A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment

Applied Energy xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment Francesco Calise a,⇑, Andrea Cipollina b, Massimo Dentice d’Accadia a, Antonio Piacentino c a b c

DII – University of Naples Federico II, Italy DICGIM – University of Palermo, Italy DEIM – University of Palermo, Italy

h i g h l i g h t s  A novel fully-renewable trigeneration system is designed and simulated.  The system uses low-enthalpy geothermal energy and solar energy.  The system produces cool, heat, electricity and desalted water.  The system is based on MED desalinization technology.  The system is very promising from energetic, economic and environmental viewpoints.

a r t i c l e

i n f o

Article history: Received 2 December 2013 Received in revised form 28 February 2014 Accepted 22 March 2014 Available online xxxx Keywords: Geothermal energy Solar heating and cooling PVT Solar desalination MED

a b s t r a c t This paper investigates the integration of solar and geothermal energy in a novel polygeneration system producing simultaneously: electricity, thermal energy, cooling energy and fresh water. The polygeneration system under analysis includes concentrating photovoltaic/thermal solar collectors (CPVT), a Geothermal Well (GW) a multi-effect distillation (MED) system for seawater desalination, a single-stage LiBr–H2O absorption chiller and additional components, such as: storage tanks, heat exchangers and balance of plant devices. The CPVT produces simultaneously electrical energy and thermal energy, at a maximum temperature of about 100 °C. The electrical energy is delivered to the grid, whereas the thermal energy can be used for different scopes. First, the thermal energy can be used for heating purposes and/or Domestic Hot Water production. As an alternative, solar thermal energy can be used to drive an absorption chiller, producing chilled water for space cooling. Finally, solar energy, in combination with the thermal energy produced by low-enthalpy (about 80 °C) geothermal wells, may be used by the MED system to convert seawater into desalinated water. Geothermal energy is also used to produce Domestic Hot Water at 45 °C. The system is dynamically simulated by means of a zero-dimensional transient simulation model. The simulation model also includes detailed control strategies, for the management of the different technologies included in such a complex system. The system is assumed to be operated in some of the several small volcanic islands in the Mediterranean Sea, assuming Pantelleria (Trapani, Italy) as main case study. Here, the availability of solar and geothermal energy is high whereas the availability of fresh water is scarce and its cost consequently high. Results show an excellent energetic performance of the system under investigation. From the economic point of view, the profitability of the system dramatically increases when user Domestic Hot Water demand is high. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction During the last few years, world is facing severe issues related to the increasing consumption of energy and water resources. This ⇑ Corresponding author. Tel.: +39 0817682301; fax: +39 081 2390364. E-mail addresses: [email protected], [email protected] (F. Calise).

increase is basically due to the huge amount of energy and water demanded by the developed Countries (USA, EU, Japan, etc.). Nevertheless, a significant and increasing amount of energy and water is also demanded by the highly crowded emerging Countries (China, India, Brazil, etc.) [1]. Unfortunately, the present energy and water policies used by the majority of the Countries are based on an intensive utilization of fossil fuels and water resources. Such

http://dx.doi.org/10.1016/j.apenergy.2014.03.064 0306-2619/Ó 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

2

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

Nomenclature A c cp cue cut cuwat e eff BPE Cop COP COPEHP,rs Eel Ibeam Itot I0 J _ M mwat NTU Q_ Q SPB t T U V X

area (m2) specific heat (kJ/kg/K) heat capacity at constant pressure (kJ/kg/°C) electricity unit cost (€/kW h) heat unit cost (€/kW h) water unit cost (€/kg) specific energy consumption in fresh water production (€/m3) heat exchange effectiveness boiling point elevation (°C) operating cost (€) coefficient of performance (–) coefficient of heat pump in reference system (–) electrical energy (kW h) beam radiation (kW h) total radiation (kW h) system capital cost (€) component capital cost (€) mass flow rate (kg/s) desalinated water mass (kg) number of transfer units thermal power (kW) thermal energy (kW h) simple pay back period (years) temperature (°C) temperature (K) overall heat transfer coefficient (kW/m2 °C) volume (m3) water salinity (parts per million, ppm)

policies scarcely take into account the sustainability of this approach, which can lead to rapid deployment of fossil fuels and water resources in few decades. In particular, in the last few years, the majority of OECD (Organisation for Economic Co-operation and Development) Governments realized that new policies must be implemented in order to promote a more sustainable use of energy resources [2]. Conversely, negligible efforts have been performed in order to promote a more careful utilization of water resources, which are becoming extremely scarce in several locations in the world. This issue was analyzed by Artistidis et al. for Crete [3], addressing the relationship between water research status and climate changes. The same relation was studied for the Southern Africa by Kusangaya et al. [4]. A similar analysis for the Mediteranean districts was analyzed by Manios et al. [5]. All these researches pointed out that in the next future the availability of water may become an issue more severe than fossil fuels accessibility. Therefore, water and energy problems must be addressed in an integrated approach aiming at promoting new technologies for a sustainable production of both water and energy. In this framework, significant improvements have been performed in the last decades aiming at promoting renewable energy sources. In fact, although conventional energy conversion technologies, based on the utilization of fossil fuels, are the most profitable option, renewable energy sources are becoming economically competitive when they are supported by public funding. Moreover, in the last decades, the fossil fuel cost is rapidly increasing and the capital cost of Renewable Energy Sources (RES) technologies is simultaneously dramatically decreasing. Therefore, it can be expected that some of the renewable energy technologies will become economically competitive with conventional ones in the near future. Simultaneously, renewable energy sources are also an interesting option for producing potable water from seawater

Greek symbols q density (kg/m3) k latent heat of vaporization/condensation (kJ/kg) gt thermal efficiency gel electrical efficiency gel,RS reference system electrical efficiency Subscripts aux auxiliary B brine bot bottom c cooling/cold d demanded ch chilled/chilling D,i distillate by evaporation at ith effect D,flash ‘‘i’’ distillate produced by flash at brine inlet at ith effect el electrical ext external f feed water ft feed-in tariff h heating/hot in inlet motive related to the hot water stream supplying energy to the MED plant o outlet p produced

using electrical or thermal energy. Therefore, a mature development of renewable systems may be helpful for both energy and water issues. It is worth noting that, in continental EU Countries these issues are mitigated by the large availability of natural water resources and fossil fuels (or nuclear power plant). Conversely, this issue is particularly severe in the islands of the Southern Mediterranean Sea, where the availability of fossil fuels and water resources is scarce or null. Nevertheless, such islands are typically rich in renewable energy (solar energy and, in case of volcanic islands, geothermal energy) and have easy access to seawater. Therefore, the scope of this paper is the design of a novel integrated process, based on the use of renewable energy sources (namely solar and geothermal), aiming at the production of energy (electricity, cool and heat) and fresh water at reasonable costs. In particular, the study is focused on volcanic islands, due to the availability of a constant heat flow from geothermal resources, to be used in order to mitigate the oscillations of solar energy availability. The following technologies are simultaneously included in the system: Solar Heating and Cooling (SHC), Concentrating Photovoltaic–Thermal collectors (CPVT), Multiple Effect Distillation (MED) for seawater desalination and Geothermal Wells (GW) for the utilization of low-enthalpy geothermal energy. Solar Heating and Cooling (SHC) systems are based on conventional solar thermal collectors, producing thermal energy. During the winter, such thermal energy is used for space heating. Conversely, during the summer, solar thermal energy is converted in cooling energy by a thermally driven chiller (absorption, adsorption, etc.). In particular, SHC technology is especially attractive in summer, when the demand for cooling is often simultaneous to the large availability of solar radiation [6].

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

Photovoltaic/thermal collectors (PVT) are solar devices which simultaneously provide electricity and heat [7]. The basic principle of a PVT collector is simple, since it can be obtained by a conventional solar thermal collector whose absorber is covered by a suitable PV layer [8]. The absorbed thermal energy is transferred to a fluid (typically air or water), whereas the PV produces electricity [9]. Concentrating PVT collectors (CPVT) are PVT collectors placed in the focus of some reflectors (Fresnel, parabolic [10], dish [11], etc.) [12]. Note that PV efficiency typically decreases with increasing temperatures and CPVT operate at higher temperatures than flat-plate PVT systems [13]. Therefore, the use of CPVT may be improved by adopting novel PV materials, such as multi-junction solar cells, able to approach a nominal efficiency of 40% [14]. The use of such materials in CPVT may even lead to a system operating up to 240 °C at reasonable conversion efficiency (slightly lower than 20%) [12]. Commercially or pre-commercially available CPVT systems are typically a small amount of the PVT under development [13]. Several technologies have been used to face fresh water scarcity by desalinating seawater or brackish water. Thermal systems, based on either Multi-Stage Flash (MSF) [15] and Multiple Effect Distillation (MED) [16] schemes, have been mainly devoted to large scale fresh water production, with average capacities usually exceeding 10,000–20,000 m3/day. The main limits of thermal desalination systems consist in their low energetic efficiency, which induces the consumption of 50–60 kWhthermal per m3 of water and increases the production cost. Several different direct and indirect solar thermal desalination systems have been designed, as accurately reviewed in [17]: (i) single effect solar stills (even in greenhouse combination or externally activated configuration), single and multiple-effect basin stills [18], wick stills and diffusion stills; (ii) Humidification–Dehumidification (HD) plants [19]; (iii) MSF and MED systems [20]; (iv) Plants based on separation by freezing [21]. Also a novel hybrid process based on the use of hydrophobic membranes, namely Membrane Distillation, has been recently presented in the literature [22], being particularly suitable for small scale systems (up to few m3/day capacity) powered by solar energy [23] but also by waste heat or an hybrid combination of the two heat sources [23]. MED systems are especially attractive for their capability to be integrated with renewable energy sources [24,25], due to the possibility to be supplied by low-temperature heat (e.g. solar and geothermal) [26,27]. In particular, the interest for the use of the solar source has been growing in the last few years, due to the large availability of this source in many areas and the possibility to work with mature technologies. The focus in this paper is posed on MED plants powered by the combination of solar and geothermal energy, because MED plants offer a large flexibility in installation capacity: in case of integration with large solar fields, in fact, this technology allows the installation of large fresh water production capacities, with the aim of supplying small-medium communities in remote areas such as islands. Among the small-scale solar MED plants available in literature, the ‘‘Sol-14’’ plant built in Almerìa represents a best example, for the high conversion efficiency induced by the use of a complex fourteen-effects design in spite of the moderate capacity of the plant. Accurate energy and exergy analyses of the plants have been presented [28]; more recently, a lay-out integrated with a doubleeffect absorption heat pump has been proposed for an effective integration of the solar source during hours with moderate irradiation [29]. The combination of geothermal energy and desalination systems is also diffusely investigated in the relevant literature presenting both theoretical and experimental studies. In particular, a detailed review of geothermal desalination processes is presented by Goosen et al. [25] also showing several applications of such systems in Algeria, Greece and Mexico, concluding that this

3

technology may be very attractive in several parts of the world. Among the different available desalination technologies, MED systems are particularly attractive for a possible integration with geothermal heat sources. In fact, geothermal energy provides a stable and reliable heat supply required for a safe operation of the desalination unit. In addition, typical range of geothermal temperatures (70–90 °C) is ideal for low-temperature MED systems [26]. A case study of a MED system powered by geothermal groundwater (75–90 °C) was developed by Karystsas et al. [26], providing 600– 800 m3/day of fresh water. A numerical study was presented by Manenti et al. [30,31], designing and simulating a MED system powered by geothermal energy for the Island of Pantelleria (South of Sicily, Italy). Some additional geothermal desalination prototypes are also analyzed in literature, showing further possibilities to couple geothermal resources and falling film evaporator [32], HD plants [33] and vacuum membrane distillation [34]. Literature review showed a number of papers investigating complex polygeneration systems. The majority of these papers also focused on the benefits achievable by polygeneration systems in micro grids [35] and/or in small cluster of buildings [36,37]. A number of studies also analyze the possibility to use renewable energy sources in complex renewable polygeneration systems. Son et al. investigated a biomass-fired polygeneration system, integrating a drying process, obtaining an increase by 3.1% in efficiency and 5.5% in power production [38]. Zhou et al. investigated the integration of a biogas generation system, a Fuel Cell system and a green house in a bioethanol plant to form a polygeneration system. The system produces 6 MW electricity and 3.4 heat flow MW [39]. Similar studies are available in literature for biomass-based polygeneration systems. For example, Starfelt et al. integrated cogeneration and lignocellulosic ethanol production [40]. Le Truong et al. investigated a biomass-based polygeneration system supplying district heat, electricity, pellets and motor fuels [41]. Among these studies, only one of them focused on the combination of renewable energy sources and products similar the ones analyzed in the present paper. In particular, he integration of renewable sources and water desalination was recently investigated by Kyriakarakos et al. [42]. This system is based on photovoltaics and a wind turbine, also including a battery bank, a Proton Exchange Membrane (PEM) fuel cell, a PEM electrolyzer, a metal hydride tank and a reverse osmosis desalination unit using energy recovery and a control system. Authors showed that a feasible polygeneration microgrid adapted to a small island is financially profitable with a probability of 90% for the present and 100% at the medium term [42]. However, the proposed arrangement is different than in the present case, as the desalination process is driven by electricity (while heat is used in the present paper) and solar energy is coupled with wind (while geothermal energy is used in this paper). All the above mentioned technologies (SHC, CPVT, GW and MED) are considered in this work, designing and simulating a novel system able to produce simultaneously electrical energy, thermal energy for heating and cooling, and fresh water. As widely discussed above, while dozens of papers are available in literature investigate separately such devices and/or polygeneration systems based on renewable sources [43,44], in the authors’ knowledge no paper is available presenting a comprehensive assessment of this kind of system. In the present paper a reliable system layout was designed and simulated in TRNSYS environment. A basic case study is presented for the island of Pantelleria in the South of Italy (36°470 N, 11°590 E). Further case studies are also presented for additional Italian volcanic islands in the Mediterranean Sea. An economic analysis was also performed, aiming at determining the set of synthesis/design parameters able to maximize the economic profitability of the system. Useful indications for system designers and manufacturers were obtained, in order to improve the performance of the system under investigation.

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

4

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

2. System layout As mentioned in the previous section, the system investigated in this paper combines SHC, CPVT, Geothermal Well (GW) and MED technologies. So, it can be considered as a polygeneration system [43,45], providing simultaneously different energies (electricity, cool and heat) and mass flows (desalinated water). For the sake of brevity, the system will be named RPS (Renewable Polygeneration System). A simplified layout of the system under investigation is shown in Fig. 1, where only the main components are displayed. The RPS includes 9 operating fluids, and corresponding loops: 1. Solar Collector Fluid, SCF: pressurized water flowing from the source sides of the tanks to the solar field. 2. Hot Fluid, HF: pressurized water flowing from the load sides of the tanks to the devices using solar thermal energy. 3. Cooling Water, CW: water flowing in the condenser and absorber of the Absorption Chiller (ACH). 4. Chilled Water, CHW: water flowing in the evaporator of the Absorption Chiller (ACH), supplying space cooling devices. 5. Domestic Hot Water, DHW: water supplying sanitary devices. 6. Hot Water, HW: water supplying space heating devices. 7. SeaWater, SW: water supplied to the MED, in order to be desalinated, or to the heat exchanger used for cooling the ACH. 8. Desalinated Water, DW: fresh water produced by the MED and supplied to final users. 9. Geothermal Well Fluid, GW: hot geothermal water supplying heat to the MED and/or to HE4 (producing DHW). The following main components are included in the system:  A Geothermal Well, GW, consisting in a set of 10 wells (80 m depth, 350 mm diameter) each one equipped with a submerged pump (nominal flow rate 12 m3/h), pumping hot geothermal water from underground at 85 °C.

 A Solar Collector field, CPVT, consisting of concentrating parabolic trough solar collectors whose absorber is covered by a triple-junction PV layer; the beam radiation is concentrated on a triangular receiver, placed on the focus of the parabola, on which a multi-junction PV panel is laminated; the triangular receiver is equipped with an internal tube, in which a cooling fluid flows; the system is also equipped with a one-axis tracking system, typical of Parabolic Trough Solar Thermal Collectors; the PVT can operate up to 100 °C.  A Thermal Storage system (TK1), supplying heat for space heating and cooling purposes, consisting of a set of stratified vertical hot storage tanks, equipped with inlet stratification devices: the entering position of the inlet fluid is varied so that fluid and tank temperature are equal.  A Thermal Storage system (TK2), supplying solar heat for seawater desalination, consisting of a set of stratified vertical hot storage tanks, equipped with inlet stratification devices: the entering position of the inlet fluid is varied so that fluid and tank temperature are equal.  A plate-fin heat exchanger in the solar loop (HE1), used to produce Domestic Hot Water when the solar irradiation is higher than the ACH/HE2/MED thermal demand.  A plate-fin heat exchanger in the HW loop (HE2), transferring heat from the HF to the hot water (CHW) to be supplied to the fan-coils during the winter.  A plate-fin heat exchanger in the GW/DHW loops (HE3), using the geothermal hot water, GW, to supply heat to the Hot Fluid, HF, which heats the MED.  A plate-fin heat exchanger in the GW/DHW loops (HE4), producing DHW using the hot stream exiting from HE3.  A plate-fin heat exchanger in the CW/SW loops (HE5), cooling the CW loop using the seawater, SW.  A LiBr–H2O single-effect absorption chiller (ACH), whose generator is fed by the hot fluid (HF) provided by the solar field; the condenser and the absorber of the ACH are cooled by seawater, through the cooling water loop (CW).

Fig. 1. System layout.

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

 A Multiple Effect Distillation (MED) desalination unit, producing desalinated water from seawater.  Some fixed-volume pumps (P1, P3, P4, P5, P6) for the HF, HW, SW, CHW and CW loops.  A variable-speed pump (P2) for the SCF loop.  A variable-speed pump (P7) for the Geothermal Wells.  An inertial chilled/hot water storage tank (TK3), included in order to reduce the number of start-up and shut-down events for the absorption chiller ACH.  Some Balance of the Plant (BOP) equipment (the majority not displayed in Fig. 1 for sake of simplicity), such as pipes, mixers, diverters, valves, and controllers required for the system operations. A number of additional mandatory components (not displayed in Fig. 1) have been also implemented in the simulation model, such as: controllers (feedback, proportional and on/off), schedulers (daily and seasonal), weather databases, printers, integrators, etc. Fig. 1 also includes some of the set point temperatures and/or temperature limits, both for the summer (s) and winter (w) operation, which could be helpful to show the RPS operating principle. The solar irradiation is converted into electrical energy and thermal energy by the CPVT. The electrical energy produced by the CPVT is used to drive all the electrical devices included in the system; electrical energy in excess is delivered to the grid. The thermal energy is used to increase the CPVT outlet temperature up to the fixed set point, Tset,CPVT. Two different values are assumed for such set-point, Tset,CPVT,summ and Tset,CPVT,wint, during the summer and winter operation modes, respectively. Set-point temperatures are achieved by the variable-speed pump P2, regulated by means of a feedback controller operating with a secant method. Such controller also deactivates the flow in the CPVT when the radiation is low, potentially determining a cooling effect of the fluid flowing in the CPVT. However, the desired set-point temperature is not always achieved. In fact, when solar radiation is low and/or user demand is high, the CPVT outlet temperature may be also lower than the corresponding set-point. Conversely, when the radiation is high and/or the user demand is low, such temperature may be higher than the set-point. Eventually, CPVT outlet temperature may approach boiling temperature, around 120 °C. This circumstance is prevented by HE1: when CPVT outlet temperature is higher than 100 °C, the heat exchanger HE1 is activated, producing Domestic Hot Water (DHW) and simultaneously cooling the SCF to 100 °C. In other words, the eventual solar heat in excess to be dissipated is converted into DHW. Conversely, when CPVT outlet temperature is below 100 °C, HE1 is by-passed. The fluid exiting from the CPVT and/or by HE1 may supply TK1 or TK2. Note that both TK1 and TK2 are stratified storage tanks, therefore the inlet flows enter the tank at the node having the same temperature of the inlet flow. Such arrangement is commonly used in solar heating and cooling systems allowing one to achieve a better system performance as shown in references [46]. TK1/TK2 flow is managed by the Diverter D1 and mixer M1. A sensor measures the temperature at the top side of TK1; when such temperature is lower than a fixed set-point, TTK,set, assumed equal to Tset,CPVT  DTTK, and/or when this temperature is below MED design temperature (75 °C), the valves supply the SCF only to TK1. Conversely, when the top temperature of TK1 reaches the maximum between TTK,set and 75 °C, the SCF is brought to the TK2, supplying heat to the MED subsystem. On the other side of TK2, P1 flow passes through a plate-fin heat exchanger (HE3) providing eventual auxiliary heat required to reach MED set point inlet temperature (75 °C). In particular, a feedback controller manages the variable speed pump P7, varying the GW flow rate in order to achieve the above-mentioned HF setpoint temperature (75 °C). The temperature of the geothermal fluid exiting from HE3 is high enough to supply heat to heat exchanger HE4

5

producing Domestic Hot Water (DHW). In other words, when the temperature at the top side of TK2 is below 75 °C, the fluid is heated up to 75 °C by the HE3, also producing DHW in HE4. Conversely, the P7 is deactivated (and no DHW is produced by HE4) when TK2 outlet temperature is higher than 75 °C. Note also that the geothermal fluid exiting from HE4 is set at 50 °C, since this fluid may be subsequently used for thermal baths or re-injected underground. The MED subsystem uses the solar/geothermal heat to convert seawater into desalinated water. In fact, it is much more convenient to desalinate cold seawater than the hot geothermal water exiting from HE4, due to high content in salinity, sulfur and other toxic substances (it is a low-grade radioactive water) of the geothermal water. This is also obvious when we consider that the evaporators of the MED section rapidly decrease their performance and increase their maintenance (due to the higher frequency of chemical cleanings) and water-pretreatment cost when processing a feed water with a high content of sulfur contaminants/solid particles. As it concerns the energy balance, in particular, when solar radiation is scarce and/or when it is used by TK1 (for space heating and cooling purposes), part of the thermal energy required by the MED is supplied by the geothermal well, GW. On the load side of tank TK1, HF is pumped by the pump P1 to the users. In particular, the summer/winter operation modes is managed by D2 and M2: during the summer the hot flow coming from TK1 supplies the generator of the ACH; conversely, during the winter, such flow supplies the hot side of the heat exchanger HE2. The pump P1 is managed by a controller, measuring the top temperature of TK1. When such temperature is lower than a minimum allowable value (45 °C and 75 °C, respectively in winter and in summer), P1 is deactivated, so that the solar energy can heat TK1 up to the minimum values required; in this way, the temperature of the hot fluid used to drive the ACH is always higher than the minimum value of 75 °C. Similarly, during the winter, the temperature of the HF going to HE2 must be higher than 45 °C. The condenser and absorber of the ACH are indirectly cooled by seawater, avoiding the use of a cooling tower. An indirect exchange by heat exchanger HE5 is required in order to avoid corrosion of the ACH by seawater. During the summer, the pump P4 is active, supplying chilled water for space cooling. Conversely, during the winter the pump P5 supplies hot water to the end user, for space heating. The valves M4 and D4 operate the seasonal switch. Finally, an inertial tank TK3 is used in order to simulate the piping capacity. Such complex control system is managed by a number of different types of controllers, on/off with hysteresis, feedback, proportional, etc. Obviously, all these controllers operate using appropriate dead-bands (around 2–4 °C), required to reduce the number of activations/deactivations. Finally, all the devices, including the MED subsystem, are completely deactivated when the incident solar radiation is zero. As clearly shown in this section, the selected system is very complex integrating a number of different technologies. Usually, such complex systems determine a scarce utilization of their components. However, this cannot occur in the system proposed in this paper as a consequence of the efficient control strategy discussed before. There is no possibility to dissipate any of the selected renewable sources and all the components are used at the maximum of their capacities, compatibly with user demand. In fact, the above discussed control strategy determines a priority order for energy utilization. In particular solar energy is first used for space heating/cooling purposes. Then, when such request is satisfied, solar energy is used to drive the desalination process. When this second demand is also completely saturated, solar energy in excess is used for Domestic Hot Water. However, in a properly designed system this last occurrence is very rare. As for geothermal energy, it is used first to drive desalination, then to produce

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

6

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

Domestic Hot Water and finally residual geothermal water is used for thermal baths. This complex network, along with the designed control strategies, allow one to achieve a full utilization of sources and components. Also, being fresh water a product that may be easily stored (i.e. production is not required to be simultaneous to consumption), a further ‘‘degree of freedom’’ is available for plant operation. Finally, it worth noting that this system is supposed to be installed in parallel with the existing energy conversion systems. Therefore, when the capacity of the RPS is lower than user demand, part of the energy will be supplied by the existing conversion systems.

day)[59]. However, examining in details the transient behavior of MED systems during start-up and shut-down (respectively at morning and late afternoon) would highly increase the complexity of the model, providing no significant improvements in the reliability of results in terms of daily freshwater production. Hence, below in the paper a stationary model for the MED section is assumed. The model essentially includes: – Mass balances for the seawater and the dissolved salts. – Energy balances. – Heat transfer equations.

3. Simulation model The RPS polygeneration system described in the previous section was dynamically simulated by TRNSYS, which is a well-known software diffusely adopted for both commercial and academic purposes, including a large library of built-in components, often validated by experimental data. This model is based on a wellknown approach, assuming quasi-stationary conditions for the non-capacitive components, whereas the capacitive ones (pipes, tanks, etc.) account for the unsteady terms in the energy balances. Such approach allows one to simulate in detail complex systems when the simulation time-step is sufficiently large (minutes). Conversely, this approach does not allow one to simulate in detail all the transient phenomena occurring at the start-up and shut-down of the components. However, these phenomena are commonly considered negligible in the operation of such system. As a consequence, the approach implemented in TRNSYS is widely adopted in literature for the simulation of several energy systems [47]. The RPS layout investigated in this paper was originated from that developed in previous works [48,49], where the models of both built-in and user-developed components are described in detail. As discussed in references [50,51], the majority of the models used for the components (e.g. pumps, mixers, diverters, valves, controllers, auxiliary heater, absorption chiller, cooling tower, plate-fin heat exchanger, building, etc.) were taken from TRNSYS library. Some new models (defined as ‘‘types’’, in TRNSYS) were developed by the authors in Fortran and then linked to TRNSYS: DHW heat exchangers, Primary Energy Calculator, Economic Costs Calculator, CPVT, MED, and others. In particular, new TRNSYS types have been created for the CPVT and the MED. The model of the CPVT was recently presented in literature [10,52], whereas the MED model will be described more in details. As for the remaining new TRNSYS types developed by the authors, the reader is referred to references [53,54] for further details. 3.1. MED model A limited number of small scale MED systems have been presented in literature, usually fed via low temperature heat cascades from cogeneration systems [55] or heat produced by solar collector fields [56]. The design and operation of MED plants fed by solar sources highly depend on the peculiar characteristics of the solar field and the desired power to water ratio; in most cases vertical stacks have been adopted, where a number of ‘‘effects’’ are vertically integrated and preheating of the feed water is also implemented [57]. Contrarily to large scale industrial MED systems that are operated 24 h/day, a small scale unit operated by solar sources may produce fresh water for 8–12 h per day (depending on the capacity of the heat storage tank and on the eventual integration of hot water supply by auxiliary components) [58]; in fact, very few examples of solar desalination systems coupled with heat storage and operated 24 h per day are available in literature, but only for extremely low scale applications (capacities up to some m3 per

Being fresh water production based on a renewable energy source, it is inconvenient to assume a plant scheme designed to achieve extremely high Performance Ratios (PRs); then, a total number of effects equal to 8 is fixed. Also, let us assume as salinity of the feed an average value for the seawater in the Pantelleria area, i.e. Total Dissolved Solids (TDS) equal to 38,000 ppm, and a plant design based on the following criteria: – Temperature of the last effect, at nominal conditions: TB,8 = 40 °C. – Salinity of the rejected brine: XB,8 = 72,000 ppm. Also, as it concerns the temperature of intake seawater, the seasonal fluctuation between 13 °C and 26 °C is kept into account by assuming an average 20 °C value. The model is briefly presented in three blocks: (1) First effect, where the heat from the tank TK2 coupled to the solar system is released to feed the process, (2) Effects from 2 to 8, (3) Condenser. The plant configuration is the feed-forward feed (where the brine and the vapor flowing from the first to the last effect), with partial feed pre-heating performed by inter-stage heat exchangers fed by the flashing vapor of the stage itself. Such configuration presents, in general, the advantage of reducing the risk of scaling (being the higher brine concentrations in the last stages at low temperature) and a good energetic performance.

3.1.1. First effect The simplified scheme of the first effect is shown in Fig. 2. The _ motiv e , enters the effect at a temperature hot water flow rate, M Tmotive,in and cools down to Tmotive,out; the released heat is first used to pre-heat the feed from its inlet temperature, Tf,2, to the temperature level of the effect, TB,1; a second and more relevant fraction of the released heat is then used to separate a fraction of distillate, _ D;1 . The distillate is assumed to be completely salts-free, i.e. M XD,1 = 0; this assumption is realistic, being usually the salinity XD,1 in the order of 5–15 ppm. The water and salts mass balances are written at ‘‘1st-effect level’’, respectively as follows:

_ D;1 þ M _ B;1 ¼ M _f M

ð1Þ

_ D;1  X D;1 þ M _ B;1  X B;1 ¼ M _ f  Xf M

ð2Þ

As it concerns the heat transfer equations, two distinct sets of equations are written for the ‘‘sensible heat transfer area’’, i.e. the upper part of the tube bundle where the feed pre-heats from Tf,2 to TB,1, and the ‘‘latent heat transfer area’’, i.e. the lower part of the bundle where part of the feed evaporates. Equations are based on the effectiveness-NTU method and, being based on well-known expressions derived from literature [60,61], they are not presented here for the sake of brevity.

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

Fig. 3. Simplified scheme of a generic effect from 2 to 8.

Fig. 2. Simplified scheme of the 1st effect, hot water powered.

Energy balances for the motive hot water and the feed are expressed as follows, respectively for the ‘‘sensible’’ and ‘‘latent’’ regions of the heat exchanger:

_ motiv e cp;motiv e  ðT motiv e;in  T motiv e;intermediate Þ Q_ sensible ¼ M _ f  cp;f  ðT B;1  T f ;2 Þ ¼M

ð3Þ

_ motiv e cp;motiv e  ðT motiv e;intermediate  T motiv e;out Þ Q_ latent ¼ M _ D;1  kD;1 ¼M

ð4Þ

In Eq. (4) kD,1 represents the latent heat of vaporization at the temperature of the distillate, TD,1. Actually, this temperature is not coincident with the temperature TB,1 in the effect, but it is lower by the Boiling Point Elevation (BPE) due to the absence of dissolved salts in the distillate:

T D;1 ¼ T B;1  BPE1

ð5Þ

According to Ref. [62], BPE can be expressed as a polynomial function of water salinity and temperature; in the present paper a quadratic expression was adopted, that has been proven sufficiently reliable over the temperature range of interest (40 °C, 80 °C):

BPE1 ¼ a1 T B;1 X B;1 þ a2 X B;1 þ a3 T B;1 þ a4

7

ð6Þ

with a1 = 7  108, a2 = 8.67  106, a3 = 7.74  104, a4 = 8.14  102 and TB,1 and XB,1 respectively expressed in Celsius degrees and parts per million. Pressure drops in the demister are neglected. 3.1.2. Generic effect (from 2 to 8) All the effects from the 2nd to the 8th can be modeled similarly, since the phenomena occurring in the different effects are absolutely similar. In Fig. 3 a simplified scheme of a generic effect is shown. _ D;i1 produced at the In the ith effect (i = 2–8) the distillate M (i  1)th effect condensates inside the tubes, releasing its latent heat of condensation; the brine exiting the (i  1)th effect is sprayed on the outer side of the tube bundles, and a new distillate

_ D;i is produced. The pressure in the ith effect is the satuflow rate M ration pressure corresponding to the temperature TB,i. The driving force for heat transfer is the difference ‘‘TD,i1  TB,i’’; once determined TD,i  1 from the study of the (i  1)th effect, TB,i can be determined by the heat transfer equation applied to the evaporator. Being the inlet temperature of the motive hot water below 100 °C, and being the temperature in the effects gradually decreasing from the 1st to the 8th, all the effects are operated under vacuum conditions and auxiliary components like vacuum pumps or, in larger MED systems, vacuum ejectors must be used during startup at morning; in this paper the MED section is examined in its stationary operation and these auxiliary components are thus excluded from our study (being the contribute related with removal of non condensable gases of moderate energetic relevance). _ B;i1 of the brine admitFirst, let us examine the inlet flow rate M ted from the (i  1)th effect, at a TB,i  1 temperature slightly higher than the saturation temperature TB,i in the ith effect; this water flow undergoes an initial flash process, where the following amount of distillate is approximately separated by flash:

_ D;flashinlet\i" ffi cp;B;i1  ðT B;i1  T B;i Þ M kD;i

ð7Þ

The remaining brine falls down on the outer side of the tubes bundle, absorbing the latent heat of condensation released by the distillate of the previous effect and thus separating some distillate. Hence, all effects from 2 to 8 share in common a same heat load, Q_ latent , since no sensible heat is here transferred (differently than in the 1st effect) and both fluids exchanging heat actually follow an approximately iso-thermal process. The amount of distillate produced by evaporation can be calculated by:

_ D;i1  kD;i1 ¼ M _ D;i  kD;i Q_ latent ¼ M

ð8Þ

Evidently, in each effect a slightly lower distillate flow rate is produced compared to the previous effect, since the latent heat of vaporization reduces with temperature, i.e. kD;i;1 < kD;i . The heat transfer equation is written as:

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

8

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

Q_ latent ¼ U i  Ai  ðT D;i1  T B;i Þ

ð9Þ

where some notes should be given on the overall heat transfer coefficient Ui. An analytical evaluation of the heat transfer coefficient is quite complex, because in ‘‘falling film evaporators’’ very complex relations are required to properly reflect the number of variables influencing the convective heat transfer coefficient on the outer side of the tube bundle; also, the formation of scaling on the outer tubes surface should be in general kept into account. More frequently, typical values derived from empirical observations are implemented in MED models. Basing on a large set of data, in [62] empirical correlations are presented for both the cases of ‘‘clean’’ and ‘‘fouled’’ evaporators; in this paper an average figure, approximately valid for plants with an adequate maintenance, is adopted:

  i2 for i ¼ 2 to 8 Ui ¼ U  1  a  6

ð10Þ

As concerns the heat transfer equation, we have:

Q_ condenser ¼ U condenser  Acondenser  DT ml;condenser

ð15Þ

Again empirical correlations provided in [62] have been adopted to evaluate Ucondenser; according to the constraints imposed (TD,i approximately equal to 40 °C), a value Ucondenser = 2.6 kW/m2 °C can be assumed. 3.1.4. System level Once set the surface of heat transfer areas, the mass flow rate, the inlet temperature of the motive hot water and some other settings, the analytical model presented above allows one to determine all the operating results. The adopted settings are presented in Table 1. However, some variables at system level are introduced to allow for an easier interpretation of results. _ D;total is introduced, which provides the First of all, a variable M total fresh water production rate: 8 8 X X _ D;i þ _ D;flashinlet\j" M M

!

Eq. (10) suggests us that the heat transfer coefficient gradually decreases from the 2nd to the last effect. Over the examined temperature range, appropriate values for the constants U* and the semi-empirical coefficient a can be respectively fixed as 3.8 kW/m2 °C and 0.3 as suggested in Ref. [62], thus resulting in U values gradually decreasing from U2 = 3.8 kW/m2 °C down to U8 = 2.66 kW/m2 °C. Of course, for each effect the mass balances for water and salts can be written as follows:

In Eq. (16) the constant b is introduced to account for the remineralization process: being the distillate salts free, it cannot be directly used as drinkable water; a preliminary mixing with an appropriate amount of seawater (after disinfection) is necessary, thus slightly increasing the total amount of fresh water produced. b is calculated by:

_ D;i þ M _ D;flashinlet\i" þ M _ B;i ¼ M _ B;i1 M

ð11Þ

bffi1þ

  _ D;i þ M _ D;flashinlet\i"  X D;i þ M _ B;i  X B;i ¼ M _ B;i1  X B;i1 M

ð12Þ

Assuming Xf = 38,000 ppm, XD = 0 and, as recommended from the World Health Organization, Xdrinkable = 420 ppm, we obtain a value b = 1.011. The total heat transfer area is:

From Eqs. (11) and (12), being X D;i ffi 0, it is evident that the salinity of the brine significantly increases from the 1st to the 8th effect; being the brine finally disposed from the last effect back to the sea, the need to design the system so as to achieve a concentration XB,8 not higher than 70–75 thousands ppm is evident due to environmental concerns. Finally, the ‘‘feed pre-heating’’ section is shown on the upper part in Fig. 3. Here the distillate flashed from the inlet brine is condensed, releasing some heat to pre-heat the feed from Tf,i+1 to Tf,i; being all effects from 2 to 8 equipped with such feed pre-heater, it is evident the significant increase the feed inlet temperature at the 1st effect, Tf,2, thus reducing its sensible heat load and increasing the amount of distillate produced per kW h released from the motive hot water. The energy balance of feed pre-heaters is written as follows:

_ D;flashinlet\i"  kD;i ¼ M _ f  cp;f  ðT f ;i  T f ;iþ1 Þ Q_ preheater;i ¼ M

ð13Þ

3.1.3. Condenser The condenser in MED systems serves as an economizer: it allows to recover part of the latent heat released from the conden_ D;8 produced at the last effect, using it to sation of the distillate M pre-heat the water from Tintake to Tf,8+1 and thus contributing to increase the final temperature Tf,2 of the feed admitted to the 1st _ CW must be supeffect. Usually, an additional cooling water flow M plied to absorb the excess heat to be finally wasted. In a typical configuration, the condensing fluid circulates shell side, while the seawater (cooling + feed) tube side, as shown in Fig. 4. As concerns the condenser, no mass balances must be written, since this component does not involve any separation process. Then, it essentially requires an energy balance and a heat transfer equation. The energy balance is written as follows:

_ D;8  kD;8 ¼ ðM _ f þM _ cw Þ  cp;f  ðT f ;8þ1  T intake Þ Q_ condenser ¼ M

ð14Þ

_ D;total ¼ b  M

i¼1

X drinkable  X D X f  X drinkable

Atotal ¼ Asensible þ Alatent þ

ð16Þ

j¼2

ð17Þ

8 X Ai þ Acondenser

ð18Þ

i¼2

In Eq. (18) a simplification is obtained when we consider that usually from the 2nd to the last effect same heat transfer areas are installed. In order to evaluate the operating costs, an indicator of the specific energy consumption is adopted:

eMED ¼

_ motiv e  cp;motiv e  ðT motiv e;in  T motiv e;out Þ kW h M _ D;total m3 3:6  M

ð19Þ

Finally, in order to calculate the investment cost, a prudential approach was adopted to estimate a reasonable cost figure. According to [63], the cost of the plant can be approximated as a function of the total heat transfer area installed:

J 1MED ¼ c  ðAtotal Þc

ð20Þ

For this particular system, reasonable values for the constants in Eq. (20) are c = 0.95 and c = 300 €/m1.9. An alternative approach to estimate the plant installation cost consists of fixing an average and literature-derived cost for the unit fresh water production capacity; a reasonable value coherent with the results presented in [64] is 800 € per m3/day capacity. To evaluate the installation cost, a 24 h per day operation must be assumed, regardless of the actual operating conditions that, in our case, involve an intermittent operation due to the availability of the solar source. Then, we may calculate:

_ D;total  3:6  24Þ J 2MED ¼ 800  ðM

ð21Þ

For the examined plant the two approaches provide similar values; then, a robust solution is assumed to be represented by the average of the above two values:

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

9

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

Fig. 4. Simplified scheme of the condenser.

Table 1 Main design parameters for the RPS. Parameter

Description

Value

Unit

NCPVT ACPVT

Number of CPVT collector CPVT Aperture Area TK1 Volume/CPVT Area TK2 Volume/ TK1 Volume CPVT outlet set-point temperature, summer CPVT outlet set-point temperature, summer Solar heat is switched from TK1 to TK2 when TK1 top temperature is higher than Tset,CPVT  DTTK P4 flow/Number of CPVT collectors P2 flow/CPVT area Number of Tank nodes Tank heat loss coefficient P3, P8 flow rate P4, P5 flow rate P7 flow rate Geothermal Well fluid temperature GW set point temperature exiting from HE4 SCF set point temperature exiting from HE1, summer SCF set point temperature exiting from HE1, winter DHW set point temperature exiting from HE1 DHW set point temperature exiting from HE4 TK1 temperature minimum temperature for system activation, summer TK1 temperature minimum temperature for system activation, winter TK1 temperature for switching to TK2, summer TK1 temperature for switching to TK2, winter P6 flow Geothermal Well rated heating capacity Rated ACH cooling capacity Rated HE1 heating capacity Rated HE2 heating capacity Rated HE3 heating capacity Rated HE4 heating capacity Rated HE5 heating capacity Rated CPVT heating capacity Rated CPVT electrical capacity Rated ACH Coefficient of Performance Number of MED effects Nominal mass flow rate of motive hot water

400 12 15 1 90 60 4 300 100 5 0.833 2.80  105 1.20  105 1.20  105 85 50 90 60 45 45 75 45 86 56 1.80  105 1200 700 2800 900 1200 3900 2050 2300 800 0.80 8 50

/ m2 L/m2 / °C °C °C kg/h kg/h m2 / W/m2 K kg/h kg/h kg/h °C °C °C °C °C °C °C °C °C °C kg/h kW kW kW kW kW kW kW kW kW / / kg/s

Sensible heat transfer area at 1st effect Latent heat transfer area at 1st effect Heat transfer area in effects from 2 to 8 Salinity of feed water Maximum salinity of disposed brine Heat transfer area at the condenser Mass flow rate of cooling water

10 50 80 38,000 72,000 62 6.75

m2 m2 m2 ppm ppm m2 kg/s

Temperature of seawater at intake facilities

25

°C

vTK1 fTK2 Tset,CPVT,summ Tset,CPVT,wint DTTK qP4 qP2 nTK Utank QP3 = QP8 QP4 = QP5 QP7 TGW TGW,HE4 TSCF,HE1,summer TSCF,HE1,winter TDHW,HE1 TDHW,HE4 Tact,summ Tact,wint TTK,set,summer TTK,set,winter QP6 PGW QACH,c,rated QHE1,rated QHE2,rated QHE3,rated QHE4,rated QHE5,rated QCPVT,rated PCPVT,rated COPACH,rated Neff

_ motiv e M Asensible Alatent Ai Xf XB,8 Acondenser _ cw M Tintake

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

10

J MED ¼

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

J 1MED þ J 2MED 2

ð22Þ

kg CO2 X Q c;i Eel;CPVT  Eel;aux þ kW h i gel;RS COP EHP;RS;c;i gel;RS ! Q h;i Q DHW;i þ þ COP EHP;RS;h;;i gt;RS

DCO2 ¼ 0:305

3.2. Energetic, environmental and economic parameters In order to evaluate the Primary Energy saving, DPE, achieved by the Renewable Polygeneration System (RPS) for each energy flow produced by the RPS, a corresponding Primary Energy consumption was calculated for a Reference System (RS) [65] representing the most common configuration in Pantelleria. With reference to Electric, Heating and Cooling Systems, this can be summarized as:  Domestic Hot Water is produced by an diesel-fired heater (with an average conversion efficiency equal to 0.80).  Space heating energy is produced by an electric heat pump (with an average Coefficient of Performance equal to 3.5).  Space cooling energy is produced by an electric heat pump (with an average Coefficient of Performance equal to 3.0).  Electric energy is produced by a conventional diesel engine, whose efficiency is gel,RS  0.35. Therefore, the overall Primary Energy saved by the RPS is calculated by integration, i.e. considering the sum over all the timesteps [6]:

DPE ¼

X i

Q c;i Q h;i Q DHW;i Eel;CPVT  Eel;aux þ þ þ gel;t COPEHP;RS;i gt;RS gt;RS gel;t

! ð23Þ

The first term on the right-hand side of Eq. (23) is the primary energy consumed by the RS for the production of the same cooling energy (by an Electrical Heat Pump, EHP) of RPS, the second and third ones represent the primary energy consumption of RS due to space heating and DHW production (respectively by the electrical heat pump and by a diesel-fired boiler whose average efficiency is gt,RS); the last term is related to the net electrical energy production achieved by the RPS [49], which is added to the first three terms expressing the energy savings with respect to the Reference System. It is worth noting that all the energy calculations performed in this paper do not consider thermal losses in pipelines. Although such losses may be up to 15% in large high-temperature district heating networks, for the selected system these losses can be considered negligible [66]. In fact, in the present work, two considerations must be taken into account: (i) well-insulated pipelines transport low-temperature fluids (<50 °C); (ii) the district heating/cooling network is very small since the system will be installed close to the island harbor, where a couple of main hotels, municipal public buildings and tourist facilities can use all the thermal and cooling energies produced. As a consequence, thermal losses in pipelines can be here neglected. The model also calculates greenhouse gases emissions and savings, using CO2 LCA (Life Cycle Analysis) emission factors [67]. These factors were taken from EU official data [68] and consider two contributions for the overall CO2 emissions: the first due to the operation of the system and the second due to the CO2 produced during the system manufacturing process. According to the EU data [68], the following LCA CO2 emission factor is assumed for diesel: 0.305 kg CO2 per 1 kW h of thermal energy (referred to the Low Heat Value) produced by the combustion of this fuel. In fact, in the Reference System electricity, space heating and cooling, Domestic Hot Water are directly or indirectly produced by diesel. In addition, according to the above mentioned data [68], LCA CO2 emissions factors of solar collectors and geothermal systems can be neglected. Therefore, in case of RPS, the CO2 emission saving (kg/year) was calculated as:

ð24Þ

A detailed cost model was also implemented in the simulation tool, considering both operating and owning costs. The following equations were used for estimating the investment cost (Ji) of the different components. Such equations were derived from manufacturers data in a wide range of capacities, including the ones adopted in the simulations [54]:

J CPVT ¼ 600ACPVT

ð25Þ

J ACH ¼ 105 P3ACH  0:0393P2ACH þ 244:53P3ACH þ 95494

ð26Þ

J TK ¼ 494:9 þ 0:808V TK

ð27Þ

J pump ¼ 1:08ð0:00000002Q 2pump þ 0:0285Q pump þ 388:14Þ

ð28Þ

 0:78 AHE J HE ¼ 150 0:093

ð29Þ

The capital cost of the geothermal well (80 m depth), including also pump, pipes and insulation is estimated using manufacturer data. The geothermal system consists of 10 wells, each one showing a nominal capacity of 12 m3/h of pumped geothermal water. Each well is equipped with a submerged pump. The geothermal system also includes pipes and heat exchangers. The unit cost of wells was reported at 200 €/m. The cost of each submerged pump is 16 k€. The total cost of the geothermal system is 416 k€. The savings in terms of operating costs, DCop, is strictly related to the energy savings. It is worth noting that a potential additional income could be derived by the geothermal fluid at HE4 outlet, which could be used for thermal baths. However, no data is available regarding the amount of that fluid demanded by the user. Therefore, such income is conservatively neglected. Furthermore, the cash flow must also consider the production of desalinated water. So, using the same procedure reported in similar studies [53], the savings achievable by the RPS with respect to the RS in terms of operating costs are given by: DC op ¼

" X i

! # Q c;i Eel;CPVT  Eel;aux cue þ cut ðQ h;i þ Q DHW;i Þ þ mwat;i cuwat þ COPEHP;RS;i gel;t ð30Þ

The unit cost of electrical energy, cue, was assumed equal to 0.15 €/ kW h; the unit cost of thermal energy, cut, was 0.075 €/kW h. Finally, a unit cost, cuwat, of 1.2 €/m3 was assumed for the desalinated water. The economic performance of the SHC system was estimated in terms of Simple Pay Back period (SPB), Net Present Value (NPV) and Profit Index (PI), in two scenarios: with and without public founding, as defined in reference [6]. Such indexes are calculated assumed 20 years operating life and 5% actualization rate. In the first scenario, the following feed-in tariffs were assumed to be available for the RPS products:    

0.45 €/kW h for electric energy. 0.20 €/kW h for thermal energy. 0.20 €/kW h for cooling energy. 1.0 €/m3 for desalinated water.

Such tariffs are summed to the prices/costs of energy and water. The feed-in tariffs considered in this work are in the order of magnitude of the ones adopted in EU for thermal, cooling and electrical production from solar energy [6]. Conversely, in authors’

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

knowledge no EU Country provides feed-in tariffs for the production of desalinated water, since the majority of funding policies consider the initial cost of the plant. Therefore, the related incentive assumed in this paper is only considered as a future possibility for promoting such technology. 4. Discussion and results As mentioned in the first part of the paper, the final goal of this work is to analyze the energetic and economic feasibility of the proposed RPS system for the island of Pantelleria, which is a small volcanic island located between Sicily and North Africa. In fact, such island suffers of scarcity of potable water and fossil fuels (for electricity, heat and cool production). Simultaneously, the demands of energy and water dramatically increase during the summer season due to high number of tourists. This problem is common for several islands in the Mediterranean Sea. For example, similar problems are also reported for the Aeolian Islands or for Ischia. However, among the cited Italian volcanic islands, Pantelleria is one of the smallest in terms of number of inhabitants. Therefore, this island has been selected in order to develop the case study presented in this section. In fact, in Pantelleria the availability of land for installing solar collectors is large, and simultaneously the energy and water demands are not that high as could occur in the more crowded Ischia and Aeolian Islands. The case study was developed using the weather data of Pantelleria, included in TRNSYS weather database. The three above-mentioned locations (Pantelleria, Ischia and Aeolian Islands) are especially attractive for the system under investigation, due to the large capacities of their geothermal resources. For example, for Pantelleria several studies reported geothermal capacities much higher than the ones assumed in this paper. In fact, the system investigated by Manenti et al. [30,31] in Pantelleria uses about 16 MW thermal power from geothermal resources. Also Sanseverino et al. used about 3.0 MW thermal from geothermal resources in Pantelleria [69]. Similar capacities are also reported for the other locations mentioned before [70]. As mentioned above, the RPS system under investigation includes dozens of components requiring hundreds of synthesis/design parameters to be fixed. Therefore, only the main design/operational parameters are summarized in Table 1. System nominal peak electrical capacity is about 800 kW, whereas space heating and cooling capacities are respectively 700 kW and 900 kW. The system can also supply Domestic Hot Water, corresponding to a nominal capacity of 3900 kW. As concerns fresh water production, the nominal capacity of the _ D;total ¼ 2:71 kg=s. Further details about the values MED section is M of some specific design parameters can be found in reference [10] for the CPVT collector and references [65] for other components. Note that some of the design parameters shown in the previous table are closely related to each other. In particular, a variation of the number of solar collectors determines a proportional variation of several design parameters (ACH and HE capacities, TK volumes, etc.). In fact, all the components were designed parametrically, so that they are automatically re-sized as a function of the solar field capacity. Similarly, a variation of qP4 determines a corresponding proportional variation in the capacity of pumps, ACH and HE, whereas the size of the CPVT field is not affected by such quantity. The system is designed in order to match the thermal demand of the MED (about 1.0 MW). Tanks, solar field, absorption chiller are designed on the basis of this data, as shown in Table 1. 4.1. Winter and summer days As mentioned above, the simulation tool developed in this work allows to analyze the results from both energetic and economic

11

points of view, considering whatever time basis. In particular, in this section results are analyzed dynamically, considering two representative winter and summer days. Then, in next sections the variation during the year of the parameters is investigated by analyzing monthly-integrated quantities. Finally, the overall yearly results are analyzed also varying some of the main design parameter and the location of the plant. The performance of the system under investigation dramatically depends on the weather conditions, being excellent in summer and poor in winter. As usual, the efficiency of the CPVT solar loop is especially sensitive to the availability of the solar radiation. In fact, CPVT thermal may achieve values up to 60% when beam radiation is close to 700 W/m2, decreasing down to 35% when beam radiation is 200 W/m2. Conversely, electrical efficiency is stable around 20%. As mentioned in reference [52], the thermal performance of the selected CPVT is close to the ones reported for similar devices in literature. Conversely, CPVT electrical efficiency is much higher as a consequence of the use of triple-junction PV cells. In addition, the amount of solar thermal energy available dramatically decreases during the winter. As a consequence, in that season the control system is not always able to achieve the CPVT outlet set point temperature, as shown in Fig. 5. Here, the temperature of the water exiting the CPVT solar field approaches the set point temperature (60 °C) only in a part of the day. In order to achieve this setpoint temperature, being low the solar energy availability, the control system reduces the mass flow rate flowing in the CPVT loop, reducing P2 speed. As a consequence, the amount of heat delivered from CPVT to TK1 is small too, as clearly shown in Fig. 5 where the maximum temperature of TK1 (ttop,TK1) is much lower than the temperature of the fluid exiting from CPVT (to,CPVT). Such graph also shows that TK1 minimum temperature (tbot,TK1) is around 45 °C. Conversely, the maximum (ttop,TK2) and minimum (tbot,TK2, not shown in figure) temperatures of TK2 are respectively around 70 °C and 15 °C. This behavior is due to the fact that both TK1 and TK2 are stratified storage tanks. In fact, for the considered winter day, the maximum temperature of TK1 has never achieved the temperature (75 °C) activating M1 and D1 in order to supply solar heat to TK2. As a consequence, TK2 has never been heated by the CPVT loop, determining such low value of its minimum temperature. This is also clear in Fig. 5, where CPVT inlet temperature (tin,CPVT) is always equal to TK1 bottom temperature (tbot,TK1), i.e. P2 always pumps water from TK1. Conversely, TK2 maximum temperature is very close to MED return temperature. Such a large temperature difference in TK2 is due to its stratification strategy and is absolutely common in tanks with limited vertical mixing and coupled with solar systems [71]: the flow coming from MED loop is supplied to the TK2 node whose temperature is equal to the inlet flow temperature. The overall consequence of this strategy is that only the upper part of TK2 is hot, avoiding the heating of the overall TK2 volume. Obviously, such circumstance allows to dramatically reduce heat losses to the environment. Differently from the trends of the solar loop (shown in Fig. 5), the temperatures of the MED loop are very stable as shown in Fig. 6. As always occurs for all the days of the year, the MED subsystem is activated by the same On/Off signal managing the solar loop, i.e. the system is active only when solar radiation is sufficient to heat the fluid in CPVT. As a consequence, this system is activated after sunshine and deactivated before sunset. During this time, the geothermal fluid is always pumped from geothermal well at 85 °C, since the solar heat is never supplied to TK2, as stated before. The geothermal fluid heats the hot water supplying the MED at 75 °C (Tin,MED = To,c,HE3). Then, the geothermal fluid enters HE4 (producing DHW at the cold side) exiting at 50 °C, which is a suitable temperature for thermal baths. Fig. 7 clearly shows the system control strategy for space heating for the selected representative winter day: space heating heat exchanger (HE2) is activated only when

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

12

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

75 ttop,TK1 tbot,TK1

70

tto,HE1 ttop,TK2

Temperature (°C)

65

tin,CPVT to,CPVT

60 55 50 45 40 384

386

388

390

392

394

396

398

400

402

404

406

408

Time (h) Fig. 5. Temperatures (1), winter day.

86 to,MED tin,MED

84

tin,GW

Temperature (°C)

82

to,h,HE3 to,c,HE3

80 78 76 74 72 70 68 384

386

388

390

392

394

396

398

400

402

404

406

408

Time (h) Fig. 6. Temperatures (2), winter day.

52 tin,user to,user

51

Temperature (°C)

50 49 48 47 46 45 44 384

386

388

390

392

394

396

398

400

402

404

406

408

Time (h) Fig. 7. Temperature (3), winter day.

TK1 top temperature is higher than 45 °C, preventing the user to be supplied by a low temperature hot stream. All the graphs showing temperature plots (Figs. 5–7) can be also better interpreted

analyzing thermal and electrical powers (Figs. 8 and 9). In particular, Fig. 8 shows that the availability of beam radiation is very unstable during the day, being dramatically low in the

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

13

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

3000 Itot Ibeam

2500

QHE1 PPVT QPVT

Power (kW)

2000

QHE2 Paux

1500

1000

500

0 384

386

388

390

392

394

396

398

400

402

404

406

408

Time (h) Fig. 8. Powers (1), winter day.

4000

QHE3 QMED

3500

QTK1,in QTK1,o

3000

Power (kW)

QTK2,in QTK2,o

2500

QHE4

2000 1500 1000 500 0 384

386

388

390

392

394

396

398

400

402

404

406

408

Time (h) Fig. 9. Powers (2), winter day.

morning and in the afternoon. Such Figure also shows the effect of TK1 thermal storage: CPVT thermal power (QPVT) produced during the central hours of the day is stored in TK1, which supplies the user (QHE2) in the afternoon with a thermal power higher than CPVT thermal production (QPVT). This can be also seen comparing the heat flows entering and exiting TK1 (QTK1,in and QTK1,o) in Fig. 9. As mentioned before, Fig. 9 shows that TK2, during the selected day, is never heated by the CPVT loop, being null the heat entering TK2 (QTK2,in). Finally, such Figure also shows that in this day the MED is heated only by geothermal energy (QHE3 = QMED) and that the heat recovered by HE4 (QHE4) is dramatically larger than all the other thermal flows. The trends of temperatures and powers completely change during the summer season, as clearly shown in the following analysis. Fig. 10 shows temperatures of the solar loop during a representative summer day. Here, all the temperatures oscillate around the fixed set point temperatures. In particular, CPVT outlet temperature (to,CPVT) is stably around 90 °C whereas TK1 and TK2 top temperatures (ttop,TK1 and ttop,TK2) oscillate respectively around 85 °C and 75 °C. All these oscillations are due to the solar loop control strategy, which continuously switches solar heat to TK1 or to TK2 depending on TK1 top temperature. The oscillations are also more significant for CPVT outlet temperature (to,CPVT) since the feedback controller managing P2 pump, encounters some unavoidable oscillations

when CPVT inlet temperature suddenly varies when the system switches from TK1 to TK2 (or vice versa). The oscillating trend of the hot water supplying TK2 also determines a moderate oscillation of the MED inlet temperature (tin,MED in Fig. 11). These small fluctuations, due to the controllers dead bands, do not generate major changes in MED performance (and in the pressure lying in each effect) and also cause the HW return temperature profile (To,HW,ACH) of the ACH shown in Fig. 12. Conversely, cooling water outlet temperature (To,CW,ACH) and chilled water supply (Tin,user) and return (To,user) temperatures are very stable. During the selected summer day, the availability of beam radiation (Ibeam) is high for a long time, as shown in Fig. 13, also determining a CPVT stable production of thermal and electrical powers (QPVT and PPVT respectively). In addition, such graph also shows that the electricity consumed by the auxiliary devices (Paux) is negligible with respect to the one produced by the CPVT solar field. It is also worth noting that the electrical power produced by the RPS is significantly lower than the one demanded by Pantelleria. In fact, data of Pantelleria report a minimum electrical demand around 3.0 MW, whereas the maximum one is around 7.0 MW. Therefore, all the electrical energy produced by the RPS will be always consumed. Finally, Fig. 14 shows energy flows in MED loop where the MED thermal energy is completely supplied by HE3 only during the early morning and late afternoon. Conversely, during the central hours of the day, the majority of the

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

14

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

95 ttop,TK1 tbot,TK1

Temperature (°C)

90

tto,HE1 tbot,TK2

85

ttop,TK2 tin,CPVT

80

to,CPVT

75 70 65 60 4008

4010

4012

4014

4016

4018

4020

4022

4024

4026

4028

4030

4032

Time (h) Fig. 10. temperature (1), summer day.

88

Temperature (°C)

to,MED

86

tin,MED

84

tin,GW

82

to,c,HE3

to,h,HE3

80 78 76 74 72 70 68 4008

4010

4012

4014

4016

4018

4020

4022

4024

4026

4028

4030

4032

Time (h) Fig. 11. Temperatures (2), summer day.

80 to,CW,ACH

70

to,HW,ACH

60

to,user

Temperature (°C)

tin,user Text

50 40 30 20 10 0 4008

4010

4012

4014

4016

4018

4020

4022

4024

4026

4028

4030

4032

Time (h) Fig. 12. Temperatures (3), summer day.

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

15

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

6000 Itot Ibeam QHE1

5000

PPVT QPVT Qcool,ACH

Power (kW)

4000

Qch,ACH QHE2 Paux

3000 2000 1000 0 4008

4010

4012

4014

4016

4018

4020

4022

4024

4026

4028

4030

4032

Time (h) Fig. 13. Powers (1), summer day.

1200

QHE3 QMED

1000

QTK1,o

Power (kW)

800

600

400

200

0 4008

4010

4012

4014

4016

4018

4020

4022

4024

4026

4028

4030

4032

Time (h) Fig. 14. Powers (2), summer day.

12

4.2. Monthly results

10

In order to improve the interpretation of the results, the oscillating trends shown before are mitigated, integrating the energy flows on a monthly basis. This analysis is shown in Figs. 15–19. In particular, Fig. 15 shows the dramatic variation of both beam and total solar energy availability during the year, determining also

Energy (kWh/month)

12

x 10

5

Energy (kWh/month)

thermal energy required to drive the MED subsystem is supplied by the solar loop via the tank TK2.

x 10

5

8

Itot Ibeam

6

PPVT Paux

4 2 0

1

2

3

4

5

6

7

8

9

10

11

12

Month

10

Itot

8

Ibeam

6

QPVT

Fig. 16. Monthly electrical energy.

QHE1 Qch,ACH

4

Qcool,ACH QHE2

2 0 1

2

3

4

5

6

7

8

9

10

11

12

Month Fig. 15. Monthly thermal energy: CPVT, ACH and HE 2 loops.

the corresponding reduction of PVT thermal energy (QPVT) and electrical energy (PPVT in Fig. 16) during the winter. The monthly production of electrical energy is significantly lower than the total demand in Pantelleria. In fact, the monthly demand of electricity in the island oscillates around 3.5 GW h/month for all the months except for July, August and September, when the demands are respectively 4.5, 5.0 and 4.0 GW h/month. This graph also shows that the RPS can produce an amount of cooling energy (Qch,ACH) much higher than the space heating one QHE2. Finally, the amount

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

16

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

x 10 5

Energy (kWh/month)

12 10 8

QHE3

6

QTK1

QHE4 QTK2

4

QMED

2 0

1

2

3

4

5

6

7

8

9

10

11

12

Month Fig. 17. Monthly thermal energy: TK and MED loops.

4

x 10

2.5

mwat (kg/month)

mwat,prod mwat,dem

2 1.5 1 0.5 0

1

2

3

4

5

6

7

8

9

10

11

12

Month Fig. 18. Monthly water production and demand.

1.4 1.2

ηt,CPVT

1

ηt,CPVT

dramatically increases due to the high number of tourists. The adopted simulation tool allows us to observe that the examined plant design is designed to cover a part of water and energy demands, being thus necessary to ensure additional supply to cover the community needs. Finally, the energetic performance of the systems are summarized in Fig. 19, which displays some non-dimensional efficiency parameters. Here, it is worth noting that the amount of solar heat delivered to the MED (QTK2/QCPVT) is null or negligible in January, February, March and December, increasing around 0.30 in the middle season. In summer this parameter is below 0.20, due to large amount of energy required for space cooling. Consequently, the ratios QHE3/QMED and QTK1/ QCPVT are specular with respect to QTK2/QCPVT. Fig. 19 also shows that the considered CPVT solar field provides ultra-high energy conversion rates, being electrical (gel,CPVT) and thermal (gt,CPVT) efficiencies respectively higher than 20% and 50%. This is a very good result considering that the operating temperature of the CPVT solar field is high (from 60 °C to 90 °C, depending on the season). However, the previous efficiencies (as common use in concentrating solar system) are calculated with respect to the beam solar radiation. When the total radiation is considered in this calculation, electrical (gel,CPVT*) and thermal (gt,CPVT*) efficiencies decrease respectively around 12% and 30%. As for heat exchangers, results showed that their heat losses toward the environment are absolutely negligible with respect to their heat transfer rate. As expected, their effectiveness [60] is high for the heat exchangers operating with small temperature differences between hot and cold streams. Conversely, the effectiveness dramatically decreases for recuperative heat exchangers operating with large temperature differences. In fact, the yearly average effectiveness of HE1 is very low (35%) where the hot stream is around 90 °C and the cold one is at an average temperature of 30 °C. Conversely, in HE2, the temperatures of the hot and cold streams are very close, determining an average effectiveness of 90%. Calculated effectivenesses for HE3, HE4, HE5 are respectively: 63%, 51%, 50%.

Parameter (/)

*

ηel,CPVT

0.8

* ηel,CPVT

0.6

COPACH QHE3/QMED

0.4

QTK1/QCPVT

0.2 0

QTK2/QCPVT

1

2

3

4

5

6

7

8

9

10 11 12

Month Fig. 19. Monthly efficiency parameters.

of energy delivered to heat exchanger HE1 (which is activated only when CPVT outlet temperature is higher than 100 °C) is null or negligible for all the months. Fig. 17 shows thermal energies involved in TK and MED loops. It is worth noting that the amount of thermal energy demanded by the MED (QMED) increases during the summer only as a consequence of the higher number of operating hours achieved in that season. This increase does not necessarily determine a corresponding growth of the heat demanded to the geothermal heat source (QHE3), since during the summer the contribution of solar energy is also higher (QTK2). The increase of operation time also causes a corresponding increase of the mass of desalinated water produced as shown in Fig. 18. Such Figure also shows that the amount of water produced by the RPS is about 5–6% of the total amount of water demand by the island. Although MED water production is maximum in summer, its contribution to the island water demand is maximum in May and in December since in summer island water demand

4.3. Yearly results and economic analysis In order to calculate the economic profitability and the energetic indexes of the proposed polygeneration system, results must be analyzed on a yearly basis (Table 2). From the energetic point of view, such table summarizes the yearly results showing basically the same results discussed in the monthly analysis above presented. Conversely, Table 2 shows some additional findings from the environmental and economic points of view. The system allows to save about 4870 t/y of CO2 emissions, as a consequence of the avoided use of a large amount of diesel fuel. In addition, the environmental analysis must also consider that the system allows to save 4.18  107 kg of fresh water per year (being such an amount of water produced by the MED unit). Finally, the expected environmental and energetic savings are even higher than the ones reported in this paper: in fact, the on-site production of fresh water by seawater desalination allows to save a large amount of fuel (and the related emission), which is usually consumed by the conventional desalination plants and the ships transporting water to the islands, especially during the summer period. The total capital cost (Jtot) of the system is about 3.95 M€, where the most expensive component is the solar field (2.9 M€). The annual income (DC), determined by the productions of electricity, cool, heat and potable water is 1.14 M€. In this case, the simple pay back period (SPB) is about 3.45 years, whereas the Net Present Value (NPV) is 10.1 M€. The ratio between the NPV and the capital cost (Jtot) is the Profit Index (PI), equal to 2.56. Such value is extremely high, showing the good economic profitability of the system under investigation. Considering the possible public funding (feed in tariffs) for the production of energy and water from renewable

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx Table 2 Annual results. Param.

PANTEL.

AEOLIAN

ISCHIA

Unit

Itot Ibeam PCPVT Paux QCPVT QC,ACH QHE1 QHE5 QHE2 QTK1 QTK2 QHE3 QMED QHE4 DPE DCO2 mwat DC Jtot SPB SPBft NPV PI NPVft PIft

8.98E+06 5.18E+06 1.06E+06 1.45E+05 2.67E+06 1.29E+06 2.65E+02 2.94E+06 5.70E+05 2.25E+06 4.55E+05 3.63E+06 4.02E+06 1.13E+07 1.60E+07 4.87E + 06 4.18E+07 1.13E+06 3.95E+06 3.45E+00 2.00E+00 1.01E+07 2.56E+00 2.07E+07 5.24E+00 2.05E01 1.18E01

9.79E+06 5.97E+06 1.22E+06 1.50E+05 3.18E+06 1.62E+06 1.33E+02 3.70E+06 5.67E+05 2.66E+06 5.33E+05 3.56E+06 4.02E+06 1.09E+07 1.57E+07 4.80E+06 4.18E+07 1.14E+06 3.95E+06 3.42E+00 1.87E+00 1.03E+07 2.59E+00 2.25E+07 1.21E+07 2.05E01 1.25E01

9.23E+06 5.69E+06 1.17E+06 1.46E+05 3.01E+06 1.41E+06 5.27E+02 3.20E+06 5.37E+05 2.36E+06 6.77E+05 3.40E+06 4.00E+06 1.09E+07 1.55E+07 4.73E+06 4.16E+07 1.11E+06 3.95E+06 3.50E+00 1.96E+00 9.94E+06 2.51E+00 2.13E+07 1.09E+07 2.05E01 1.26E01

kW h/y kW h/y kW h/y kW h/y kW h/y kW h/y kW h/y kW h/y kW h/y kW h/y kW h/y kW h/y kW h/y kW h/y kW h/y t/y kg/y €/y € y y € / € / / /

5.16E01 2.98E01

5.32E01 3.24E01

5.29E01 3.26E01

/ /

7.90E01 9.05E01

7.90E01 8.85E01

7.90E01 8.50E01

/ /

gel,CPVT gel;CPVT gt,CPVT gt;CPVT COPACH QHE3/QMED

energy sources, the corresponding economic indexes (SPBft, NPVft, PIft in Table 2) are respectively: 2.0 years, 20.7 M€ and 5.42. Therefore, results show that the system under investigation is extremely profitable from the economic point of view, even without funding. It is worth noting that this result is obtained assuming a predicted future cost for CPVT collectors (achievable in case of massive commercialization). In addition, such result is also greatly affected by the assumption to use all the thermal, electrical and cooling energy produced. As discussed later, a variation of the amount of energy/water demanded by the user dramatically affects the economic performance of the system.

17

simple pay back period of about 3.4 years. Thus, it can be concluded that the proposed system can be economically competitive for a number of locations. Once again, note that this analysis is based on a specific cost of the CPVT collector, which will be achieved only in case of large commercial productions. Conversely, the present costs of such CPVT prototypes are still very high completely changing the results of the above-discussed economic analysis. 4.5. Sensitivity analysis: energy utilization Finally, this study is concluded by a parametric analysis aiming at evaluating the variation of energetic and economic performances as a function of the ratio of energy and water produced by the system and demanded by the user. As mentioned above, the production of water of the system under investigation is significantly lower than the one demanded by the island. Therefore, it can be concluded that all the produced desalinated water is consumed. The same conclusion is obtained also for the electricity. Similarly, data regarding the main canter of the islands under investigation show that all the space heating and cooling energy produced by the system will be used. Conversely, the amount of DHW produced by HE4 is significantly higher than the possible demand of DHW located nearby the location where the system is supposed to be installed. Therefore, a parametric analysis is presented in Fig. 20, showing the variations of SPB periods and Profit Index as a function of the amount of QHE4 consumed. Such plot shows that similar drop offs of the SPB are detected for all the considered locations. Similarly, Profit Index dramatically decreases in case of scarce DHW demand. In particular, a fraction of about 20% may be estimated for the Island of Pantelleria. In this case the SPB period would increase up 8.50 years (2.26 years in case of public funding) and the Profit Index decreases down to 0.47. However, in case of feed in tariffs the system remains extremely profitable even when DHW demand is scarce. This result is similar to the ones achieved for similar renewable energy systems.

4.4. Sensitivity analysis: location As above discussed, all the previous analysis is performed for the island of Pantelleria. However, several additional islands in the Italian Mediterranean Sea show a similar potential in terms of availability of geothermal and solar energy and face the same issues of scarcity of potable water and energy, especially during the summer season. Therefore, on the basis of the same design parameters shown in Table 1, the simulations have been performed also for two additional representative small Italian volcanic islands, namely Ischia (close to Naples, 40°430 N, 13°540 E) and the Aeolian Islands (North of Sicily, 38°320 N, 14°540 E). The corresponding results are shown in Table 2. Surprisingly, the availability of beam radiation in Pantelleria is lower than the ones in Ischia and in Aeolian Islands. This circumstance determines higher CPVT thermal and electrical productions in these two additional islands (from 10% to 15%). The highest production of cooling energy QC,ACH is achieved in the Aeolian islands, whereas the highest space heating production QHE2 occurs in Pantelleria. No significant differences in terms of desalinated water production are detected among the three considered location. From the economic point view, results are also very similar, though the Aeolian islands show the lowest

Fig. 20. Sensitivity analysis: SPB periods and profit index.

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

18

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx

5. Conclusion A polygeneration system based on the integration of three technologies (Concentrating Photovoltaic-Thermal collectors, Geothermal Wells and Multi-Effect Distillation) was presented and dynamically simulated in TRNSYS. A numerical case study was developed for the volcanic island of Pantelleria in the southern Mediterranean sea. The system is designed on the basis of the thermal demand of the MED. Simulations results show that the energetic performance of the Polygeneration system is excellent especially during the summer, when the availability of beam radiation is high. In fact, during the summer, the system produces large amounts of space cooling energy, Domestic Hot Water, electrical energy and desalinated water only using the combination of solar energy and low-enthalpy geothermal energy. Conversely, during the winter, the performance of the CPVT collectors dramatically decreases determining a corresponding reduction of heat and electricity produced. This results also in a reduction of the space heating energy delivered to the user. As a consequence, in that season the amount of solar energy used for desalination is low and the majority of the heat is supplied to the MED by the geothermal well. Therefore, the system under investigation is especially promising for a number of Mediterranean Volcanic islands where the demands of water and energy dramatically increase during summer as a consequence of the high of number of tourists. The economic analysis - assuming a possible capital cost of CPVT in case of full commercialization and a full utilization of energy and water produced – showed pay back periods extremely low. In case of lower utilization of thermal energy SPB periods may increase up to 10 years. Such result is comparable with that of other renewable systems. The system becomes extremely convenient in case of feed-in tariffs comparable to those presently adopted for PV collectors in many EU areas. Finally, a sensitivity analysis shows a similar thermo-economic performance also for other volcanic islands in the Italian Mediterranean sea.

[15]

References

[32]

[1] IEA – International Energy Agency. World Energy Outlook 2013; 2013. [2] Lund Henrik, Østergaard Poul Alberg, Stadler Ingo. Towards 100% renewable energy systems. Appl Energy 2011;88(2):419–21. [3] Koutroulis Aristeidis G, Tsanis Ioannis K, Daliakopoulos Ioannis N, Jacob Daniela. Impact of climate change on water resources status: a case study for Crete Island, Greece. J Hydrol 2013;479:146–58. [4] Kusangaya Samuel, Warburton Michele L, van Garderen Emma Archer, Jewitt Graham PW. Impacts of climate change on water resources in southern Africa: a review. Phys Chem Earth Parts A/B/C; 2013 [in press]. http://dx.doi.org/ 10.1016/j.pce.2013.09.014. [5] Manios Thrassyvoulos, Tsanis Ioannis K. Evaluating water resources availability and wastewater reuse importance in the water resources management of small Mediterranean municipal districts. Resour Conserv Recycl 2006;47(3):245–59. [6] Dentice d’Accadia M, Calise F, Vanoli L. Thermoeconomic optimization of solar heating and cooling systems. Energy Convers Manage 2010;52(2):1562–73. [7] Zondag HA. Flat-plate PV-thermal collectors and systems: a review. Renew Sustain Energy Rev 2008;12(4):891–5. [8] Chow TT. A review on photovoltaic/thermal hybrid solar technology. Appl Energy 2010;87(2):365–79. [9] Chow TT, Tiwari GN, Menezo C. Hybrid solar: a review on photovoltaic and thermal power integration. Int J Photoenergy 2012;2012:1–17. [10] Calise F, Vanoli L. Parabolic trough photovoltaic/thermal collectors: design and simulation model. Energies 2012;5(10):4186–208. [11] Buonomano A, Calise F, Palombo A. Solar heating and cooling systems by CPVT and ET solar collectors: a novel transient simulation model. Appl Energy 2013;103:588–606. [12] Mittelman G, Kribus A, Dayan A. Solar cooling with concentrating photovoltaic/thermal (CPVT) systems. Energy Convers Manage 2007;48: 2481–90. [13] Zhang Xingxing, Zhao Xudong, Smith Stefan, Xu Jihuan, Yu Xiaotong. Review of R&D progress and practical application of the solar photovoltaic/thermal (PV/ T) technologies. Renew Sustain Energy Rev 2011;16(1):599–617. [14] Nishioka K, Takamoto T, Agui T, Kaneiwa M, Uraoka Y, Fuyuki T. Annual output estimation of concentrator photovoltaic systems using high-efficiency InGaP/

[16]

[17] [18]

[19] [20]

[21] [22]

[23]

[24]

[25] [26]

[27] [28]

[29]

[30] [31]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

InGaAs/Ge triple-junction solar cells based on experimental solar cell’s characteristics and field-test meteorological data. Sol Energy Mater Sol Cells 2006;90:57–67. Abduljawad M, Ezzeghni U. Optimization of Tajoura MSF desalination plant. Desalination 2010;254:23–8. Tian L, Tang Y, Wang Y. Economic evaluation of seawater desalination for a nuclear heating reactor with multi-effect distillation. Desalination 2005;180:53–61. Qiblawey HM, Banat F. Solar thermal desalination technologies. Desalination 2008;220:633–44. Malik M, Tiwari G, Kumar A, Sodha M. Solar distillation: a practical study of a wide range of stills and their optimum design construction and performance. University of Michigan (USA): Pergamon Press; 1996. Fath H, Ghazy A. Solar desalination using humidification–dehumidification technology. Desalination 2002;142:119–33. Szacsvay T, Hofer-Noser P, Posnansky M. Technical and economic aspects of small-scale solar pond powered seawater desalination systems. Desalination 1999;122:185–93. Williams PM, Ahmad M, Connolly BS. Freeze desalination: an assessment of an ice maker machine for desalting brines. Desalination 2013;308:219–24. Koschikowski J, Wieghaus M, Rommel M. Membrane distillation for solar desalination. In: Cipollina A, Micale G, Rizzuti L, editors. Seawater desalination. Conventional and renewable energy processes. Berlin & Heidelberg: SpringerVerlag; 2009. Schwantes R, Cipollina A, Gross F, Koschikowski J, Pfeifle D, Rolletschek M, et al. Membrane distillation: solar and waste heat driven demonstration plants for desalination. Desalination 2013;323:93–106. Fernández-López C, Viedma A, Herrero R, Kaiser AS. Seawater integrated desalination plant without brine discharge and powered by renewable energy systems. Desalination 2009;235:179–98. Goosen Mattheus, Mahmoudi Hacene, Ghaffour Noreddine. Water desalination using geothermal energy. Energies 2010;3:1423–42. Gude Veera Gnaneswar, Nirmalakhandan Nagamany, Deng Shuguang. Renewable and sustainable approaches for desalination. Renew Sustain Energy Rev 2010;14:2641–54. Mathioulakis E, Belessiotis V, Delyannis E. Desalination by using alternative energy: review and state-of-the-art. Desalination 2007;203(1–3):346–65. García-Rodríguez L, Gómez-Camacho C. 2001. Desalination, exergy analysis of the SOL-14 plant. Plataforma Solar de Almería, Spain, vol. 137; 2001. p. 251–8. Alarcón-Padilla DC, García-Rodríguez L, Blanco-Gálvez J. Design recommendations for a multi-effect distillation plant connected to a doubleeffect absorption heat pump: a solar desalination case study. Desalination 2010;262:11–4. Manenti Flavio, Masi Maurizio, Santucci Giorgio. Start-up operations of MED desalination plants. Desalination 2013;329:57–61. Manenti Flavio, Masi Maurizio, Santucci Giorgio, Manenti Giovanni. Parametric simulation and economic assessment of a heat integrated geothermal desalination plant. Desalination 2013;317:193–205. Bourouni Karim, Deronzier Jean Claude, Tadrist Lounes. Experimentation and modelling of an innovative geothermal desalination unit. Desalination 1999;125(1–3):147–53. Mohamed AMI, El-Minshawy NAS. Humidification dehumidification desalination system driven by geothermal energy. Desalination 2009;249(2): 602–8. Sarbatly Rosalam, Chiam Chel-Ken. Evaluation of geothermal energy in desalination by vacuum membrane distillation. Appl Energy 2013;112: 737–46. Kyriakarakos George, Piromalis Dimitrios D, Dounis Anastasios I, Arvanitis Konstantinos G, Papadakis George. Intelligent demand side energy management system for autonomous polygeneration microgrids. Appl Energy 2013;103:39–51. Piacentino Antonio, Barbaro Chiara, Cardona Fabio, Gallea Roberto, Cardona Ennio. A comprehensive tool for efficient design and operation of polygeneration-based energy lgrids serving a cluster of buildings. Part I: description of the method. Appl Energy 2013;111:1204–21. Piacentino Antonio, Barbaro Chiara. A comprehensive tool for efficient design and operation of polygeneration-based energy lgrids serving a cluster of buildings. Part II: analysis of the applicative potential. Appl Energy 2013;111:1222–38. Song Han, Starfelt Fredrik, Daianova Lilia, Yan Jinyue. Influence of drying process on the biomass-based polygeneration system of bioethanol, power and heat. Appl Energy 2012;90(1):32–7. Zhou Wei, Yang Hongxing, Rissanen Markku, Nygren Bertil, Yan Jinyue. Decrease of energy demand for bioethanol-based polygeneration system through case study. Appl Energy 2012;95:305–11. Starfelt Fredrik, Daianova Lilia, Yan Jinyue, Thorin Eva, Dotzauer Erik. The impact of lignocellulosic ethanol yields in polygeneration with district heating – a case study. Appl Energy 2012;92:791–9. Le Truong Nguyen, Gustavsson Leif. Integrated biomass-based production of district heat, electricity, motor fuels and pellets of different scales. Appl Energy 2013;104:623–32. Kyriakarakos George, Dounis Anastasios I, Rozakis Stelios, Arvanitis Konstantinos G, Papadakis George. Polygeneration microgrids: a viable solution in remote areas for supplying power, potable water and hydrogen as transportation fuel. Appl Energy 2011;88(12):4517–26.

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064

F. Calise et al. / Applied Energy xxx (2014) xxx–xxx [43] Rubio-Maya Carlos, Uche-Marcuello Javier, Martínez-Gracia Amaya, BayodRújula Angel A. Design optimization of a polygeneration plant fuelled by natural gas and renewable energy sources. Appl Energy 2011;88(2):449–57. [44] Buonomano Annamaria, Calise Francesco, d’Accadia Massimo Dentice, Vanoli Laura. A novel solar trigeneration system based on concentrating photovoltaic/ thermal collectors. Part 1: design and simulation model. Energy 2013;61:59–71. [45] Narvaez A, Chadwick D, Kershenbaum L. Small-medium scale polygeneration systems: methanol and power production. Appl Energy 2014;113:1109–17. [46] Buonomano Annamaria, Calise Francesco, Ferruzzi Gabriele. Thermoeconomic analysis of storage systems for solar heating and cooling systems: a comparison between variable-volume and fixed-volume tanks. Energy 2013;59:600–16. [47] Klein SA et al. TRNSYS. A transient system simulation program. Madison: Solar Energy Laboratory, University of Wisconsin; 2006. [48] Calise F, Dentice d’Accadia M, Palombo A. Transient analysis and energy optimization of solar heating and cooling systems in various configurations. Sol Energy 2010;84(3):432–49. [49] Calise F. Design of a hybrid polygeneration system with solar collectors and a solid oxide fuel cell: dynamic simulation and economic assessment. Int J Hydrogen Energy 2011;36(10):6128–50. [50] Calise F. Thermoeconomic analysis and optimization of high efficiency solar heating and cooling systems for different Italian school buildings and climates. Energy Build 2010;42:992–1003. [51] Dentice d’Accadia M, Calise F, Vanoli R. Dynamic simulation and parametric optimization of a solar-assisted heating and cooling system. Int J Ambient Energy 2010;31(4):173–94. [52] Calise Francesco, Palombo Adolfo, Vanoli Laura. A finite-volume model of a parabolic trough photovoltaic/thermal collector: energetic and exergetic analyses. Energy 2012;46(1):283–94. [53] Calise Francesco, d Accadia Massimo Dentice, Vanoli Laura. Design and dynamic simulation of a novel solar trigeneration system based on hybrid photovoltaic/ thermal collectors (PVT). Energy Convers Manage 2012;60:214–25. [54] Calise F, Dentice d’Accadia M, Piacentino A. A novel solar trigeneration system integrating PVT (photovoltaic/thermal collectors) and SW (seawater) desalination: dynamic simulation and economic assessment. Energy 2014. [55] Cardona E, Piacentino A, Marchese F. Performance evaluation of CHP hybrid seawater desalination plants. Desalination 2007;205:1–14. [56] El-Nashar Ali M, Samad M. The solar desalination plant in Abu Dhabi: 13 years of performance and operation history. Renew Energy 1998;14:263–74.

19

[57] Milow B, Zarza E. Advanced MED solar desalination plants. Configurations, costs, future—seven years of experience at the Plataforma Solar de Almeria (Spain). Desalination 1997;108:51–8. [58] Abu-Jabal MS, Kamiya I, Narasaki Y. Proving test for a solar-powered desalination system in Gaza-Palestine. Desalination 2001;137:1–6. [59] Gude Veera Gnaneswar, Nirmalakhandan Nagamany, Deng Shuguang, Maganti Anand. Low temperature desalination using solar collectors augmented by thermal energy storage. Appl Energy 2012;91(1):466–74. [60] Kakac, Liu. Heat exchanger selection, rating and thermal design. Florida (USA): Boca Raton CRC Press; 1998. [61] Serth RW. Process heat transfer e principles and applications. Burlington (USA): Oxford C. Academic Press – Imprint of Elsevier; 2007. [62] El-Dessouky HT, Ettouney HM. Fundamentals of salt water desalination. Amsterdam (The Netherlands): Elsevier; 2002. [63] Kemp IC. Pinch analysis and process integration. Oxford (UK): ButterworthImprint of Elsevier; 2007. [64] El-Nashar AM. The economic feasibility of small solar MED seawater desalination plants for remote arid areas. Desalination 2001;134:173–86. [65] Calise Francesco, d’Accadia Massimo Dentice, Palombo Adolfo, Vanoli Laura. Dynamic simulation of a novel high-temperature solar trigeneration system based on concentrating photovoltaic/thermal collectors. Energy 2013;61:72–86. [66] Rezaie Behnaz, Rosen Marc A. District heating and cooling: Review of technology and potential enhancements. Appl Energy 2012;93:2–10. [67] Calise Francesco. High temperature solar heating and cooling systems for different Mediterranean climates: dynamic simulation and economic assessment. Appl Therm Eng 2012;32:108–24. [68] Energy and environment report 2008, 2008, European Environment Agency. [accessed 02.14]. [69] Sanseverino Eleonora Riva, Sanseverino Raffaella Riva, Favuzza Salvatore, Vaccaro Valentina. Near zero energy islands in the Mediterranean: supporting policies and local obstacles. Energy Policy 2014;66:592–602. [70] Carlino S, Somma R, Troiano A, Di Giuseppe MG, Troise C, De Natale G. The geothermal system of Ischia Island (southern Italy): critical review and sustainability analysis of geothermal resource for electricity generation. Renew Energy 2014;62:177–96. [71] Dickinson Ryan M, Cruickshank Cynthia A, Harrison Stephen J. Charge and discharge strategies for a multi-tank thermal energy storage. Appl Energy 2013;109:366–73.

Please cite this article in press as: Calise F et al. A novel renewable polygeneration system for a small Mediterranean volcanic island for the combined production of energy and water: Dynamic simulation and economic assessment. Appl Energy (2014), http://dx.doi.org/10.1016/j.apenergy.2014.03.064