydrogen · hhydrogen,d= ṁhydrogen · hhydrogen,e + Qout	ṁhydrogen · exhydrogen,d= ṁhydrogen · exhydrogen,e + Exout + Exdest

.

Taking into account all the aforementioned equations, the system is evaluated thermodynamically. This evaluation is made through the main energy and exergy indexes, which are presented in this section. First, the overall net electricity load ($P_{el,net}$), which is absorbed by the grid, can be described as below:

$$P_{el,net}=P_{el,ORC}-P_{el,PEM}-P_{el,comp}$$

(24)

The system’s energy efficiency (${\eta }_{en}$) is calculated as:

$${\eta }_{en}=\frac{P_{el,net}+LH{V}_{hydrogen}·{\dot{m}}_{hydrogen}}{Q_{solar}}$$

(25)

The system’s exergy efficiency (${\eta }_{ex}$) is calculated as:

$${\eta }_{ex}=\frac{E{x}_{el,net}+e{x}_{hydrogen}·{\dot{N}}_{hydrogen}}{E{x}_{solar}}$$

(26)

More specifically, ($LH{V}_{hydrogen}$) is the hydrogen low heating value, and the hydrogen exergy ($e{x}_{hydrogen}$) is defined by the following expression [36]:

$$e{x}_{hydrogen}=e{x}_{hydrogen,chem}+e{x}_{hydrogen,phy}$$

(27)

(ex_{hydrogen,chem}) is the chemical exergy of hydrogen which is equal to 236.1 kJ/mol [39], and (ex_hydrogen,phy) is the physical exergy which is calculated as below [36]:

$$e{x}_{hydrogen,phy}=\left(h-h_0\right)-T_0·\left(s-s_0\right)$$

(28)

Enthalpy (h) and entropy (s) are defined taking into account the electrolyzer conditions, while subscript “0” refers to the reference ambient conditions. Additionally, the exergy rate of solar irradiation is determined according to the Petela model as below [30]:

$$E{x}_{solar}=Q_{solar}·\left(1-\frac{4}{3}·\frac{T_0}{T_{sun}}+\frac{1}{3}{\left(\frac{T_0}{T_{sun}}\right)}^4\right)$$

(29)

($T_0$) is the reference environment temperature which is supposed at 298.15 K (25 °C) and ($T_{sun}$) is the sun’s temperature which is assumed at 5770 K [30]. Apart from the aforementioned definitions which refer to a specific time, the annual energy and exergy efficiencies can also be considered, taking into account the corresponding annual energy (Y) and exergy (YE) values; therefore, the following expressions are made:

$${\eta }_{en,annual}=\frac{Y_{el,net}+Y_{hydrogen}}{Y_{solar}}$$

(30)

$${\eta }_{ex,annual}=\frac{Y{E}_{el,net}+Y{E}_{hydrogen}}{Y{E}_{solar}}$$

(31)

2.6. Financial Analysis

In this section, the main economic indexes are presented. Before their calculation, it is important to define the capital cost of the whole installation ($C_0$), and the annual cash flow (CF). First of all, the capital cost is made up of the cost of the collectors, the TES tank, the ORC module, the PEM water electrolyzer, the hydrogen compressor, the hydrogen storage tank, and the heat exchangers (water preheater, and intercooler); therefore, the capital cost is calculated by the following equation:

$$C_0=C_{PTC}+C_{tank}+C_{ORC}+C_{PEM}+C_{comp}+C_{hydrogen\_tank}+C_{HEX}$$

(32)

In addition, the annual cash flow is defined by considering the inflows from the production of electricity and green hydrogen, which are described by the cost of electricity ($K_{el}$), and the cost of hydrogen ($K_{hydrogen}$), as well as the outflows from the operation and maintenance costs ($K_{O\&M}$); therefore, the following expression is made:

$$CF=Y_{el,net}·K_{el}+M_{hydrogen}·K_{hydrogen}-K_{O\&M}$$

(33)

(M_hydrogen) is the total annual hydrogen production in kg. Taking into account the above definitions, the four main economic indexes (simple payback period, payback period, net present value, and internal rate of return) are assumed. More specifically, the simple payback period (SPBP) is defined as:

$$SPBP=\frac{C_0}{CF}$$

(34)

The payback period (PBP) is defined as:

$$PBP=\frac{\ln \left(\frac{CF}{CF-C_0·i}\right)}{\ln \left(1+i\right)}$$

(35)

The net present value (NPV) is defined as:

$$NPV=-C_0+CF·\frac{{\left(1+i\right)}^N-1}{i·{\left(1+i\right)}^N}$$

(36)

The internal rate of return (IRR) is defined by the following non-linear expression:

$$IRR=\frac{CF}{C_0}·\left(1-\frac{1}{{\left(1+IRR\right)}^N}\right)$$

(37)

The parameters of the financial study are presented in

Table 3. Parameters of financial analysis.

Parameters	Values
$\mathrm{Cost} \mathrm{of} \mathrm{electricity} (K_{el}$)	0.30 EUR/kWh
Specific cost of PTC	230 EUR/m2 [13]
Specific cost of the TES tank	1000 EUR/m3 [30]
Specific cost of the ORC	1800 EUR/kWel [13]
Specific cost of the hydrogen compressor	1600 EUR/kWel [38]
Specific cost of the hydrogen storage tank	600 EUR/kg [40]
Specific cost of the PEM water electrolyzer	800 EUR/kWel [41]
Specific cost of the heat exchangers	100 EUR/kW [30]
Project lifetime (N)	25 years [30]
Discount factor (i)	4% [30]
$\mathrm{Operation} \mathrm{and} \mathrm{maintenance} \mathrm{costs} (K_{O\&M}$)	2% of the capital cost [13]

.

2.7. Simulation Methodology

The configuration was placed on Kythnos island, Greece. Due to the limited available space on the island, a solar field of about 975 m², which included 25 PTC modules, was selected. The first step of the analysis was the conduction of a preliminary analysis, which was carried out in steady-state conditions through the EES software [42]. The preliminary analysis included the investigation of the operation of the main sub-devices, which were the solar collector, the ORC system, and the electrolyzer; thus, the collector thermal efficiency according to Equation (6), the ORC efficiency, and the hydrogen production rate were defined.

Afterward, the work was focused on the dynamic performance of the system, studying its performance during the day and the year. The dynamic model was developed in Matlab. At first, in order to carry out the dynamic simulation, typical meteorological and irradiation data for this location were used from the Photovoltaic Geographical Information System (PVGIS) interactive tool, which is a large database developed by the European Commission [43]. In addition, the steady-state analysis results that defined the performance of the PTC, the ORC, and the electrolyzer, were provided as inputs to the dynamic model. Subsequently, the storage tank differential equations were solved for each time step, for every single day of the year. Each day was solved multiple times until the numerical simulation met the convergence criteria. The simulation ended if the absolute errors of the temperature values T_c,in, T_c,out, T_s,in, and T_s,out,2 between the beginning and the end of the day, were lower than 10⁻⁴; therefore, the operation during the days of the year was determined.

The main system outputs, which were the net electricity and the hydrogen production, were strongly dependent on the ORC electricity production. This value relies on the ORC heat input, which comes from the available solar energy and the ORC thermal efficiency. The cycle efficiency was assumed to be determined through the evaporator saturation temperature and the ambient temperature, which affected the condenser temperature and was known for each time step from the meteorological data. Nevertheless, the evaporator temperature was not known a priori; therefore, different values of this temperature were examined, to find the one that led to the maximum daily electricity production. At this level, the examined thermodynamic parameters were the net (base) electricity load absorbed by the local grid, which ranged from 0 to 175 kW, and the TES tank volume, which varied from 10 to 50 m³. These two parameters represented the two different storage methods integrated into this unit. Apart from the energetic and exergetic analysis, the system was also evaluated economically. In the financial study, another parameter was defined, which was the price of hydrogen. This value ranged from 15 to 25 EUR/kg [44], as it was assumed the hydrogen price was going to increase in the following years, due to the rising demand. Therefore, through the transient simulations, the main thermodynamic and financial indexes were determined for each combination of the aforementioned parameters.

Finally, to define an overall optimum operation scenario, multi-objective optimization was carried out. Thus, the proper combination of the net electricity load and the TES tank volume was determined. The following objective function (F), which has to be as minimum as possible, takes into account all the main evaluation indexes. An optimum case must achieve the maximum possible annual energy efficiency, annual exergy efficiency, net present value, and the minimum possible payback period. The objective function is described by the following expression [30]:

$$F=\sqrt{{\left(\frac{{\eta }_{en,annual}-{{\eta }_{en,annual}}_{max}}{{{\eta }_{en,annual}}_{max}-{{\eta }_{en,annual}}_{min}}\right)}^2+{\left(\frac{{\eta }_{ex,annual}-{{\eta }_{ex,annual}}_{max}}{{{\eta }_{ex,annual}}_{max}-{{\eta }_{ex,annual}}_{min}}\right)}^2+{\left(\frac{PBP-PB{P}_{min}}{PB{P}_{max}-PB{P}_{min}}\right)}^2+{\left(\frac{NPV-NP{V}_{max}}{NP{V}_{max}-NP{V}_{min}}\right)}^2}$$

(38)

A flow chart of the methodology of the solution process is depicted in Figure 2.

2.8. Validation of the Developed Model

As the configuration was made up of three sub-plants, the model of each one of them was validated separately. At first, the LS-2 parabolic collector model was validated and found to be accurate enough, as it is shown in

Table 4. Validation of the PTC with experimental data from the literature.

	Examined Cases				Present Model		Reference [45]		Deviation
Gbn (W/m2)	uair	${\dot{\mathbf{V}}}_{\mathbf{fluid}}$	Tamb	Tcol,in	ηth,col	qloss,th	ηth,col	qloss,th	ηth,col	qloss
	(m/s)	(L/min)	(°C)	(°C)	(%)	(W/m2)	(%)	(W/m2)	(%)	(%)
933.7	2.6	47.7	21.2	102	72.72	5.44	72.5	5.62	0.30	3.10
968.2	3.7	47.8	22.4	151	72.3	9.67	72.1	9.39	0.28	2.99
982.3	2.5	49.1	24.3	197	71.71	15.65	71.6	15.18	0.15	3.10
909.5	3.3	54.7	26.2	250	70.63	24.26	70.4	24.51	0.33	1.02
937.9	1	55.5	28.8	297	69.26	37.92	69.1	37.86	0.23	0.16
880.6	2.9	55.6	27.5	299	69	37.9	68.7	38.38	0.44	1.25
920.9	2.6	56.8	29.5	379	64.92	77.19	64.8	76.71	0.19	0.63
903.2	4.2	56.3	31.1	355	66.35	62.77	66.1	63.35	0.38	0.92

. The mean deviation of the collector thermal efficiency and thermal losses per collector aperture, between the proposed model and the experimental work of Forristall [45], was calculated at 0.29%, and 1.64%, respectively. Additionally, the ORC module model was compared to the data from the experimental analysis of Eyerer et al. [46], and the results are presented in

Table 5. Validation of the ORC with experimental data from the literature.

Examined Cases				Present Model		Reference [46]		Deviation
Organic Fluid	$\dot{\mathbf{m}}$	Phigh	Plow	Pel,ORC	ηORC	Pel,ORC	ηORC	Pel,ORC	ηORC
	(kg/s)	(bar)	(bar)	(W)	(%)	(W)	(%)	(%)	(%)
R1233zd(E)	20 × 10−3	6.82	1.82	174.1	3.66	174.8	3.67	0.39	0.33
R1233zd(E)	35 × 10−3	8.74	1.94	268.1	3.46	270.2	3.45	0.79	0.32
R1224yd(Z)	20 × 10−3	6.10	1.94	76.5	1.80	76.9	1.81	0.64	0.88
R1224yd(Z)	35 × 10−3	9.21	2.19	286.7	4.19	286.5	4.15	0.06	1.01
R245fa	20 × 10−3	7.05	2.09	139.1	2.83	139.1	2.84	0.01	0.39
R245fa	35 × 10−3	9.68	2.23	311.7	4.07	317.5	4.08	1.81	0.20

. More specifically, the mean deviation of the electricity production rate and thermal efficiency was found at 0.62% and 0.52%, respectively. These values were considered low and acceptable. The electrolyzer model was also tested through experimental data, which were provided from the study by Ioroi et al. [47]. The model was considered to be valid enough as the mean cell potential deviation was calculated at 2.88%. Finally, the analytical results are given in

Table 6. Validation of the PEM water electrolyzer with experimental data from the literature.

Examined Cases	Present Model	Reference [47]	Deviation
Current Density	Cell Potential	Cell Potential	Cell Potential
(A/m2)	(Vcell)	(Vcell)	(%)
185.35	1.78	1.70	4.96
394.18	1.83	1.76	4.09
590.80	1.85	1.79	3.54
793.62	1.87	1.81	3.70
990.27	1.89	1.83	3.22
1494.27	1.91	1.86	2.77
1998.28	1.93	1.89	2.29
3000.23	1.96	1.92	2.19
3996.03	1.99	1.96	1.52
5004.11	2.01	1.99	0.56

and Figure 3.

3. Results and Discussion

3.1. Preliminary Analysis

The performance of the main sub-systems was investigated via the preliminary analysis. At first, the collector performance was determined via the thermal losses analysis in steady-state conditions, as it is presented in Ref. [29]. According to these calculations, the collector thermal efficiency can be approached through a polynomial expression, as described in Equation (6). The defined polynomial equation, for which the coefficient of determination (R²) was found at almost 1, is presented below:

$${\eta }_{th,col}=0.7358·K\left(\theta \right)-0.0789·\frac{T_{col,in}-T_{amb}}{G_{bn}}+3.407·{10}^{-4}·\frac{{\left(T_{col,in}-T_{amb}\right)}^2}{G_{bn}}-1.771·{10}^{-6}·\frac{{\left(T_{col,in}-T_{amb}\right)}^3}{G_{bn}}$$

(39)

The PTC thermal efficiency as a function of the temperature difference between the collector inlet and the ambient temperature is depicted in Figure 4. The collector efficiency decreased with the increase in the aforementioned temperature difference, while the maximum achieved value was equal to 72.8%. Additionally, through steady-state calculations, the ORC thermal efficiency was also specified, taking into account two important parameters, i.e., the evaporator saturation temperature (T_sat), and the ambient temperature. These two temperature levels defined both the higher and the lower pressure values, which strongly affected the cycle’s efficiency. The ORC efficiency, which is illustrated by a tri-dimensional plot in Figure 5, increased if the evaporator saturation temperature increased and the ambient temperature decreased. The calculated values ranged from 19.8% to 35.4%. Moreover, the main output of the PEM electrolyzer was the hydrogen production rate, which highly depended on the electricity input; therefore, according to the results that are shown in Figure 6, there was almost a linear correlation between the produced hydrogen mass flow rate and the electricity load provided to the PEM water electrolyzer. All the above results were exploited as inputs to the transient operation simulation.

3.2. Dynamic Analysis

3.2.1. Thermodynamic Analysis

The thermodynamic results, which were selected to be presented, were the annual energy efficiency, the annual exergy efficiency, and the annual hydrogen production, for different values of the net electricity load, and the TES tank volume. Particularly, according to Figure 7 and Figure 8, both the annual energy and exergy efficiencies increased when the net electricity load increased; therefore, when the net electricity production was getting higher and the hydrogen generation decreased, the configuration performed better energetically and exergetically. Moreover, the aforementioned efficiency rates had a slightly decreasing rate if the TES tank volume increased. Figure 7 indicates that the energy efficiency reached values from 9.77% to 17.7%, while, as it is shown in Figure 8, the exergy efficiency ranged from 10.24% to 18.99%. The maximum energy efficiency value of 17.7%, and the maximum exergy efficiency value of 18.99%, were achieved when the net electricity load was considered to be equal to 175 kW, and the tank volume was assumed at 10 m³. Furthermore, the annual hydrogen production in kg is illustrated in Figure 9. The hydrogen production amount decreased slightly with the increase in the tank volume, but noticeably when the net electricity load increased. When the net electricity generation increased, the electricity which was provided to the PEM water electrolyzer was getting lower. As a result, the hydrogen production rate decreased. If the entire ORC electricity production fed the electrolyzer, and the net electricity load was equal to zero, the annual hydrogen production amount was varied from 5836 to 5913 kg. If the net electricity load was equal to 175 kW, the produced hydrogen amount was negligible. That is why this value was selected as the maximum examined one for the net (base) electricity load.

3.2.2. Financial Analysis

To evaluate the system from an economic point of view, the selling price of hydrogen was considered as an additional parameter, except for the net electricity load, and the TES tank volume. Different values of the hydrogen price were assumed to take into account different future market conditions. The fundamental financial indexes, which are the payback period and the net present value, are presented in Figure 10, Figure 11, Figure 12, Figure 13, Figure 14 and Figure 15. At first, all these figures indicated that the payback period increased slightly with the increase in the storage tank volume, while the net present value had a minor decreasing rate. Moreover, Figure 10 shows that the payback period decreased when the net electricity load increased. On the other hand, the net present value had exactly the opposite trend, as it is shown in Figure 11. Figure 10 and Figure 11 present the results when the hydrogen price was equal to 15 EUR/kg. The lowest payback period value was calculated at 6.98 years, while the highest net present value is EUR 908,998. Both values were defined when the net electricity load was equal to 175 kW, and the TES tank volume was at 10 m³; therefore, it is important to mention that if the hydrogen price was assumed at this level, the greater the net electricity production, which would lead to lower hydrogen generation, which is more economically viable.

In addition, when the hydrogen price was considered to be equal to 20 EUR/kg, the aforementioned trends tended to change. First of all, lower payback period values were calculated for the net electricity load values from 125 to 175 kW according to Figure 12. The lowest payback period, which was equal to 6.98, was defined when the net electricity load was equal to 175 kW, and the TES tank volume was at 10 m³; however, larger levels of the net present value were determined when the net electricity load was varied from 0 to 50 kW, as it is illustrated in Figure 13. The highest net present value of EUR 947,740 was calculated when there was no net electricity load, and the TES tank volume was equal to 10 m³. In parallel, the lowest financial performance, i.e., when the net present value was found at lower levels, and the payback period at higher levels, was determined for a 100-kW net electricity load. Thus, hydrogen generation appears to be more financially sustainable compared to net electricity production, in some cases. The value of 20 EUR/kg can be characterized as a “critical” one, as the previously defined tendency with which net electricity generation performs better financially is now overturned.

Furthermore, Figure 14 and Figure 15 depict clearly that hydrogen production was more economically viable than net electricity generation, as the hydrogen price was assumed at 25 EUR/kg. Figure 14 indicates that the payback had a decreasing rate if the net electricity load decreased, and the electricity input to the electrolyzer increased. The net present had an increasing rate with the decrease in the net electricity load, which led to an increase in hydrogen production, as it is shown in Figure 15. In this case, assuming the price of 25 EUR/kg, the highest net present value levels and the lowest payback period values were specified to be equal to about EUR 1,409,579 and 5.83 years, respectively if the net electricity load was equal to zero, and the TES tank volume at 10 m³. Consequently, the increase in the price of hydrogen makes hydrogen production, and the overall configuration, more profitable.

3.3. Dynamic Optimization

After the investigation of the configuration performance taking into account the different thermodynamic and financial parameters, it was crucial to carry out an optimization procedure, considering all the previously defined cases. As a result, the best combination of the net electricity load and TES tank volume was determined. A multi-criteria objective function was assumed, according to Equation (38). For this analysis, the hydrogen price was considered at 25 EUR/kg. The objective function (F) took various values and, more specifically, the lower ones, which calculated a net electricity load that was equal to 75 or 175 kW, or for lower values of the TES tank volume, in general, as illustrated in Figure 16. The lowest value was found at 0.92 and was determined when the net electricity load was equal to 75 kW and the storage tank volume at 10 m³. For this optimum operation scenario, the annual electricity and hydrogen production was calculated at 210,163 kWh and 2356.5 kg, respectively, while the annual energy efficiency was found at 14.52%, and the corresponding exergy efficiency at 15.48%. Finally, the payback period, the simple payback period, the internal rate of return, and the net present value were defined at 6.73 years, 5.8 years, 16.89%, and EUR 1,073,384, respectively. All the optimum case outputs are presented in

Table 7. Outputs of the optimum case.

Outputs	Values
Annual net electricity production ($Y_{el,net}$)	210,163 kWh
Annual hydrogen energy production ($Y_{hydrogen}$)	78,550 kWh
Annual hydrogen production amount ($M_{hydrogen}$)	2356.5 kg
Annual energy efficiency (${\eta }_{en,annual}$)	14.52%
Annual exergy efficiency (${\eta }_{ex,annual}$)	15.48%
$\mathrm{Capital} \mathrm{cost} (C_0$)	€633,862
Simple payback period (SPBP)	5.80 years
Payback period (PBP)	6.73 years
Internal rate of return (IRR)	16.89%
Net present value (NPV)	€1,073,384

.

3.4. Daily and Monthly Operation of the Optimum Scenario

The performance of the previously determined optimum operation scenario is presented in this section on a daily and monthly basis. For this case, the net electricity load and storage tank volume was equal to 75 kW and 10 m³, respectively. First of all, three typical days of the year, i.e., 16 February, 15 June and 15 October were selected indicatively to examine both the ORC and net electricity production during the day. The results are illustrated in Figure 17. The difference between these values regards the electricity provided to the PEM water electrolyzer and the hydrogen compressor. Figure 16 depicts that the maximum ORC electricity load during the day was 178.6 kW on 15 June, 104.2 kW on 16 February, and 107.1 kW on 15 October. Additionally, the system operated for only 3.7 h on 16 February, while on 15 October and 15 June, the operational hours were about 9 and 11.3, respectively; therefore, the ORC electricity production was larger in the summer period compared to the other year’s seasons, due to the larger irradiation rates, and the sunshine duration. Moreover, as the net electricity load absorbed by the local grid was constant at 75 kW, the surplus electricity was greater in June, rather than in February or October. Finally, it was obvious that the TES tank contributed to the performance stability.

In addition, the hydrogen production rate during the typical days (i.e., 16 February, 15 June and 15 October) is illustrated in Figure 18. According to the results, the maximum hydrogen production mass flow rate during the day was about 1.64 kg/h on 15 June, 0.47 kg/h on 16 February, and 0.51 kg/h on 15 October. It is important to mention that the curves had similar trends to the ORC electricity production curves in Figure 17. As a result, the produced hydrogen amounts that could be stored for later use were higher in June, rather than in February and October.

Furthermore, the total net electricity and hydrogen production in kWh, as well as the available solar energy for each month is shown in Figure 19. In general, greater energy values were determined in the summer period for both the products (electricity, and hydrogen) and the inputs (solar irradiation). The maximum values were achieved in July when the net electricity generation was equal to 28,046 kWh, and the hydrogen production in terms of energy was determined at 14,040 kWh. The hydrogen production amount in mass units is presented in Figure 20, and its maximum value, which was equal to 421.2 kg, was defined in July. Figure 20 also depicts the monthly energy and exergy efficiency of the overall configuration. The highest level of energetic and exergetic performance was determined in May, with an energy efficiency of 16.4% and an exergy efficiency of 17.46%.

At this point, it is remarkable to state that the production profile of the electricity and hydrogen in February was very different compared to the other months. This fact is explained by the restricted solar beam potential during the winter period in Greece. Thus, a typical February day had no available solar beam irradiation for many hours per day and this result created the non-symmetrical profile in the production curves of Figure 17 and Figure 18.

4. Conclusions

The work is focused on a renewable energy plant based on solar energy, which provides two useful outputs, i.e., electricity from an ORC module, and green hydrogen via PEM water electrolysis. Additionally, two energy storage technologies are integrated, which are TES, using thermal oil, and hydrogen storage. The configuration, which is placed on the Greek island of Kythnos, provides electricity, and, in parallel, can store the excess electricity generation through the production of hydrogen, which can be exploited as a fuel in plenty of applications on the island. Thus, the installation was studied on dynamic conditions, through the weather and irradiation data of Kythnos, and optimized to be thermodynamically efficient and economically viable. The most important aspects of the present analysis are summed up below:

• Different combinations of the net (base) electrical load, and the TES tank volume, as well as various hydrogen price values, were examined.
• In the optimum scenario, a collector aperture of 975 m², a storage tank volume of 10 m³, and a net electricity load of 75 kW were determined.
• The net electricity generation was equal to 210163 kWh per year, while the hydrogen production amount was found at 2356.5 kg per year.
• The overall energy efficiency was calculated at 14.52%, and the exergy efficiency at 15.48%.
• The payback period and the net present value of the investment were defined at 6.73 years and EUR 1,073,384, respectively.
• The proposed system can meet the various demands of an island, or a remote area, in a low-carbon and sustainable way, with acceptable financial viability.