Dynamic Simulation and Thermoeconomic Analysis of a Trigeneration System in a Hospital Application

Hospitals are very attractive for Combined Heat and Power (CHP) applications, due to their high and continuous demand for electric and thermal energy. However, both design and control strategies of CHP systems are usually based on an empiric and very simplified approach, and this may lead to non-optimal solutions. The paper presents a novel approach based on the dynamic simulation of a trigeneration system to be installed in a hospital located in Puglia (South Italy), with around 600 beds, aiming to investigate the energy and economic performance of the system, for a given control strategy (electric-load tracking). The system includes a natural gas fired reciprocating engine (with a rated power of 2.0 MW), a single-stage LiBr-H2O absorption chiller (with a cooling capacity of around 770 kW), auxiliary gas-fired boilers and steam generators, electric chillers, cooling towers, heat exchangers, storage tanks and several additional components (pipes, valves, etc.). Suitable control strategies, including proportional–integral–derivative (PID) and ON/OFF controllers, were implemented to optimize the trigeneration performance. The model includes a detailed simulation of the main components of the system and a specific routine for evaluating the heating and cooling demand of the building, based on a 3-D model of the building envelope. All component models were validated against experimental data provided by the manufacturers. Energy and economic models were also included in the simulation tool, to calculate the thermoeconomic performance of the system. The results show an excellent economic performance of the trigeneration system, with a payback period equal to 1.5 years and a profitability index (ratio of the Net Present Value to the capital cost) equal to 3.88, also due to the significant contribution of the subsidies provided by the current Italian regulation for CHP systems (energy savings certificates).


Introduction
Cogeneration (or CHP, Combined Heat and Power) represents a mature and well-known technology, able to ensure remarkable energy and economic savings, due to the combined production of thermal end electric energy, from a single primary energy input [1,2]. Trigeneration systems are CHP units whose thermal waste energy is also used to drive a thermally-driven chiller. Such systems are often referred to as Combined Cooling, Heat and Power (CCHP) systems [3]. CCHP systems can be also used to produce other products (e.g., hydrogen, alcohols, glycerine), configuring a polygeneration system [4].
CHP and CCHP are based on the recovery of the exhaust heat rejected by an engine, also called the prime mover (PM) [5]. Such thermal energy may be used for heating, process, production of sanitary hot water, etc., and/or to drive a thermally-driven chiller to produce cooling energy [1,6]. The main engine is off and/or its capacity is not sufficient to this scope. The demand for thermal energy includes the production of sanitary hot water, steam, space heating and cooling. The electric energy is used for lights, machinery, medical devices, pumps, fans. An electrical load tracking mode was selected by the company who manage the system, and the maximum electric power of the CHP system is lower than the maximum power demand, in order to reduce the capital cost of the system and the running time in part load conditions. Therefore, when the power demand is lower than the cogenerator maximum capacity, such demand is fully supplied by the CHP unit; in other cases, the cogeneration unit operates at full load and a certain amount of power must be imported from the public grid. As a consequence, the availability of waste heat for heating and/or cooling purposes varies, according to the operating conditions of the engine. At any rate, the heat demand is usually much higher than the capacity of the CHP system, so that the auxiliary heaters are always active, and the same happens for the cooling demand. The heat recovery system includes various loops, as described in the following. In the Exhaust Gases Loop (EGL), the Exhaust Gases (EG) leaving the engine (at a temperature of about 411 °C) are used to produce steam in the Steam-Exhaust Gas heat exchanger (S-EG-HE), evolving within the Steam Loop (SL); afterwards, in the EG-JW-HE, the exhaust gases are used to enhance the temperature of the engine jacket water (JW), flowing within the JW Loop (JWL), and finally leave the heat exchange section at a temperature of at least 120 °C (to avoid acid condensation phenomena inside the heat exchangers).
As described in Figure 2, if the vapor produced in S-EG-HE (Qsteam,S-EG-HE) is lower than the hospital demand (Qsteam), the remaining amount of steam is produced by the auxiliary steam generator (SG). In particular, to this end, a suitable amount of water is sent from TKs to SG, by means of the pump PS2. The steam flows produced by S-EG-HE and/or SG are mixed in the steam collector (SColl), and finally are made available to the end users.  As described in Figure 2, if the vapor produced in S-EG-HE (Q steam,S-EG-HE ) is lower than the hospital demand (Q steam ), the remaining amount of steam is produced by the auxiliary steam generator (SG). In particular, to this end, a suitable amount of water is sent from TKs to SG, by means of the pump PS2. The steam flows produced by S-EG-HE and/or SG are mixed in the steam collector (SColl), and finally are made available to the end users.    For the JW Loop, two different operation strategies are adopted, for heating and cooling periods, respectively. During the cooling period, the jacket water is primarily supplied to the absorption chiller (ACH), as shown in Figure 3; subsequently, the JW leaving the ACH is supplied to the JW-HW-HE ( Figure 1) to provide heat to the heating water Loop. As shown in Figure 4, if the jacket water temperature decreases below 80 • C, the diverter D4 addresses the flow to the mixer M4, keeping the JW temperature within the range suggested by the engine manufacturer [19]. Conversely, if temperature of the jacket water rises over 80 • C, the dissipator unit is activated, preventing engine overheating. Clearly, if the cooling energy demand is null, the JW directly heats the heating Loop through the HE-H. Finally, JW cools down the engine oil in the HE-O1 exchanger.     The engine lubricating oil, flowing in the Oil Loop (OL), is cooled in HE-O1, supplying heat to jacket water, and in and HE-O2, where low temperature heat is dissipated in the heat exchanger DISS.
The Engine Water Cooling Loop (EWCL) is used to cool the second stage of intercooling (IC2) and the engine oil (HE-O2)-since the temperature within this loop is too low, the thermal energy of the loop is entirely dissipated in the dissipator (DISS).
The heating water Loop (HWL) consists of the water heated in JW-HW-HE, which supplies thermal energy for building space heating and domestic hot water (DHW). In particular, in JW-HW-HE the heat exchange between JW and heating water (HW) is performed, increasing the temperature of HW. Subsequently, HW is supplied to the heating water collector (Coll 1). Note that the setpoint temperature of HWL is 85 • C. If the temperature of HWL is below 85 • C, the auxiliary condensation boiler (CB) is activated, raising such temperature to 85 • C ( Figure 5). The HWL is managed by pump P2, which is turned on only when there is a demand for space heating and/or the top temperature of the domestic hot water tank (T top,TKDHW ) decreases below 45 • C ( Figure 6). In particular, when T top,TKDHW decreases below 45 • C, the pump P7 is also switched on, supplying thermal energy to TKDWH. The pump P7 is switched off when T top,TKDHW reaches the value of 60 • C.
The Domestic Hot Water Loop (DHWL) refers to the sanitary hot water required by the hospital. When the DHW demand is not null, the pump P8 is switched on. Thus, a proper amount of water flows through the heat exchanger towards the tank TKDHW, heating the water from 15 • C (tap water) to 45 • C.
The Building space Heating loop (BHL) refers to the hot water supplied to the fain-coils inside the hospital. In the heating period, the pump P5 is switched on, keeping the indoor temperature equal to the its setpoint value ( Table 2).
The Building space Cooling Loop (BCL) is used to provide space cooling by means of Chilled Water (CW), supplied to fan coils located in the building. The cooling demand is matched by the ACH, as far as possible; when the ACH capacity is not sufficient, the auxiliary electric chillers (CH) are activated.
The Cooling Tower Water Loop (CTWL) is used to cool all the chillers included in the plant (ACH and CH); therefore, when ACH and/or CH are running, the pump P4 must be activated.
As mentioned before, an electric load tracking operating strategy was adopted for the system under evaluation. The engine can work with a partial load ratio (R el , defined as the ratio of the power demand on the maximum rated power) ranging between 50% and 100%; when R el falls below 50%, the engine must be switched off.

Simulation Model
The system presented was modelled in TRNSYS 17 environment. TRNSYS is widely employed in the academic community, because of its high accuracy and reliability; moreover, a large library of user-developed models and experimentally validated built-in components [20] is included in this software. The model includes several components (engine, heat exchangers, tanks, pumps, etc). Some of these models are based on thermodynamic models and they are mainly taken from TRNSYS library, being previously validated against experimental data and successfully used in several works. The Type 56 (Building) is considered a reference in the field of building dynamic simulation. The 3D geometric model of the hospital was built with the Google SketchUp TRNSYS 3d plug-in; such geometric model was then imported in the Type 56.
The main TRNSYS libraries used in this work are described below.
• Type 56: this library dynamically models the energy performance of the building considering its 3D geometry, thermophysical proprieties of the envelope, dynamic weather conditions (i.e., solar radiation, ambient temperature, wind velocity, relative humidity, etc.); further details are available in Ref. [21].

•
Type 907: this library describes the performance of internal combustion engines, considering oil cooling water, after-cooling water, jacket water, and using inlet air temperature as a parameter. The mass and thermal balance performed by this library are based on manufactured data [18].

Simulation Model
The system presented was modelled in TRNSYS 17 environment. TRNSYS is widely employed in the academic community, because of its high accuracy and reliability; moreover, a large library of user-developed models and experimentally validated built-in components [20] is included in this software. The model includes several components (engine, heat exchangers, tanks, pumps, etc). Some of these models are based on thermodynamic models and they are mainly taken from TRNSYS library, being previously validated against experimental data and successfully used in several works. The Type 56 (Building) is considered a reference in the field of building dynamic simulation. The 3D geometric model of the hospital was built with the Google SketchUp TRNSYS 3d plug-in; such geometric model was then imported in the Type 56.
The main TRNSYS libraries used in this work are described below.
• Type 56: this library dynamically models the energy performance of the building considering its 3D geometry, thermophysical proprieties of the envelope, dynamic weather conditions (i.e., solar radiation, ambient temperature, wind velocity, relative humidity, etc.); further details are available in Ref. [21].

•
Type 907: this library describes the performance of internal combustion engines, considering oil cooling water, after-cooling water, jacket water, and using inlet air temperature as a parameter. The mass and thermal balance performed by this library are based on manufactured data [18]. For the sake of brevity, the equations of this model not are described here, but a detailed description is available in Ref. [1].

•
Type 107: this library models the performance of a single-effect LiBr-H 2 O absorption chiller. The library adopts a normalized catalogue data look-up principle [22]. Thus, the performance of ACH is evaluated by using an operating map provided by manufacturers [23]. Further details are reported in Ref. [1].

Energy Analysis
A comparison between the Proposed System (PS) described above and a conventional, Reference System (RS) was carried out. The energy comparison was performed by evaluating the Primary Energy consumptions (PE) of both RS and PS, and the corresponding Primary Energy Saving (PES) achieved by the PS. In particular, PE and PES were calculated as follows: Here, E el,load is the overall electric energy demand of the hospital, E th,cool and E th,heat are the thermal energy demand for space cooling and heating, respectively, E th,DHW is the thermal energy demand for sanitary hot water production, E th,steam is the thermal energy needed for steam production, E el,fromGRID is the electric energy withdrawn from the grid, E th,CB and E th,SG are the thermal energy provided by auxiliary boilers and steam generators, respectively; the remaining terms in Equation (1) are explained in Table 1.

Economic Analysis
The yearly operating cost (C) of RS and PS are calculated as follows: Table 1 explains the main terms of these equations. The subsidies available in Italy for cogeneration (Energy Saving Certificates, ESC) were considered; they are proportional to the primary energy saving acknowledged as "High-Efficiency cogeneration" units, and can be calculated with an algorithm described in Ref. [1]. The economic performance of the proposed system was evaluated by means of the following economic parameters: (i) yearly economic savings (∆C); (ii) Simple payback period (SPB) of the investment; (iii) net present value (NPV); and (iv) profitability index (PI): Energies 2020, 13, 3558 8 of 16 Here, C inv is the capital cost of the cogeneration plant, equal to 2.0 M€, and AF is the annuity factor, assumed equal to eight years (based on a discount rate of 5% and a time horizon of 10 years- Table 1).

Hospital Energy Loads
The system described above is to be installed in the "F. Miulli" hospital, located in Acquaviva delle Fonti, near Bari, in South Italy (Figures 7 and 8). The facility has 603 beds and a gross volume of 370 × 10 3 m 3 .
To limit the computational cost of the simulations, while keeping the accuracy high, the complex of buildings was thermally modelled by defining four thermal zones (Table 2  The occupancy schedule of such thermal zones is reported in Table 2. Here, data about the main geometric characteristics, ventilation, setpoint temperatures and heating and cooling periods for each thermal zone are reported, too. The elements of the building envelope are described in Table 3. Note that, as it is well-known, the thickness and U-value of each element affect the energy performance of the building in terms of space heating and cooling energy consumption. The hospital is located in the Italian weather zone D (1.610 heating degree-day, Table 2). It is useful to remember that the heating degree-days of a site are defined as the sum of the difference (considered only if positive) between the indoor setpoint temperature and the mean daily outdoor temperature of that site. In the heating season, such parameter is commonly used to estimate the energy consumption for space heating of buildings.
According to the Italian legislation about healthcare facilities, no restrictions are present concerning the duration of the heating periods.
Presently, the energy needs of the facility are matched as follows: electric air-to-air chillers, natural gas-fueled steam generators, for steam, and condensing boilers for sanitary hot water preparation and space heating; electric energy and natural gas; this configuration was adopted as the Reference System for the energy and economic analyses of the new CCHP system presented in the next section.
The hospital consumes about 15.89 GWh/y of electric energy and 1.76 × 10 6 Sm 3 of natural gas. Figure 9 shows the current electric energy and natural gas consumptions on a monthly basis. The occupancy schedule of such thermal zones is reported in Table 2. Here, data about the main geometric characteristics, ventilation, setpoint temperatures and heating and cooling periods for each thermal zone are reported, too. The elements of the building envelope are described in Table 3. Note that, as it is well-known, the thickness and U-value of each element affect the energy performance of the building in terms of space heating and cooling energy consumption. The hospital is located in the Italian weather zone D (1.610 heating degree-day, Table 2). It is useful to remember that the heating degree-days of a site are defined as the sum of the difference (considered only if positive) between the indoor setpoint temperature and the mean daily outdoor temperature of that site. In the heating season, such parameter is commonly used to estimate the energy consumption for space heating of buildings.
According to the Italian legislation about healthcare facilities, no restrictions are present concerning the duration of the heating periods.
Presently, the energy needs of the facility are matched as follows: electric air-to-air chillers, natural gas-fueled steam generators, for steam, and condensing boilers for sanitary hot water preparation and space heating; electric energy and natural gas; this configuration was adopted as the Reference System for the energy and economic analyses of the new CCHP system presented in the next section.
The hospital consumes about 15.89 GWh/y of electric energy and 1.76 × 10 6 Sm 3 of natural gas. Figure 9 shows the current electric energy and natural gas consumptions on a monthly basis.
In particular, three 3 MW TRANE CVGF1000 electric chillers combined with four 3.5 MW cooling towers are included for the production of chilled water; one chiller is used as a backup system. The power absorbed by each chiller is equal to equal to 504 kW. The total electric consumption of chillers and cooling towers is  Presently, the energy needs of the facility are matched as follows: electric air-to-air chillers, natural gas-fueled steam generators, for steam, and condensing boilers for sanitary hot water preparation and space heating; electric energy and natural gas; this configuration was adopted as the Reference System for the energy and economic analyses of the new CCHP system presented in the next section.
The hospital consumes about 15.89 GWh/y of electric energy and 1.76 × 10 6 Sm 3 of natural gas. Figure 9 shows the current electric energy and natural gas consumptions on a monthly basis. The layout of the cogeneration system to be installed was already presented in Figure 1. The CHP unit is an internal combustion engine model ABB Ecomax 20 , with a rated power of 2.00 MW condensing boilers, are employed as auxiliary systems, see Figure. 1.
In conclusion, the case study presented above was solved in the TRNSYS environment, selecting an equation-resolution time step of 0.05 h, as shown in Table 2. 15th November-31stMarch 1st May-30th September Occupancy schedule 00:00-24:00 00:00-24:00 08:00-18:00 00:00-24:00 Simulation Time step (h) 0.05 Figure 9. Electric energy and Natural gas monthly demand. Figure 9. Electric energy and Natural gas monthly demand.  The layout of the cogeneration system to be installed was already presented in Figure 1. The CHP unit is an internal combustion engine model ABB Ecomax 20, with a rated power of 2.00 MW and a thermal capacity of 1.95 MW [18] (Table 4); the nominal electric efficiency is 44.1%, the thermal one 43.0%; therefore, the global efficiency is equal to 87.1%. According to the indications provided by the ESCO, who designed the plant, an electric load tracking operating strategy was adopted. Table 4 also displays the characteristics of other devices included in the layout under analysis. Note that the existing devices, such as steam generators and condensing boilers, are employed as auxiliary systems, see Figure 1.
In conclusion, the case study presented above was solved in the TRNSYS environment, selecting an equation-resolution time step of 0.05 h, as shown in Table 2.

Results
In this section, the calibration of the building model is presented as well as the results of the simulations carried out are discussed. The calibration is performed using the data regarding the energy bills. In particular, for each month of the year, the comparison between the real data of thermal energy demand for space heating and domestic hot water and the simulated value by the implementation of the building model is presented (Figure 10). The same comparison is presented also for the total electric demand, including also the electric consumption to supply the electric chillers for the hospital space cooling. Figure 10 clearly shows a good agreement between the numerical results and real data. Therefore, the building model resulted to be calibrated. chillers for the hospital space cooling. Figure 10 clearly shows a good agreement between the numerical results and real data. Therefore, the building model resulted to be calibrated.  Figure 11 displays the dynamic performance of the CCHP under evaluation. The power (Pel,CHP) varies over the hours, due to the control strategy assumed (electric load tracking). In particular, in the case shown in the figure, Pel,CHP is constantly equal to the power demand of the hospital (Pel,LOAD). The latter, in fact, is never higher than the maximum power of the CCHP system. Moreover, Figure  11 displays the thermal control strategy adopted, aiming at keeping the jacket water temperature entering the engine (TJW,inCHP) equal to about 80 °C.  Figure 11 displays the dynamic performance of the CCHP under evaluation. The power (P el,CHP ) varies over the hours, due to the control strategy assumed (electric load tracking). In particular, in the case shown in the figure, P el,CHP is constantly equal to the power demand of the hospital (P el,LOAD ). The latter, in fact, is never higher than the maximum power of the CCHP system. Moreover, Figure 11 displays the thermal control strategy adopted, aiming at keeping the jacket water temperature entering the engine (T JW,inCHP ) equal to about 80 • C.

Daily Results
Energies 2020, 13, x FOR PEER REVIEW 13 of 18 Figure 11. Dynamic performance of the cogeneration system-daily results. Figure 12 displays the weekly energy performance of the proposed layout. According to the operating strategy adopted for the system, it is able to constantly match the electric energy demand of the hospital (Figure 11 and Figure 12-left side). Note that some weeks are characterized by zero or limited energy production, due to scheduled maintenance interventions. Clearly, when the engine is switched off, the public grid is used as a backup system, meeting the electric energy demand of the hospital. The electric energy delivered to the auxiliary systems is very low, achieving the maximum value of 10 MWh/week ( Figure 12-left side). The thermal energy recovered by the engine (Eth,CHP) is equal to about 281-310 MWhth/week ( Figure 12-right side). In particular, the medium temperature thermal energy (Eth,CHP,MT) ranges between 281-310 MWhth/week, whereas the high temperature one (Eth,CHP,HT) ranges between 109 and 129 MWhth/week (Figure 12-right side). Figure 13 displays the performance of the Steam Loop (SL), and, in particular, the thermal energy required for steam production (Eth,Steam), and the ratio of the thermal energy provided by the steam generator (SG) and by the CHP system to the overall request, Eth,Steam (RSG and RCGP,HT, respectively). From Week 28 to week 35 (summer period), the thermal energy provided by SG for producing steam is about null, i.e., RSG is equal to zero. This trend is due to the reduction of the demand of steam during these weeks. During the remaining weeks, the CHP system is able to meet a limited share of Eth,Steam: RCGP,HT ranges between 42% and 57%, except the weeks in which the programmed maintenance operations occur (weeks 17 and 44). Figure 14 displays the weekly energy balance regarding the cooling service; in particular, the following quantities are shown: cooling energy demand (Eth,cool), ratio of the cooling energy provided by ACH on Eth,cool (RACH), and ratio of cooling energy provided by the electric-driven auxiliary chiller on Eth,cool (RCH). When the cooling demand is high (weeks from 25 to 36), the ACH is able to match the major share of cooling energy demand (RCH ranging from 56% to 76%).  Figure 12 displays the weekly energy performance of the proposed layout. According to the operating strategy adopted for the system, it is able to constantly match the electric energy demand of the hospital (Figures 11 and 12-left side). Note that some weeks are characterized by zero or limited energy production, due to scheduled maintenance interventions. Clearly, when the engine is switched off, the public grid is used as a backup system, meeting the electric energy demand of the hospital. The electric energy delivered to the auxiliary systems is very low, achieving the maximum value of 10 MWh/week ( Figure 12-left side). The thermal energy recovered by the engine (E th,CHP ) is equal to about 281-310 MWh th /week ( Figure 12-right side). In particular, the medium temperature thermal energy (E th,CHP,MT ) ranges between 281-310 MWh th /week, whereas the high temperature one (E th,CHP,HT ) ranges between 109 and 129 MWh th /week (Figure 12-right side).  , and, in particular, the thermal energy required for steam production (E th,Steam ), and the ratio of the thermal energy provided by the steam generator (SG) and by the CHP system to the overall request, E th,Steam (R SG and R CGP,HT , respectively). From Week 28 to week 35 (summer period), the thermal energy provided by SG for producing steam is about null, i.e., R SG is equal to zero. This trend is due to the reduction of the demand of steam during these weeks. During the remaining weeks, the CHP system is able to meet a limited share of E th,Steam : R CGP,HT ranges between 42% and 57%, except the weeks in which the programmed maintenance operations occur (weeks 17 and 44).    Table 5 summarizes the yearly performance of the proposed layout. CHP produces 13.74 GWh/year of electric energy, achieving an electric efficiency of 42%. The proposed system exhibits a limited energy saving (PES equal to 9%). This is mainly due to the fact that the energy saving was calculated with respect to the overall primary energy consumption of the reference scenario, whereas the Proposed System only covers a part of such consumption. in particular, the share of thermal energy produced by the CHP on the overall request is only 41% ( Table 6). Should the energy saving be calculated for the cogeneration system only, it would be significantly higher.

Yearly Results
In any case, from the economical point of view, the Proposed System achieves significant results, with an SPB of 1.5 years and an NPV of 7.10 M€ ( Table 5). The economic results are mainly due to  Figure 14 displays the weekly energy balance regarding the cooling service; in particular, the following quantities are shown: cooling energy demand (E th,cool ), ratio of the cooling energy provided by ACH on E th,cool (R ACH ), and ratio of cooling energy provided by the electric-driven auxiliary chiller on E th,cool (R CH ). When the cooling demand is high (weeks from 25 to 36), the ACH is able to match the major share of cooling energy demand (R CH ranging from 56% to 76%).    Table 5 summarizes the yearly performance of the proposed layout. CHP produces 13.74 GWh/year of electric energy, achieving an electric efficiency of 42%. The proposed system exhibits a limited energy saving (PES equal to 9%). This is mainly due to the fact that the energy saving was calculated with respect to the overall primary energy consumption of the reference scenario, whereas the Proposed System only covers a part of such consumption. in particular, the share of thermal energy produced by the CHP on the overall request is only 41% (Table 6). Should the energy saving be calculated for the cogeneration system only, it would be significantly higher.

Yearly Results
In any case, from the economical point of view, the Proposed System achieves significant results, with an SPB of 1.5 years and an NPV of 7.10 M€ ( Table 5). The economic results are mainly due to  Table 5 summarizes the yearly performance of the proposed layout. CHP produces 13.74 GWh/year of electric energy, achieving an electric efficiency of 42%. The proposed system exhibits a limited energy saving (PES equal to 9%). This is mainly due to the fact that the energy saving was calculated with respect to the overall primary energy consumption of the reference scenario, whereas the Proposed System only covers a part of such consumption. in particular, the share of thermal energy produced by the CHP on the overall request is only 41% (Table 6). Should the energy saving be calculated for the cogeneration system only, it would be significantly higher.

Yearly Results
In any case, from the economical point of view, the Proposed System achieves significant results, with an SPB of 1.5 years and an NPV of 7.10 M€ (Table 5). The economic results are mainly due to two elements: first, the proposed CHP is able to achieve a remarkable amount of energy savings certificates (ESC), whose market value amounts to 0.207 M€/year; second, energy prices are considerably high in Italy, therefore self-production causes significant economic savings. Therefore, even if the primary energy saving is moderate (about 9%), the economic profitability is excellent.

Conclusions
The paper presents the energy and economic analysis of a trigeneration system, installed in a hospital located in South Italy. In this scope, a detailed dynamic simulation model was developed in TRNSYS, in order to analyse the behaviour of the proposed system, whereas real data were used about the energy consumption, people scheduling and thermophysical characteristics of the hospital.
The main results are summarized below: • the trigeneration system produces about 13.74 GWh/year of electric energy with an electric efficiency of 42%; in particular, 92% of the electric energy demand of the hospital is met by the system; • a limited share of the overall thermal energy demand of the hospital is matched by the cogeneration unit (41%); this is due to the fact that the size of the system was mainly selected on the basis of the power demand of the hospital (much lower than the thermal one), to avoid: (i) the production of excess electric energy (in case of full-load operation or thermal load tracking); (ii) the need for continuous part-load operation, in case of electric load tracking (this latter represents the option actually selected for the system under consideration); • in spite of a limited primary energy savings index (about 9%), mainly due to the limited share of thermal energy covered by the cogeneration system, excellent economic performance was highlighted by the simulations (payback period equal to 1.5 years, profitability index of 3.88), also due to the significant contribution of the subsidies provided by the current Italian regulation for CHP systems (energy savings certificates, with a market value of 0.207 M€/years).
Further analyses will be carried out, aiming to investigate the influence of the design choices on the performance of the system and the opportunities for improving the efficiency of the trigeneration system, as well as that of the hospital, as a whole.
Author Contributions: All authors provide equality contribute. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.

Conflicts of Interest:
The authors declare no conflict of interest.