Energy and Buildings 43 (2011) 787795

Contents lists available at ScienceDirect

Energy and Buildings

journal homepage: www.elsevier.com/locate/enbuild

Economic and exergy analysis of alternative plants for a zero carbon building complex
Tiziano Terlizzese, Enzo Zanchini
Dipartimento di Ingegneria Energetica, Nucleare e del Controllo Ambientale, Universit di Bologna, Viale Risorgimento 2, I-40136, Bologna, Italy

a r t i c l e

i n f o

a b s t r a c t
The feasibility of zero carbon emission plants for heating, air conditioning and domestic hot water (DHW) supply, is analyzed, with respect to conventional plants, for a new residential building complex to be constructed, in Northern Italy. Two zero carbon plants are considered: the rst is composed of air-towater heat pumps for space heating and cooling, PV solar collectors, air dehumidiers, thermal solar collectors and a wood pellet boiler for DHW supply; in the second, the air-to-water heat pumps are replaced by ground-coupled heat pumps. The conventional plant is composed of a condensing gas boiler, single-apartment air to air heat pumps, and thermal solar collectors. The economic analysis shows that both zero carbon plants are feasible, and that the air-to air heat pumps yield a shorter payback time. The exergy analysis conrms the feasibility of both plants, and shows that the ground coupled heat pumps yield a higher exergy saving. 2010 Elsevier B.V. All rights reserved.

Article history: Received 5 March 2010 Received in revised form 12 November 2010 Accepted 23 November 2010 Keywords: Zero carbon buildings Heat pumps Solar collectors Dynamic simulation Exergy analysis

1. Introduction Since a few decades, improving the energy efciency of buildings, and possibly reach zero energy use for space heating and cooling and DHW production, is considered as an important technical target both in industrialized and in developing countries; thus, much research activity in this eld has been performed worldwide. Peippo et al. [1] proposed a procedure for the optimum design trade-off strategy for solar low energy buildings, and reported some qualitative results of the procedure for a single family residential house and a large electricity intensive ofce building, with reference to three different climatic zones in Europe. Balaras [2] audited 8 apartment buildings, located in three climatic zones of Greece, and showed that a considerable energy saving in heating, air conditioning, DHW production and lighting can be obtained by proper retrot actions. Iqbal [3] studied the feasibility of a zero energy one family home in Newfoundland, Canada, in which a grid connected 10 kW wind turbine provides the electric energy for space and water heating, cooking, lighting and appliances; he found that the total cost of the wind energy system is about 30% of the cost of the house. Rijksen et al. [4] studied, both experimentally and through dynamic simulation, the reduction of peak cooling requirement for an ofce building obtainable by means of thermally activated building systems (TABS); TABS have pipes embedded in the concrete oor, to carry water for heating and

Corresponding author. Tel.: +39 051 2093295; fax: +39 051 2093296. E-mail address: enzo.zanchini@unibo.it (E. Zanchini). 0378-7788/$ see front matter 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2010.11.019

cooling. Zhao et al. [5] designed and studied numerically a novel dew point air conditioning systems, and Zhao et al. [6] investigated the feasibility of this system in several China regions. Chan et al. [7] pointed out advantages and limitations of passive solar heating and cooling technologies and suggested research guidelines to improve the economic feasibility of these techniques. Wang et al. [8], discussed possible solutions for zero energy building design in UK. They showed that zero energy buildings, in which energy for heating, air conditioning, DHW, lighting and home appliances is provided by PV and thermal solar collectors and wind turbines, are theoretically possible in UK. They also provided optimization criteria for the building insulation and orientation, but did not perform an economic or exergy feasibility analysis. The aim of the present paper is to analyze the economic and exergy feasibility of zero carbon emission plants for heating, cooling and DHW supply, for a residential building complex planned for construction in a village close to Bologna, in Northern Italy. The transmittance of walls and windows is assumed as xed, and two alternative zero carbon plants are designed and studied by dynamic simulations, performed through TRNSYS 16, and life cycle analysis. The rst plant is based on air-to-air heat pumps and PV collectors, the second on ground coupled heat pumps and PV collectors. The economic and exergy payback time of these plants is determined with respect to a traditional plant, composed of a condensing gas boiler and single-apartment heat pumps for air conditioning. This kind of plant is still the most commonly employed for residential buildings in Northern Italy, where winter loads are important and air conditioning is usually not provided by the building constructor, but installed by single apartment owners. For all the plants

788

T. Terlizzese, E. Zanchini / Energy and Buildings 43 (2011) 787795

els are employed and fresh air is supplied by a forced ventilation circuit, provided with a humidity control and heat recovery system. 3. Energy demand for heating, cooling and DHW supply The component materials of the external wall, between the timber pillars, and their thermal properties are listed in Table 1, starting from outside. Oriented Strand Board (OSB) is manufactured from waterproof wood strands, that are arranged in cross-oriented layers. For air layers, the effective thermal conductivity is reported in Table 1, evaluated as = s R (1)

Fig. 1. Layout of the building complex.

considered, about 70% of the DHW energy use is supplied by thermal solar collectors. The economic analysis shows that both zero carbon plants are feasible, and that the air-to air heat pumps yield a shorter payback time. On the other hand, the ground coupled heat pumps appear as preferable from the exergy analysis viewpoint. 2. Description of the building complex and of the plants The building complex is composed of seven four-apartment houses and ve two-apartment houses; a layout of the complex is reported in Fig. 1. Each apartment has a heated oor area of 111.41 m2 , so that the total heated oor area of the complex (38 apartments) is about 4234 m2 . Each apartment has two oors. The ground oor is composed of an entrance hall, a living room, a bathroom and a garage (unheated). The rst oor is composed of a kitchen with dining room, two bedrooms, a bathroom and a small terrace. All houses have a timber frame and wooden walls, and are insulated with wood-derived insulating materials. A view of a house with 4 apartments and a map of the rst oor are illustrated in Fig. 2. Two alternative zero carbon plants, named Plant A and Plant B, and a conventional plant, named Plant C are considered. Plant A is composed of air-to-water heat pumps (AWHPs), with electric energy supplied by PV collectors, which provide heating and cooling; air dehumidiers; thermal solar collectors and a wood pellet boiler, which provide DHW. Plant B is similar to Plant A, but AWHPs are replaced by ground-coupled heat pumps (GCHPs). Plant C is composed of a central condensing gas boiler for heating and single apartment heat pumps for air conditioning; DHW is supplied by thermal solar collectors and by the gas boiler. The thermal solar plant for DWH, designed by the f-chart method [9] as illustrated in Section 3, is the same in all cases: it provides 70% of DHW energy use. In each case, oor radiant pan-

where s is the thickness and R is the thermal resistance per unit area of the layer. The transmittance of the external wall, evaluated according to EN ISO 6946:2008, is 0.170 W/(m2 K) in correspondence of the wood ber insulation (layers listed in Table 1) and 0.326 W/(m2 K) in correspondence of the timber frame; the latter covers about 10% of the total wall area, so that the average transmittance of the external wall is about 0.186 W/(m2 K). In the dynamic simulation, the thermal resistance of the external surface has been evaluated as a function of the wind velocity and of the external surface temperature. The roof has a composition similar to that of the external vertical wall. The wood beams, which are placed under the roof, provide an additional thermal resistance. The roof transmittance, evaluated according to EN ISO 6946:2008, is 0.15 W/(m2 K) in correspondence of the timber frame, which covers about 22% of the total roof area, and 0.21 W/(m2 K) elsewhere; therefore, the average transmittance of the roof is about 0.197 W/(m2 K). The heat exchange between building and ground has been evaluated by considering the real, time-dependent, temperature distribution in the soil, determined by means of TRNSYS Type 501. The ground is composed of heavy clay with 15% water content. The following values of the ground thermal conductivity kgd and heat capacity per unit volume ( cp )gd have been considered: kgd = 1.70 W/(m K); ( cp )gd = 2.938 MJ/(m3 K) [10]. Double glazed windows with 4 mm thick panes separated by a 16 mm thick argon layer have been considered. The window transmittance, including frame, is 1.4 W/(m2 K); the frame area is 20% of the total window area, and the glazed surface solar factor is g = 0.589. Shadowing effects have been considered to evaluate solar energy gains. The width of the shading devices placed above the windows (see Fig. 2) has been designed in order to shelter completely the direct solar radiation from April 15th to September 15th for windows facing South. The heat capacity of internal walls has been taken into account. The internal heat loads have been evaluated, for each hour, according to ISO 13790:2008. The heat loss due to ventilation has been determined by assuming an air change rate of 0.3 h1 and the employment of a heat recovery system with efciency 0.6. The weather data for Bologna have been considered, with reference to the typical meteorological year (TMY) determined by Remund and Kunz [11]; these data are available in the default TRNSYS 16 climatic data packages. For the TMY considered, the monthly averaged air temperatures are reported in Table 2, while the values of both the beam and the diffuse solar radiation incident on a horizontal surface, during each month, are illustrated in Fig. 3. During winter, the internal air temperature is set at 20 C during the day and at 18 C during the night, except for bathrooms, where it is kept 2 C higher. During summer, the internal air temperature is set at 28 C during the day, the cooling system is turned off during the night, while the relative humidity of the internal air is kept at 50% both night and day. The heating and cooling heat loads for the

T. Terlizzese, E. Zanchini / Energy and Buildings 43 (2011) 787795

789

Fig. 2. House with 4 apartments: view of the building and map of the rst oor.

Table 1 Materials of the external wall: s = thickness [cm]; = thermal conductivity (for air, effective thermal conductivity) [W/(m K)]; c = heat capacity per unit volume [MJ/(m3 K)]; e = emissivity. Material 1: Plaster 2: Mineralized wood ber 3: Air 4: Vapour barrier 5: Air 6: Low e layer 7: Mineralized wood ber 8: OSB 9: Wood ber 10: Air 11: Vapour barrier 12: OSB 13: Mineralized wood ber 14: Cellulosegypsum board s 0.5 5 4 0.1 4.5 0.1 3.5 1.2 12 2 0.1 1.2 5 1.3 0.9 0.083 0.222 0.077 0.149 0.071 0.083 0.13 0.038 0.111 0.071 0.13 0.083 0.32 c 1.638 0.840 0.000 0.034 0.000 0.034 0.756 1.701 0.105 0.000 0.034 1.701 0.756 1.265

whole building complex, in kW, are illustrated in Fig. 4. The annual energy need for the whole building complex is: 131.75 MWh for heating, 64.00 MWh for cooling, 25.85 MWh for dehumidifying. The domestic hot water demand has been determined by employing the national Technical Standard UNI TS 11,300, as is
Table 2 Monthly averaged temperatures of the TMY. Month January February March April May June July August September October November December Average temperature [ C] 1.72 4.35 9.43 13.84 20.19 21.55 24.45 24.12 20.96 14.44 8.39 3.88

790
400 2 MJ/m 350 300 250 200 150 100 50 0 January March May July

T. Terlizzese, E. Zanchini / Energy and Buildings 43 (2011) 787795

Beam Diffuse

September

November

Fig. 3. Monthly values of beam and diffuse radiation on a horizontal surface for Bologna, during a typical meteorological year.

kW 160 heating 120 cooling 80 40 0 0 1460 2920 4380 5840 7300 8760 heating

hours
Fig. 4. Heating (dark gray) and cooling (gray) heat load for the whole building complex during a typical meteorological year.

prescribed by the Regional Law 156/2008. The result, for the DHW demand, is 165.66 L per day, per each apartment. By assuming a temperature rise from 15 C to 40 C, one obtains a total energy need, for the whole building complex, given by Endhw = 66.70 MWh. The annual energy needs for heating, cooling, dehumidifying, and DHW supply are summarized in Table 3. 4. Plant sizing and primary energy use A oor radiant panel heat distribution system is adopted, in each plant for heating, in Plants A and B also for cooling. The distribution efciency, the emission efciency and the control efciency have been evaluated according to the national Technical Standard UNI/TS 11300; their values are, respectively: d = 0.97, e = 0.99, c = 0.99. For Plant B, double U tube borehole heat exchangers (BHEs) with the following features have been considered: high density polyethylene tubes SDR 11 with external diameter 32 mm; borehole diameter 156 mm; grout thermal conductivity 1.1 W/(m K), hence borehole thermal resistance 0.095 m K/W. The undisturbed ground temperature, i.e., the average temperature of the ground from the soil surface to the BHE bottom (100 m), before the beginning of the BHE eld operation, has been assumed equal to 14 C. The GCHP system has two water tanks: WT1, between BHEs and
Table 3 Annual energy needs for the whole building complex. Kind of service Heating Cooling Dehumidifying DHW Energy need (MWh) 131.75 64.00 25.85 66.70

heat pumps; WT2, between heat pumps and radiant panels. The total length of the BHEs has been designed by iterative simulations, performed through TRNSYS. A scheme of Plant B during winter operation is reported in Fig. 5, where red lines represent warmer water, blue lines cooler water, arrows in lines denote the water ow direction, and large arrows at the sides of the heat pumps denote the energy ow direction. The water tank WT2 is present in all the plants considered. For this tank, a maximum water temperature equal to 35 C has been assumed; the latter is sufcient to match the design heat load of 166.9 kW (external temperature 5 C). Both for Plant A and for Plant B, two heat pumps, with a heating power of 79.5 kW each, have been selected, so that the maximum max = 159 kW. For heating power supplied by the heat pumps is Qhp each plant, the coefcient of performance (COP) of the heat pumps has been evaluated for each hour, by considering the external air temperature (Plant A) or the water temperature in WT1 (Plant B), with a constant value of the water temperature in WT2 (35 C). For Plant A, the COP of the air-to-water heat pumps as a function of the external air temperature and of the supply water temperature provided by the manufacturer has been employed, after a comparison with available experimental data. A reliable experimental evaluation of the long term COP of air-to-water heat pumps operating in conditions similar to those considered in this paper has been provided by Marcic [12]. The author presents the results of the monitoring, during the period 19881998, of an air-to-water heat pump installed in 1988 which supplies water at a mean temperature of 40 C. In Fig. 6, three plots of the COP of air-to-air heat pumps versus the external air temperature are reported: the plot in light gray refers to the heat pumps considered in this paper, with a supply water temperature of 35 C; the plot in dark gray refers to the heat pumps considered in this paper, with a supply water temperature of 40 C; the plot in black refers to the heat pump monitored by Marcic (supply water temperature 40 C). The gure shows that, with reference to the same supply water temperature, the COP of the heat pumps considered in this paper is about 19% higher than that measured by Marcic. A recent report available in the literature [13] shows that, on account of technological improvement, the percent COP increase of air-to-water heat pumps from 1986 to 2004 is about 25%. Therefore, the COP data provided by the manufacturer of the air-to-water heat pumps considered in this paper have been considered as reliable and employed in calculations. For Plant B, which uses GCHPs, reliable experimental data in working conditions similar to those employed in this paper are not available in the literature. Therefore, the COP data provided by the manufacturer have been employed. For a water temperature in WT2 equal to 35 C, the COP as a function of the water temperature in WT1 is given by 0.12 TWT1 + 4.4, where TWT1 is the water temperature in WT1 expressed in degrees Celsius. The seasonal weighted mean values of the COP obtained are as follows: for Plant A, COP = 3.81 during the heating period and COP = 3.60 during the cooling period; for Plant B, COP = 5.32 during the heating period, while the heat pumps are not used for cooling (free cooling). For Plants A and B, the power supplied to the building is

s = Q

n Q hp + Q aux , =Q d e c

(2)

hp n is the net thermal power required by the building, Q where Q is the thermal power supplied by the heat pumps and Qaux is the auxiliary thermal power for heating supplied by the wood pellet boiler.

T. Terlizzese, E. Zanchini / Energy and Buildings 43 (2011) 787795

791

thermal solar collectors PV solar collectors

building complex

DHW tank boiler

WT2 WT1 heat pumps BHEs

Fig. 5. Scheme of Plant B.

For Plant B, the power extracted from the ground to meet the winter heat load is given by gd = Q hp 1 Q 1 COP (3)

Simulations of the BHEs have been performed through TRNSYS Type 557, by employing the data obtained with Eq. (3). The total length of the BHEs has been determined by iterations, in order to obtain a minimum temperature of WT1 not lower than 4 C. A total length of 4000 m has been obtained, which corresponds to 40 BHEs 100 m deep. A plot of the temperature of WT1 versus time, for a period of two years, is reported in Fig. 7. The gure shows that the temperature of WT1 during summer exceeds 18 C only exceptionally. The thermal power subtracted from the building by the radiant panels, with a water inlet temperature of 18 C and an internal air temperature of 28 C, is 28.9 W/m2 . Simulations of the apartments have been performed by TRNSYS under the following constraints: the maximum heating power during winter, per unit oor area, is equal to the design heating power, for each room, and the maximum cooling power per unit oor area during summer is equal to 28.9 W/m2 . By means of these simulations, the electric energy required by the heat pump system has been determined, for Plants A and B. Moreover, the thermal energy
5 COP 4
Present paper, 35 C Present paper, 40 C

supplied for heating by the wood pellet boiler during one year has been evaluated. Finally, it has been veried that, for Plant B, the internal set point temperature (28 C) is reached in summer by free cooling, i.e., sending water directly from WT1 to the radiant panels. For Plants A and B, during summer nights the water ow in radiant panels is stopped; nevertheless, the internal air temperature is usually lower than 29 C and exceeds this value only exceptionally. These temperature conditions and 50% relative humidity have been considered as satisfying. The electric energy consumed by the heat pumps per year, Ehp , hp /COP. has been determined as the integral, during one year of Q The results are: Ehp = 55.10 MWh for Plant A (36.38 MWh for heating and 18.72 MWh for cooling); Ehp = 26.05 MWh for Plant B, for heating (free cooling is adopted). The electric energy use for water circulation, dehumidication (Plant A and Plant B) and single apartment air conditioning (Plant C) has been evaluated as follows. For the piping system between WT2 and the radiant panels, the total head loss and ow rate are respectively 69.9 kPa and 8.03 L/s. The estimated electric energy consumption is 5.34 MWh per year for Plants A and B; 3.59 MWh per year for Plant C (where radiant

20 C 18 16 14 12
Marcic

10 8 6

4 2

0 0 2 4 6 8 10 12 C 14

2920

5840

8760

11680

14600

17520

hours
Fig. 7. Temperature of WT1 versus time, for a period of 2 years.

Fig. 6. COP of air-to-air heat pumps versus external air temperature.

792 Table 4 Annual electric energy use for Plants A, B and C. Plant A (MWh) Heating Cooling Dehumidifying Radiant panel pumping DHW loop pumping BHE loop pumping Total 36.38 18.72 12.30 5.34 0.05 72.79

T. Terlizzese, E. Zanchini / Energy and Buildings 43 (2011) 787795

0.9

f
Plant C (MWh) 20.68 8.35 3.59 0.05 32.67

Plant B (MWh) 26.05 12.30 5.34 0.05 8.51 52.25

0.8 0.7 0.6 0.5 0.4

panels are used only for heating). The electric energy use for pumping domestic hot water is about 0.05 MWh per year. The electric energy use for dehumidication, for Plant A and Plant B, has been evaluated by assuming the COP of air dehumidiers equal to 2.1; the result is 12.30 MWh per year. For Plant C, the electric energy consumed by the singleapartment heat pumps for cooling and dehumidifying has been determined by considering the hourly thermal loads evaluated by TRNSYS and the COP data provided by the constructor. In analogy with Plants A and B, the thermal loads have been evaluated by assuming that both temperature and relative humidity are controlled during the day, while only the relative humidity is controlled during the night. The result is 29.03 MWh per year (20.68 MWh for cooling and 8.35 MWh for dehumidifying), and the weighted mean value of the COP is 3.10. For Plant B, also the energy use for the BHE loop pumping must be considered. The BHE piping system is composed of 8 parallel loops, each with 5 BHEs piped in parallel. The water ow rate is 20 L per minute, for each BHE. The total head loss, evaluated as suggested in Ref. [14], is 93.9 kPa, and the estimated energy consumption is 8.51 MWh per year. The values of the electric energy used for heating, cooling, dehumidifying and pumping, for Plants A, B and C, are summarized in Table 4. For Plants A and B, the PV collectors have been sized in order to supply exactly the total use of electric energy reported in Table 4, in a typical meteorological year, namely 72.79 MWh of electric energy for Plant A, and 52.25 MWh of electric energy for Plant B; the design software available in Ref. [15] has been employed. The following PV system features have been considered: tilt angle 14 , azimuth angle 21 , combined PV system losses 25.5%. The desired energy supply is obtained by a PV system with 71.2 kWp (peak power) for Plant A, with 51.2 kWp for Plant B. The PV collectors are roof-integrated, in each house. The total PV collector area is 569.6 m2 for plant A and 409.6 m2 for plant B, i.e., about 60 m2 for Plant A and about 43 m2 for Plant B, for a house with four apartments. For Plants A and B, the auxiliary thermal energy for heating per year, supplied by the wood pellet boiler, is Eaux = 0.11 MWh. For all the plants considered, the thermal energy supplied per year to the DHW system is Esdhw = Endhw , (d st s )dhw (4)

40

60

80

100
2

120

140

Collector area [m ]
Fig. 8. Plot of the fraction f of the annual thermal energy use for DHW provided by solar collectors as a function of the transparent collector area.

the fraction f of the annual thermal energy use for DHW provided by solar collectors, as a function of monthly thermal loads, climatic data, collector performance parameters, storage volume and collector area. Clearly, the same climatic data employed for the building simulation [11] have been used. The total radiation per unit area incident on the collector surface has been evaluated by TRNSYS Type 16. The latter employs hourly data of both direct and diffuse radiation on a horizontal surface, and thus accounts for the effects of clouds. In Fig. 8, the fraction f of the annual thermal energy use for DHW provided by solar collectors is plotted versus the transparent collector area. The gure shows that solar collectors provide 70% of Esdhw with a transparent area of about 87.5 m2 . Thermal collectors are placed on the roof of a detached plant room, which contains WT2 and the DHW tank, the wood pellet boiler and the heat pumps for Plants A and B, the condensing gas boiler for Plant C, and WT1 for Plant B. For Plants A and B, the total thermal energy supplied by the wood pellet boiler during one year is given by Ewpb = Esdhw (1 f ) + Eaux , wpb (5)

where d , st and s are the distribution, storage and supply efciencies for the domestic hot water system, which have been evaluated according to EN 15316-3-1:2007. Their product is 0.89; thus, since Endhw = 66.70 MWh, one obtains Esdhw = 74.94 MWh. To meet a part of the thermal load Esdhw , single glazed at plane thermal solar collectors with a selective absorbing surface have been chosen, with the following plant features: tilt angle 45 ; azimuth angle 0 ; FR ( )0 = 0.824, where FR is the heat removal factor and ( )0 is the effective transmittanceabsorptance product at normal incidence; FR UL = 3.66 W/(m2 K), where UL is the overall heat transfer coefcient; storage volume 75 kg/m2 . The plant has been sized by the f-chart method [9], which allows to determine

where f is the fraction of Esdhw supplied by the thermal solar collectors, Eaux is the auxiliary thermal energy for heating per year supplied by the boiler, and wpb is the boiler efciency. A wood pellet boiler with 200 kW power and an efciency equal to 0.92 has been chosen. Indeed, an analysis of the technical data provided by constructors has shown that the efciency of wood pellet boilers produced nowadays ranges from 0.9 to 0.95. Thermal solar collectors have been sized to yield f = 0.70, so that Ewpb = 24.56 MWh. Since electric energy for heat pumps, dehumidiers and water circulation is provided by PV collectors, for Plants A and B Ewpb = 24.56 MWh is the total primary energy use of the building complex for heating, cooling, dehumidifying and DHW supply. This consumption corresponds to 5.80 kWh/(m2 year), with zero carbon emission. For Plant C, the efciency of the condensing gas boiler has been considered as equal to 1.05. The product of the distribution, the emission and the control efciency for the heating system is d e c = 0.95, and the product of the distribution, the storage and the supply efciency for the DHW system is (d st s )dhw = 0.89. Thus, the total plant efciency is 1.00 for heating and 0.94 for DHW supply, and the energy use to provide space heating and 30% of the DHW energy need is 153.04 MWh. This consumption corresponds to 36.15 kWh/(m2 year), to which the use of 32.67 MWh of electric energy, for cooling, dehumidifying and pumping, must be added. The use of primary energy which corresponds to this use of electricity has been determined according to the Resolution EEN 3/08 of the Italian Agency for Electric Energy and Gas (AEEG), which states

T. Terlizzese, E. Zanchini / Energy and Buildings 43 (2011) 787795 Table 5 Plant costs. Plant A Heat pumps Dehumidiers PV solar collectors Pellet boiler Total Plant B BHE, loop and pump Cold tank Heat pumps Dehumidiers PV solar collectors Pellet boiler Total Plant C Gas boiler Air to air heat pumps Total 11,000 D 114,000 D 125,000 D 205,200 D 2000 D 40,000 D 20,500 D 245,800 D 12,000 D 525,500 D 40,000 D 20,500 D 341,800 D 12,000 D 414,300 D Plant A Wood pellet PV electricity Maintenance Annual income Plant B Wood pellet PV electricity Maintenance Annual income Plant C Methane Electricity Annual cost Table 6 Annual cost (income) for energy use (production). Cost 1200 D 30,700 D 3300 D 26,200 D 1200 D 22,000 D 2400 D 18,400 D 11,200 D 8200 D 19,400 D

793

Income

that 1 kW h of electric energy corresponds to 2.175 kWh of primary energy; the result is 78.63 MWh of primary energy. Therefore, the total use of primary energy per year for plant C is 231.67 MWh. 5. Economic analysis The economic feasibility of Plants A and B has been analyzed by comparison with a conventional heating and cooling plant, called Plant C. The same thermal solar collector system has been considered, for all plants. Since a comparative economic analysis of Plants A, B and C has been performed, the costs of the common components, present in all plants, have not been considered. These components are: radiant panels, water distribution system to radiant panels, tank WT2, thermal solar collector system, DHW distribution circuit. The capital costs of Plants A, B and C, excluding the common components, are reported in Table 5. For the PV systems, a capital cost of 4800 D /kWp has been considered. For the BHE system of Plant B, a cost of 50 D /m has been considered for the BHEs (length 4000 m and total cost 200,000 D ), plus a cost of 2600 D for pipes and pumps and a cost of 2600 D for labour and machinery use. The table shows that Plant B is the most expensive and that the ratio between the capital cost of Plant B and that of Plant A is about 1.27. The operating costs have been evaluated as follows. The current rates of fuels and electricity in Bologna have been considered: 0.23 D /kg for wood pellet; 0.70 D /m3 for natural gas; 0.25 D /kW h for electricity. The State nancial support given for PV electricity production in Italy has been taken into account: only the annual difference between the electric energy consumed by the plant and the electric energy produced by the PV system is paid by the user (zero in this case); all the PV electricity produced is paid by the State at the rate 0.422 D /kW h, for roof-integrated PV Panels. For Plant A and Plant B, an additional maintenance cost has been considered, with respect to Plant C, by assuming that the additive maintenance cost is due only to the PV system, because also Plant C has heat pumps, for summer cooling and dehumidifying. Indeed, PV systems require periodical maintenance activities such as module cleaning, visual checking of the electrical wiring system, and checking of module watertight seals. An annual maintenance cost equal to 46 D /kWp, which corresponds to the average service cost of local maintenance companies, has been assumed. Hence, the additive maintenance cost has been evaluated as equal to 3300 D /year for Plant A, and to 2400 D /year for Plant B. The annual operating costs/incomes for Plants A, B and C are reported in Table 6.

On account of the uncertainty in the previsions of the cost of money and on the annual increase of the unit costs of fuels and electricity, we have performed our economic analysis by assuming zero cost of money and zero annual increase of fuels and electricity costs. The total capital plus operating cost versus time is plotted in Fig. 9, for each plant, for a period of 20 years. The gure shows that Plant A is the most convenient. Its payback time, with respect to Plant C, is about 6 years, while that of Plant B is about 11 years; moreover it has a total cost always lower than that of Plant B. Clearly, the results illustrated in Fig. 9 are strongly inuenced by the presence of PV systems with different areas and by the State incentives to PV electricity production. Therefore, it may be interesting to perform a comparative economic analysis of Plants A, B, C, in the absence of PV systems. The results of this analysis are reported in Fig. 10, and show that Plant A remains the most convenient, for a time interval of 20 years. 6. Exergy analysis A comparative exergy analysis of Plants A, B and C has been performed. As usual, we will call embodied energy of a plant component the exergy loss due to its construction and installation. In analogy with the economic analysis, the embodied energy of the common components of Plants A, B, and C has not been considered. For each plant, the embodied energy of each non-common component has been evaluated as follows. For heat pumps, boiler, dehumidiers and tanks, the real mass has been considered, together with the mass fractions of the constituent materials given in Ref. [16], while the value of the embodied energy of each material, per unit mass, has been taken from Ref. [17]. For the high density polyethylene tubes of BHEs, the real mass has been con-

Fig. 9. Capital plus operating cost versus time, for Plants A, B, C.

794
600000

T. Terlizzese, E. Zanchini / Energy and Buildings 43 (2011) 787795

4000 MWh Plant A Plant B Plant C 3500 3000 2500 2000 1500 1000 Plant A Plant B Plant C

500000 400000 300000 200000 100000

500 0 0 5 10 15 0 0 5 10 15

years

20

years

20

Fig. 10. Capital plus operating cost versus time, for Plants A, B, C, in the absence of PV collectors.

Fig. 11. Total (construction + operation) exergy use versus time, for Plants A, B, C.
4000 3500 3000 2500 2000 1500 1000 500 0 0 5 10 15 Plant A Plant B Plant C

sidered and the value of the embodied energy per unit mass has been taken from Ref. [17], by considering the feedstock energy as no longer available. The embodied energy of PV collectors has been evaluated by assuming an embodied energy per unit peak power equal to 8.5 MWh/kWp, as reported in Ref. [18]. The exergy loss due to borehole drilling has been evaluated by considering a diesel fuel consumption of 1 L per each meter of borehole (typical consumption for the soil considered), and by approximating the diesel fuel exergy with its lower heating value, namely 10.02 kW h/L [19,20]. The values of the embodied energy for the non common components of Plants A, B, and C are summarized in Table 7. The table shows that the total embodied energy for Plant B is greater than that for Plant A, and that (excluding the components common to all plants) the ratio between the embodied energy of Plant B and that of Plant A is about 1.43. For each plant, the exergy loss due to the plant operation during a typical meteorological year has been evaluated, by approximating the fuel (methane or wood pellet) exergy with its lower heating value. With this approximation, the annual exergy use for Plants A and B is 24.56 MWh, while the annual exergy use for Plant C is 231.67 MWh, as is shown in Section 3. Plots of the total exergy loss due to the plant construction and operation versus time, for a period of 20 years, are illustrated in Fig. 11 for Plants A, B and C. The gure shows that the lowest exergy use after 20 years is obtained by Plant A.
Table 7 Values of the embodied energy for the non common components of Plants A, B, and C. Plant A Heat pumps Dehumidiers PV solar collectors Pellet boiler Total Plant B Heat pumps Cold tank Boreholes BHE pipes Dehumidiers PV solar collectors Pellet boiler Total Plant C Gas boiler Air to air heat pumps Total 8.5 MWh 38.0 MWh 46.5 MWh 18.6 MWh 8.4 MWh 40.1 MWh 392.7 MWh 3.4 MWh 435.2 MWh 6.8 MWh 905.2 MWh 18.6 MWh 3.4 MWh 605.2 MWh 6.8 MWh 634.0 MWh

MWh

years

20

Fig. 12. Total (construction + operation) exergy use versus time, for Plants A, B, C, in the absence of PV collectors.

The exergy analysis illustrated in Fig. 10 does not yield a direct comparison between the exergy use of an air-to-water heat pump system and that of a ground-coupled heat pump system, because Plant A and Plant B have different PV collector areas. To obtain a direct comparison, the exergy analysis has been repeated by excluding the embodied energy and the annual exergy production of the PV system. Clearly, the data for Plant C do not change. The total embodied energy becomes 28.8 MWh for Plant A and 470 MWh Plant B. The annual exergy use for Plant A is given by the sum of 24.56 MWh, due to the consumption of wood pellet, and of the primary energy equivalent of the electric energy use per year, namely 72.79 2.175 = 158.32 MWh; the total is 182.88 MWh. Similarly, for plant B one obtains a total exergy use per year equal to 24.56 + 52.25 2.175 = 138.20 MWh. Plots of the total exergy loss due to the plant construction and operation versus time, for a period of 20 years, in this scenario, are illustrated in Fig. 12. The gure shows that, in the absence of the PV system, the lowest exergy use after 20 years is obtained by Plant B. Therefore, the exergy analysis reveals that ground-coupled heat pump systems yield the lowest consumption of primary energy sources, even in a ground with a rather low thermal conductivity (kgd = 1.70 W/(m K)), as in the case considered here. 7. Conclusions Two alternative zero carbon plants for heating, cooling, humidity control and domestic hot water supply, for a new building complex in Northern Italy, have been studied by means of the simulation code TRNSYS and compared with a conventional plant. Both plants employ heat pumps which receive electricity by PV panels and thermal solar collectors for DHW supply. Plant A employs air-

T. Terlizzese, E. Zanchini / Energy and Buildings 43 (2011) 787795

795

to-water heat pumps, whereas Plant B employs ground-coupled heat pumps; they have the same, nearly vanishing, primary energy use (wood pellet) and different PV collector areas. The economic analysis has shown that both Plant A and Plant B are feasible, and that Plant A has a lower nancial payback time (6 years) than Plant B (11 years). The exergy analysis has shown that Plant A yields also a lower total exergy consumption after 20 years of operation. However, this result is due to the higher PV collector area employed in Plant A. If the exergy analysis is repeated without considering the PV panels, then the lowest exergy consumption after 20 years is obtained by Plant B. The results point out that ground-coupled heat pumps ensure a lower environmental impact than air-to-water heat pumps, but are economically less feasible, at least in a ground with a low or medium thermal conductivity. Therefore, a specic nancial support for the installation of ground-coupled heat pumps should be given by public administrations. References
[1] K. Peippo, P.D. Lund, E. Vartiainen, Multivariate optimization of design trade-offs for solar low energy buildings, Energy and Buildings 29 (1999) 189205. [2] C.A. Balaras, K. Droutsa, A.A. Argiriou, D.N. Asimakopoulos, Potential for energy conservation in apartment buildings, Energy and Buildings 31 (2000) 143154. [3] M.T. Iqbal, A feasibility study of a zero energy home in Newfoundland, Renewable Energy 29 (2004) 277289. [4] D.O. Rijksen, C.J. Wisse, A.W.M van Schijndel, Reducing peak requirements for cooling by using thermally activated building systems, Energy and Buildings 42 (2010) 298304.

[5] X. Zhao, J.M. Lee, S.B. Riffat, Numerical study of a novel counter-ow heat and mass exchanger for dew point evaporative cooling, Applied Thermal Engineering 28 (2008) 19421951. [6] X. Zhao, S. Yang, Z. Duan, S.B. Riffat, Feasibility study of a novel dew point air conditioning system for China building application, Building and Environment 44 (2009) 19901999. [7] H.Y. Chan, S.B. Riffat, J. Zhu, Review of passive solar heating and cooling technologies, Renewable and Sustainable Energy Reviews 14 (2010) 781789. [8] L. Wang, J. Gwilliam, P. Jones, Case study of zero energy house design in UK, Energy and Buildings 41 (2009) 12151222. [9] S.A. Klein, W.A Beckman, J.A. Dufe, A design procedure for solar heating systems, Solar Energy 18 (1976) 113127. [10] ASHRAE Handbook HVAC Applications, Ch. 32. [11] J. Remund, S. Kunz, Meteonorm Version 5, METEOTEST, http://www. meteotest.com. [12] M. Marcic, Long-term performance of central heat pumps in Slovenian homes, Energy and Buildings 36 (2004) 185193. [13] M. Forsn, Heat pumps technology and environmental impact, Report of the Swedish Heat pump Association and of the European Heat pump Association, 2005, http://ec.europa.eu/environment/ ecolabel/about ecolabel/reports/hp tech env impact aug2005.pdf. [14] S.P. Kavanaugh, K. Rafferty, Ground-source heat pumps: design of geothermal systems for commercial and institutional buildings, ASHRAE (1997). [15] Joint Research Centre, Institute for Energy, http://re.jrc.ec.europa.eu/ pvgis/apps3/pvest.php. [16] G. Cammarata, L. Marletta, Embodied energy versus energy efciency in building heating systems, Proceedings of CLIMA 2000/Napoli 2001 World Congress, 1, 1518. (Naples, 2001). [17] G.P. Hammond, C.I. Jones, Embodied energy and carbon in construction materials, Proceedings of Institution of Civil Engineers, Energy 161 (2) (2008) 8798. [18] J.K. Kaldellis, D. Zarakis, E. Kondili, Optimum autonomous stand-alone photovoltaic system design on the basis of energy pay-back analysis, Energy 34 (2009) 11871198. [19] UNI 10389-1, Heat generators. Flue gases analysis and measurement on site of combustion efciency, September 2009. [20] EN 590, Automotive fuels Diesel Requirements and test methods, October 1st 2009.