Innovative Solutions to Use Ground-Coupled Heat Pumps in Historical Buildings: A Test Case in the City of Napoli, Southern Italy

: The new standards on energy saving for new and existing buildings have animated both researchers and technicians in recent years, aiming at reducing the dependence on fossil fuels, improving indoor comfort, and systems efﬁciency. In this scenario, special attention must be paid to historical buildings that need to preserve their key testimonial heritage within the society. This paper describes the design and realization stages of a pilot system based on a ground-coupled heat pump, operating both in heating and cooling modes, installed in the monumental site of Saints Marcellino and Festo (SM&F), in Naples, Southern Italy. This study aims to demonstrate that low-enthalpy geothermal systems can be employed as energy retroﬁt applications in buildings of historical, artistic, and cultural interest and, at the same time, to prove that the use of this technology allows achieving the objectives, set at global level by the current regulations, and requiring a reduction of carbon dioxide emissions (tCO 2 ) of 53% compared to technology using fossil fuels.


Introduction
Today, climate change and environmental damage are acute problems that must be solved. Europe needs to develop all the available energy strategies to become a modern, resource-efficient, and competitive economic system, and in which net greenhouse gas emissions are no longer generated by 2050. These are the assumptions of the European Green Deal [1], providing the actions to be carried out in order to use efficiently energy resources; to move towards a circular economy, to restore biodiversity, and to reduce pollution. The objective is that by 2050, the European Union (EU) will have zero climate impact. This target requires the employment of various actions, such as investing in environmentally friendly technologies, decarbonizing the energy sector, and ensuring the higher energy efficiency of buildings.
The total energy consumption in the EU is represented, for about 42%, by the building sector, including both residential and services sectors [2]. In order to achieve the EU's energy targets, retrofit operations in the building sector are considered essential, due to the large energy saving potential.
Due to the environmental problems related to the use of fossil fuels, such as the emission of carbon dioxide and other pollutants, the EU promotes the employment of energy efficient systems and the use of renewable energy sources (RES). In the last decade, several targets have been defined in order reduce the greenhouse gas (GHG) emissions and to increase the use of RES.
In addition to the EU targets imposed for 2020, reported in the Directive 2209/29/EC [3], the EU has established more restrictive targets to be achieved by 2030 [4], concerning: the reduction of 40% of GHG emissions, compared to 1990 level; a RES employment of 32%; and an increase of energy efficiency of 32.5%. In Italy, the laws implementing the European directives in the field of energy efficiency, namely European Directives 2010/31/EU [5] and 2009/28/EC [6], require a certain percentage of primary energy needs to be covered by the use of renewable energy sources in buildings for both new and existing buildings.
Therefore, at a time when EU member states are faced with the challenges associated with climate change, the reduction of fossil fuel consumption and the promotion of highefficiency systems using renewable energy sources, for space heating and cooling, becomes a necessary condition. In this context, heat pumps are one of the most appropriate solutions to achieve the decarbonisation of the building sector, and to reduce primary energy consumption in European countries [7,8]. In particular, the technology of ground source heat pumps (GSHP) is currently being studied, allowing the achievement of significant energy savings, and contributing to the energy balance targeted at net zero energy buildings (NZEB) [9]. Novel solutions like geothermal systems can be based on energy piles, on the recovery of freezing probes (employed for the realization of underground tunnels) being reused as geothermal probes, and on ground-coupled heat pumps (GCHP) [10][11][12][13][14][15][16][17].
On the other hand, historical buildings (i.e., those built before 1945), are generally considered as low-performance buildings [18], and make up about 30-40% of the entire building heritage in EU countries [19].
The trend in recent years has seen heritage buildings being converted to other functions, both for private and public use [20]. Churches and historic buildings in Italy, for example, have been transformed into museums, shops, showrooms, or commercial buildings [21,22]. However, as their preservation is essential to maintain the integrity of historic city centres [23], the majority of historic buildings have urban constraints, and are often protected by laws forbidding alterations to their visual appearance, employed materials, and construction techniques [24]. Fabbri and Pretelli [25] state that a heritage building, due to its immeasurable significance, is a historical building under legal protection. According to Fabbri [26], the monuments and buildings of particular architectural interest are the buildings constructed before an agreed historical date, and the buildings that present distinctive constructive and technological features. For example, in the Italian scenario, historical buildings made before 1919 make up about 19% of the total stock, while those that were built between 1919 and 1945 represent about 12% [27].
Taking into account the current regulatory restrictions, different feasibility actions have been analysed by several authors in order to optimize the energy performance of the historical building [28]. Therefore, in this context, where the installation of solar panels or photovoltaic systems for the production of thermal energy and electricity is not allowed, heat pumps are the most appropriate solution, allowing significant improvement in energy efficiency and the possibility to exploit RES, such as geothermal energy [29].
This paper describes the design and realization stages of a pilot plant based on groundcoupled heat pump technology, and allowing both heating and cooling of the room of a real historic building, i.e., the monumental site of Saints Marcellino and Festo (hereinafter SM&F), in Naples, Southern Italy. The building is a part of the historical and artistic regional estates bound by the Soprintendenza Archeologica, Belle Arti e Paesaggio of the city of Naples. The monumental site houses the headquarters of the Department of Earth Sciences, Environment, and Resources of the University Federico II. This study aims to highlight that low-enthalpy geothermal systems can be successfully employed as energy retrofit applications in buildings of historical, artistic, and cultural interest and, simultaneously, do not alter the building itself. The heat pump is connected to a double U geothermal probe inserted in a 120 m depth borehole. The present pilot application demonstrates that geothermal heat pumps can be employed even in densely built places, such as the historic centre of Naples.
The research activity has been carried out within the frame of the SNECS project, Social Network of Historic Centre Entities, and with the support of the DATABENC district (http://www.databenc.it/), with the aim of developing a low-enthalpy geothermal plant in an urban environment, using groundwater temperature. Specifically, a low-enthalpy geothermal plant has been designed and built at the monumental site of SM&F in the ancient Centre of Naples (Figure 1b). The novelty of this research lies in the technology itself and on its application: the heat pump is equipped with multiple sensors and coupled with unconventional geothermal probes, that allow the monitoring of the efficiency of the thermal exchange with the underground water. Moreover, this technology is used for the retrofit operation of a constrained historical building.
The paper is structured as follows: the next section describes the experimental set up designed and realized at the historical site of SM&F, while section three reports the main experimental results obtained, together with the corresponding measurement uncertainties. In the last section, some conclusions are drawn. The city of Naples is in the south-western coastal sector of the Campanian plain, in southern Italy (Figure 1a). It is bordered to the east by the active volcanic district of the Phlegraean Fields, and to the west by the Sebeto valley and Somma-Vesuvius volcano. The stratigraphic setting of the city mainly consists of Neapolitan Yellow Tuff (NYT;~15 ka BP; [31]), a variably lithified pyroclastic current deposit, overlying ancient marine and fluvial deposits, and blanketed by several Phlegraean incoherent ash-fall pyroclastic products, as well as by historical and recent anthropogenic fillings. Along the coastal sector of the ancient Centre of Naples, the top of the tufaceous horizon is at a mean depth ranging between 10 and 20 m b.g.l. (below ground level), and at the bottom the pyroclastic sequence is interdigitated with ancient marine and palustrine sediments (Figure 1b,c).

The Geothermal Plant
From the hydrogeological point of view, the main aquifer is hosted in the NYT complex, characterized by a medium permeability degree due to porosity and fracturing, and a groundwater circulation occurs with a prevailing NNW-SSE flow direction [32]. At the site of SM&F, the water table depth is about 19 m b.g.l. (Figure 1c), thus lying in the tuffaceous aquifer.
The research activity was carried out within the SNECS Project, Social Network of Historic Centre Entities, and is focused on the development of low-enthalpy geothermal plants in urban environments. Specifically, a low-enthalpy geothermal plant was designed and realized at the monumental site of SM&F. Located in the heart of the ancient Centre of Napoli, in Southern Italy, the monumental site of SM&F, with attached cloister and internal garden, has an origin dating back to the early Middle Ages (500÷1000 AD). Of high architectural, artistic, and cultural value, the structure currently houses the headquarters of the Department of Earth Sciences, Environment, and Resources of the University Federico II. Over the years, the structure of SM&F has undergone various modifications, extensions, and restorations. Among these, archaeological excavations carried out in the 1970s and 1990s allowed the reconstruction of the morphology of the soil, and thus of the ancient stratification.
Within this ancient architecture, a pilot geothermal heat pump has been installed, in order to air-condition, both in summer and winter, a classroom of the historical building (area of 22 m 2 and volume of 66 m 3 ), hosting the Association of Geological Sciences Students, to ensure indoor comfort. The archaeological framing of the demonstrator site has been taken from the available archaeological investigation documents.
The urban context in which the geothermal plant was built is characterized by a very high density of buildings. In highly urbanized centres, there are many problems related to the construction of a geothermal plant, mainly related to: (i) the difficulty in drilling in the places identified for the realization of the geothermal well, due to the narrow alleys; (ii) difficulties related to interception, during the excavation phase, of cavities spread in the Neapolitan territory, which would result in the ineffectiveness of the geothermal well. Therefore, before proceeding to the construction of a geothermal plant in an urban centre, a series of investigations must be undertaken to eliminate these problems. Moreover, such a kind of geothermal plant in a historical building needs to undergo a long administrative and technical process, under the control of the Soprintendenza Archeologica, Belle Arti e Paesaggio of the city of Naples, demonstrating that the architecture of the site is not altered.
The present pilot installation demonstrates how it is possible to realize geothermal systems to heat and cool buildings of historical and architectural interest, which are subject to architectural constraints, and often difficult to access by mechanical devices, such as drills for the construction of deep wells. Figures 2 and 3 show the state of the places, regarding the cloister and the portico, before and after the intervention, highlighting how the excavation and drilling processes did not alter the original architectural layout.

The Geothermal System
The essential part of a low-enthalpy geothermal system is made up by the geothermal probes, i.e., a system of pipes inserted in the ground with the function of exchanging heat between the surrounding subsoil and the fluid circulating inside them.
The design of the geothermal plant is strongly influenced by the thermal characteristics of the ground. For this reason, in addition to the geological and hydrogeological surveys, investigations were also conducted to analyze the thermal response of the subsoil in proximity of the probes. The geothermal probes are inserted inside vertical perforations, with the double function of: (i) taking heat from the subsoil (winter), and (ii) transferring heat to the subsoil (summer). Figure 4 shows a simplified rendering of the geothermal system realized at the historical site. Specifically, double U probes have been installed, with a length of 80 m, diameter of 32 mm, and placed inside a borehole with a diameter of 160 mm. The probes, with quality control under SKZ HR 3.26, have the following characteristics: (i) operating temperatures from −20 • C to 30 • C; (ii) maximum operating pressure of 16 bar; (iii) probe foot diameter of 96 mm. Figure 5 shows the double U probes installed at the SM&F site. To facilitate the installation of the geothermal probes, a steel ballast weighing 30 kg was used, with a length of 670 mm and diameter of 80 mm. The weight was applied to the geothermal probe using a set of flat steel bars, to maintain the verticality of the probes. The probes used were manufactured by the German company REHAU, and comply with UNI 10910 and DIN 8074/8075 (type RAUGEO PE-RC duo double U probe made of PE 100-RC material). The probes have been curved at the bottom through a special production process, and further protected by a particular polyester resin core, reinforced with glass fibers. In this way, any risk of non-sealing of the welded connections is excluded, and maximum safety is ensured at the deepest point of the probes. Before being connected to the heat pump, the probes were filled with the working fluid, i.e., water, that circulates within the probes, allowing the heat exchange with the ground.
Two wells have been realized, one for the heat exchange (production well) and the other for controlling and monitoring the aquifer during the operation of the geothermal system (control well). For the construction of the production well, where the double U probes were inserted, a rotation and continuous core drilling geognostic survey was carried out, while for the control well, a rotation and core destruction survey was carried out. The working fluid is water, without antifreeze additives. Figure 6 displays the overall scheme of the geothermal system, with the location of the two wells (S1 and S2) and the classroom that is served by the system (Classroom ASGU-Unina Student Association of Geological Sciences), and on which the tests have been carried out. Moreover, Figure 6 shows the location of the two wells, the geothermal collector, the distribution network, and the heat pump. The experimental facility includes the presence of temperature sensors inside the two boreholes, every 10 m, connected to a data acquisition system, to assess the thermal changes induced in the ground due to the presence of the geothermal probes.
A geothermal collector (Figure 7) to connect the double U probes (two delivery pipes and two return pipes) to the secondary loop of the heat pump has been used. The collector capacity is 10 ÷ 30 L/min. The collector used for the SM&F demonstration plant was supplied by REHAU (type Raugeo Clik pre-assembled with two outlets, tested by the manufacturer). The technical characteristics are: (i) operating temperature from −20 • C to 40 • C; (ii) maximum control pressure 10 bar; (iii) maximum working pressure 6 bar; and (iv) suitable for water and glycol mixtures with a maximum content of 35%. The geothermal heat pump installed at the SM&F demonstrator site is a GSI AQUA DC, with total inverter technology, designed for applications with geothermal type systems powered by a closed-circuit system, and compliant with UNI EN 14511:2004 regulations. The refrigerant fluid within the heat pump is R410a. The compressor has a brushless scroll engine with permanent magnets controlled by a DC electronic driver. A 100 L inertial storage is connected to the system on the user side, in order to prevent the heat pump from experiencing frequent intermittent operation. Figure 8 shows the heat pump and the inertial storage. Finally, fan coils have been chosen as thermal exchange terminals. Due to the low water temperature needs during winter, these kind of terminals allow obtaining high energy efficiency for the system, together with a high comfort level both in summer and winter. The user can control the system by an infrared remote controller. In particular, the installed terminal is characterized by thermal and refrigeration powers to meet the actual loads, i.e., 3410 W for heating and 2650 W for cooling. This terminal is also appropriate for acoustic comfort, as its sound pressure varies from a minimum of 25.2 dBA to a maximum of 42.4 dBA. The electric motor (three-speed) is capable of processing 230 m 3 /h, 340 m 3 /h, and 430 m 3 /h of air at minimum, medium, and maximum fan speed, respectively.
The geothermal system is equipped with a series of sensors that allow the control and the management of the plant remotely, via a web interface. Remote control and management ensure the maximum energy efficiency of the system. Continuous monitoring and remote management of the plant allows obtaining advantages, such as checking the environmental and economic savings compared to a traditional air conditioning system. It is essential to control the state of the system and the thermostatic valve, pump, compressor, and general switches. The remote control system designed and built for the pilot plant is of the SCADA type (supervisory control and data acquisition). Figure 9 shows the main data monitored by the sensors installed on the heat pump. The monitored parameters are: (i) temperature at entrance and exit of the primary and secondary heat exchangers; (ii) speed of the pumps; (iii) pressure values at the entrance and exit of the compressor; (iv) operating power of the compressor; (v) the opening of the thermostatic valve. winter. The user can control the system by an infrared remote controller. In particular, the installed terminal is characterized by thermal and refrigeration powers to meet the actual loads, i.e., 3410 W for heating and 2650 W for cooling. This terminal is also appropriate for acoustic comfort, as its sound pressure varies from a minimum of 25.2 dBA to a maximum of 42.4 dBA. The electric motor (three-speed) is capable of processing 230 m 3 /h, 340 m 3 /h, and 430 m 3 /h of air at minimum, medium, and maximum fan speed, respectively. The geothermal system is equipped with a series of sensors that allow the control and the management of the plant remotely, via a web interface. Remote control and management ensure the maximum energy efficiency of the system. Continuous monitoring and remote management of the plant allows obtaining advantages, such as checking the environmental and economic savings compared to a traditional air conditioning system. It is essential to control the state of the system and the thermostatic valve, pump, compressor, and general switches. The remote control system designed and built for the pilot plant is of the SCADA type (supervisory control and data acquisition). Figure 9 shows the main data monitored by the sensors installed on the heat pump. The monitored parameters are: (i) temperature at entrance and exit of the primary and secondary heat exchangers; (ii) speed of the pumps; (iii) pressure values at the entrance and exit of the compressor; (iv) operating power of the compressor; (v) the opening of the thermostatic valve.

Temperature Sensors
For the temperature measurements of the SM&F plant, two types of temperature sensors have been used. In particular, Pt1000 temperature sensors, using platinum (1000 Ω) technology. The probe (50 mm × Φ60 mm) bulb is made of stainless steel, while the conductor material is tinned copper. The probe is equipped with silicone cables. Ten Pt1000 temperature sensors have been installed, to monitor the temperatures in the geothermal system, as follows:

•
Four sensors positioned at the head of the geothermal well, to monitor the send and return temperatures of the double U geothermal probes; • One sensor inside the classroom, for monitoring ambient air temperature; • One external air temperature sensor, positioned on the north side of the building, bordering the thermal power plant room; • Two sensors used to monitor the sending and return temperatures of the geothermal plant, respectively, located inside the thermal power plant;

Temperature Sensors
For the temperature measurements of the SM&F plant, two types of temperature sensors have been used. In particular, Pt1000 temperature sensors, using platinum (1000 Ω) technology. The probe (50 mm × Φ60 mm) bulb is made of stainless steel, while the conductor material is tinned copper. The probe is equipped with silicone cables. Ten Pt1000 temperature sensors have been installed, to monitor the temperatures in the geothermal system, as follows:

•
Four sensors positioned at the head of the geothermal well, to monitor the send and return temperatures of the double U geothermal probes; • One sensor inside the classroom, for monitoring ambient air temperature; • One external air temperature sensor, positioned on the north side of the building, bordering the thermal power plant room; • Two sensors used to monitor the sending and return temperatures of the geothermal plant, respectively, located inside the thermal power plant; • Two sensors were used to monitor the sending and return temperatures of the primary plant, located inside the thermal power plant.
The second sensor type used are NTC (negative temperature coefficient) sensors. These sensors are located onboard the machine, and are housed at the entrance and exit of the evaporators of the primary and secondary system, respectively.

Heat Meter
A heat meter is usually employed for heat accounting systems, designed for the detection of thermal energy consumption. For the SM&F pilot plant, a compact device, Sensonic II, was chosen, measuring from 0.6 to 2.5 m 3 /h. It is a multi-jet turbine counter, in which the rotation of the turner is detected electronically. The accuracy and stability of measurement over time is ensured by electronic detection and the use of corrosion-resistant materials. Maximum power and thermal energy values are automatically updated every 15 min.

Electrical Energy Analyzer
An electrical energy analyzer was used to manage, monitor, and account for electricity consumption. A single-phase eM111 electricity meter has been installed at the SM&F plant. This counter is equipped with a backlit LCD display with a built-in touch keypad. It is particularly suitable for active electrical energy measurement and cost allocation in applications up to 45 A (direct connection), with dual tariff management availability. It can be used to measure both imported and exported energy, or be programmed to consider only imported energy. It is equipped with housing for DIN guide mounting, and with a ip51 front protection rank.

Results: On Site Experiments and Metrological Analysis
This section describes the results obtained from the pilot geothermal system. The analysis has been carried out on several parameters, i.e., temperatures, thermal power, coefficient of performance (COP), and carbon dioxide emissions.
The experimental results are described in the following sections.

Winter Temperature Analysis
The send and return temperatures on the geothermal side of the heat pump were measured, and reported in Figure 10, together with the temperature difference. The reference period is 8 December 2016 to 8 March 2017. It can be noticed that the average inlet temperature into the probes is equal to 10.9 • C, while the average outlet temperature from the probes is equal to 12.4 • C, and the average temperature difference is equal to 1.5 • C.
The external air temperature was also monitored, to assess its influence on the system. This is reported in Figure 11, observing an average value of 8.8 • C. Table 1 reports the data analyzed on an average weekly basis, in particular the temperature of water sent to the probes (T s ), the return temperature from the probes (T R ), and their difference.

Temperature Analysis in the Control Well (S2)
To verify the influence of the geothermal probe on the temperature of the ground, the temperature at the depths of 25 m, 40 m, 55 m, 65 m, 75 m, 85 m, and 95 m were monitored inside the control well (S2). Table 2 and Figure 12 show the values of these temperatures. The highest temperature was recorded at −25 m, observing a maximum value of 16.9 • C, and an average value of 16.8 • C. From the analysis of Figure 12, it can be noticed that the temperature values are almost constant throughout the data acquisition period. In addition, Figure 12 reports the monitoring on the day 23 December 2016, that is before the operation of the geothermal plant, observing that the temperature does not have a significant variation.   Figure 13 shows the stratigraphy of the ground and temperatures distribution within the control well (S2) over time. In particular, the first line (dotted red) refers to a day where the plant was not yet in operation, while the other lines refer to the plant in operation. It is clear from this figure that the geothermal system did not have a significant influence on the ground temperature surrounding the production well (S1).

Summer Temperature Analysis
The send and return temperatures on the geothermal side were measured and are reported in Figure 14, together with the temperature difference. In particular, the data acquired directly on the probes have been reported. The reference period is from 8 June 2018 to 28 August 2018. The average, maximum, and minimum values monitored by the sensors are reported in Table 3. Table 3. Temperature values acquired by the sensors mounted on the probes (summer).  32  31  29  29  3  2  Max  46  46  43  43  3  3  Min  15  17  16 16 −1 1

Thermal Power Exchanged with the Ground: Winter
The analysis of the thermal power exchanged with the subsoil has been carried out by means of the temperatures read by the management and monitoring system installed inside the heat pump, and of the mass flow rate of the heat transfer fluid, equal to 1500 L/h. The reference period considered is 28 December to 8 March 2017. Table 4 and Figure 15 report the thermal power data recorded in the winter period. Figure 15 also reports the temperature difference between inlet and outlet temperature of the working fluid in the heat pump on the geothermal side. The analysis of these data indicated a maximum thermal power per meter of hole equal to 61 W/m has been obtained, with a minimum value of 32 W/m, and an average value 41 W/m.   Figure 16 shows the thermal power exchanged with the subsoil by the geothermal probe in the summer period. The figure reports the power exchanged by the double U probes, referred to each probe, based on the temperature data acquired by the sensors installed on each probe, in correspondence with the geothermal collector (refer to Figure 7). In particular, both the maximum and average thermal power are equal to 18.1 kW and 4.6 kW, respectively, in the period from 13 June 2018 to 26 June 2018.   The summary of the results related to the thermal power exchanged with the subsoil, referred to the measurements both on the probes and on the heat pump, is reported in Table 5. The heat transfer per unit of probe length is obtained by dividing the total power by the length of the probe, which in this case is 320 m, while heat transfer per unit of well length is obtained by dividing the power exchanged by 80 m or the depth of the well. The analysis of the average values reported in Table 5 and of the data reported in Figure 17 shows that in the distribution network from the geothermal collector to the heat pump (length of the tubes of about 25 m), about 4.8% of the thermal power exchanged with the ground is lost.

Coefficient of Performance Analysis
This section displays the analysis on the coefficient of performance (COP) of the heat pump, with the heating system running (fan coil in operation) for at least two hours. The days considered for the analysis are:  Figure 18 displays the dynamic COP values, the COP certified by the manufacturer equal to 4.32, the average value of the dynamic COP, and its minimum and maximum values. The reference period is the one referred to by letter A above (18/01/2017 from 09:06:36 to 11:06:46). The experimental analysis shows that the actual values are larger than the COP value certified by the manufacturer. The heat pump attains its minimum value, equal to the one declared by the manufacturer, only at start-up. Based on these real data, the geothermal system can be considered efficient.

Carbon Dioxide Emissions Analysis
This section displays the analysis of carbon dioxide emissions, comparing the average emissions due to the geothermal heat pump installed at the SM&F demonstrator site (we refer to the carbon dioxide equivalent emissions due to the generation of consumed electricity), with those of state-of-the-art condensing boilers with an efficiency of 0.95. Figure 22 shows the CO 2 emissions due to a typical condensing boiler, due to the present geothermal heat pump, the corresponding average values, and the avoided emissions, derived from the comparison between the two systems. The weekly average boiler emission was 0.20 tCO 2 , while the average heat pump emission was 0.09 tCO 2 , with a consequent average avoided emission of 0.12 tCO 2 .

Metrological Analysis
An analysis of type A and type B measurement uncertainties [33] was conducted based on the experimental data collected during the campaign on the geothermal system.
The data were acquired on the system every 30 s. For the day taken into account in this analysis, the data acquired were 2880. To calculate the type A uncertainties, the acquired data were referred to intervals of five minutes, obtaining ten measurements for each interval. Type B uncertainties were calculated based on the information declared by the manufacturer of the measurement instruments.
Type A uncertainty was calculated as u A = s/ √ N, s being the standard deviation of the data sample, and N the number of data of the sample. The type B uncertainty declared by the manufacturer f the temperature sensors is equal to 0.25 • C. Finally, the expanded combined uncertainty is calculated as U c = k · u c = k · u 2 A + u 2 B , being k the coverage factor, considered equal to 2, corresponding to a confidence level of 95.4%. Table 6 reports the temperature values acquired on the heat pump inlet and outlet sections, together with the relative uncertainty. The uncertainty values can be considered acceptable for the present analysis. Figure 23 reports the inlet and outlet temperatures on the heat pump with the corresponding uncertainties, recorded during the 24 h of July 5, 2018. In the figure, a drop in temperature can be observed at minute 556. This is due to the mechanical operation on the machine, because the temperature of the water in the storage tank was equal to the flow temperature of the fan coils.

Conclusions
This paper describes the experimental activity carried out on the pilot system, based on a geothermal heat pump, designed and installed at the Monumental Site of Saint Marcellino and Festo (SM&F). The authors have taken into account both winter and summer operation of the system, that has been employed to heat and cool a classroom by means of a fan coil.
From the present analyses, it is possible to notice that, based on the dynamic values acquired on site, the maximum and average thermal power exchanged during winter are 4.5 kW and 2.5 kW, respectively. The heat exchange takes place for about 60 m of borehole, obtaining a maximum thermal power for meter of well during winter equal to 61 W/m, a minimum value of 32 W/m, and an average value 41 W/m. During summer, the maximum and average thermal powers are equal to 18.1 kW and 4.6 kW, respectively.
As concerns the coefficient of performance (COP), the data acquired on site have been compared with the value certified by the manufacturer, equal to 4.32. From the present analysis, it is possible to observe that the actual COP is always larger than the one declared by the manufacturer, with a peak value of 6.98 and average values around 5.65. By using the present technology, it is possible to reduce the CO 2 emissions with respect to the use of condensing boilers, obtaining up to 0.14 tCO 2 avoided over the analyzed period of time.
The present work has demonstrated that the energy efficiency improvement of a historic building can be achieved by means of geothermal heat pumps, without causing any alteration of the artistic-cultural aspects of the building, protected by the competent authorities.