- Email: [email protected]
Thermal charging of boreholes
Renewable Energy 67 (2014) 165e172
Contents lists available at ScienceDirect
Renewable Energy journal homepage: www.elsevier.com/locate/renene
Thermal charging of boreholes Tshewang Lhendup, Lu Aye*, Robert James Fuller Renewable Energy and Energy Efficiency Group, Department of Infrastructure Engineering, Melbourne School of Engineering, The University of Melbourne, Victoria 3010, Australia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 October 2013 Accepted 14 November 2013 Available online 8 December 2013
This paper presents experimental study of thermal charging the boreholes that are used for interseasonal thermal storage of heat and coolth integrated with ground-coupled heat pump and unglazed solar collectors. After 180 days of thermal charging, it was observed that the temperature of the ground at 21 m depth and 1 m distance from the borehole had increased by 2.5 C. The unglazed collectors are able to collect heat for charging the heat storage borehole at an average of 43.9 MJ day1. The system is able to charge borehole at an average heat transfer rate of 57 W m1. A comparison of the experimental results with the simulated results of a TRNSYS model of the system showed a good agreement. The mean efficiency of the unglazed solar collector during 180 days charging operation was found to be 30% and the mean efficiency of the system was found to be 38%. Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved.
Keywords: Thermal charging Unglazed solar collector Inter-seasonal thermal storage
1. Introduction Globally 40% of the total energy consumption is by buildings [1]. In Australia, household energy consumption accounted for 12% of the total energy end use during the year 2010e2011 [2], out of which more than 35% was used for space heating [3]. The demand for the space heating and cooling energy is increasing due to increasing number of buildings associated with increase in population. This demand is expected to increase over the years. Therefore there is a need to look for alternative technologies which have lower energy consumption for space heating and cooling. While renewable energy is an alternative technology which is environmentally friendly, the problem is its intermittent availability. Most of the renewable energy sources are not available continuously. Building heating and cooling loads vary continuously. Moreover the peak loads and the availability of the renewable energy sources may not match. Therefore storage of energy is necessary to enable the system to match the demand. Inter-seasonal thermal storage integrated with a heat pump and solar collector has been gaining attention in the past [4e10]. Inter-seasonal thermal storage is defined as the storing of heat/coolth from one season to another season. The accumulated heat/coolth is extracted by the heat pump for space heating in winter and cooling in summer respectively. The heat/coolth charging is expected to enhance the performance of the heat pump. Several experimental and theoretical studies have been
* Corresponding author. E-mail address: [email protected] (L. Aye).
conducted in the past regarding the performance of thermal storage with heat pump and solar collectors. Nordell and Hellstrom [11] found that 60% of the total heat demand (both for space heating and domestic hot water) of 9000 m2 building floor area can be supplied by 3000 m2 glazed flat plate solar collectors mounted on roof and 60 000 m3 of borehole storage system. The mean temperature of the borehole storage varied from 30 to 45 C and a heat pump is not required to upgrade the heat delivered. The storage loss was estimated to be about 40% of the total energy charged by the solar collectors. Chiasson and Yavuzturk [12] studied the performance of hybrid geothermal heat pump systems with solar thermal collectors in six cities in USA. They applied the same school building parameters for all locations but different climatic data sets. From their findings the borehole length could be reduced from 16% to 33% when a fixed solar collector was used compared to a system without solar thermal collectors. Trillat-Berdal et al. [13] performed simulation using a validated model to study the performance of a glazed solar assisted ground-coupled heat pump (SAGCHP) consisting of two boreholes of 90 m deep and 12 m2 roof top glazed solar thermal collectors for space heating and domestic hot water production for a residential house (180 m2 floor area). They concluded that the proposed system could help to reduce the borehole length and initial cost of the system. Wang and Qi [14] studied the performance of borehole thermal storage and GCHP system for the heating and cooling of residential building. The system comprised of four boreholes of 50 m deep and 220 mm diameter and 25 m2 glazed flat plate collector for a 120 m2 floor area residential house. They reported that out of 42 GJ of useful energy delivered at the collector outlet, only 30 GJ (71%) could be charged to the boreholes and the rest was lost.
0960-1481/$ e see front matter Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.renene.2013.11.045
166
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
Xi et al. [15] simulated a SAGCHP system for space heating and hot water supply over 20 years for Beijing climatic conditions. The system consisted of 270 m borehole, 45 m2 glazed flat plate collector and 1.8 m3 water storage tank. They concluded that the SAGCHP system coefficient of performance (SCOP) is 26% higher than the SCOP of conventional GCHP without solar collector and electricity consumption for heating could be reduced by 1.6 MWh annually. Moreover, SAGCHP could save borehole length up to 3.6 m per m2 of solar collector area compared to the conventional GCHP system. Wang et al. [16] investigated the relative performance of a conventional GCHP and a SAGCHP. The conventional GCHP comprised of 66 boreholes of 120 m deep and 4 m apart while the SAGCHP comprised of 25 boreholes of 50 deep and 2.5 m apart and 280 m2 flat plate collector. The load for the system was a four storey building having a floor area of 4953.4 m2. The study was based on a multi-year simulation using TRNSYS. They found the simulated system SCOP varied from 3.42 to 3.17 and 2.99 to 2.95 for SAGCHP and conventional GCHP respectively at the end of 25 years of operation. A comparative study on performance of SAGCHP in three cities in Canada (Edmonton, Montreal and Vancouver) by Eslami-Nejad and Bernier [17] found that by charging solar energy collected into the boreholes using double U-tubes, the design borehole length could be reduced in the respective locations by 13%, 12% and 18% with a corresponding reduction of annual heat pump electricity consumption by 6%, 5.3% and 6.3%. Mempouo [8] also found that borehole length could be reduced up to 60% by using a SAGCHP system compared to system without solar collectors. A recent study by Rad et al. [4] found that by charging heat from 6.81 m2 solar collector into four boreholes of 55 m deep each, the borehole length could be reduced by 15% compared to the system without thermal charging. The system was used for heating a two-storey detached residential house having 498 m2 floor area with design heating and cooling demand of 11.5 kW and 9.5 kW respectively. The presented past studies indicated that the solar thermal charging of the boreholes provides two potential benefits, improvement of COP of the system or reduction in design borehole length compared to the system without solar collectors with same or slight increase in COP. The above studies are limited to heating application only, whereas in most parts of Australia, there is demand for both heating and cooling. Therefore an inter-seasonal heat and coolth storage system is being investigated at the University of Melbourne. The system consists of a separate heat and coolth storage boreholes integrated with a GCHP and unglazed solar collectors. The unglazed solar collectors allow charging of both heat and coolth. The heat collected during the day is charged into the heat storage borehole (HSB). During the night the heat extracted from the coolth storage borehole (CSB) is dissipated from the unglazed solar collector surface to the ambient. The aim of this paper is to present the potential of heat charging using unglazed solar collector conducted experimentally on a small scale set up in Melbourne. The coolth charging experiment was presented in Lhendup et al. [18]. 2. Experimental set up The experimental set up consists of two 40 m deep boreholes, two 3.84 m2 unglazed solar collectors connected in parallel, two circulating pumps, piping networks and six 21 m deep monitoring boreholes. One 40 m deep borehole is used as HSB and another 40 m deep borehole as a CSB. Each borehole has two U-tubes which enable independent charging and discharging operations. For detailed descriptions of the experimental set up, refer to Lhendup et al. [18]. The design parameters used in this study are summarised in Table 1. Fig. 1 shows the schematic diagram of the experimental set up. Water pump PP1 is switched ON when (TcoeTgd_HSB > 1) and
Table 1 Design parameters used in the study. Parameter
Value
Borehole depth (m) Distance between the boreholes (m) Diameter of borehole (mm) No of U-tube (HDPE pipe) U-tube pipe inside diameter (mm) U-tube pipe outside diameter (mm) Total solar collector area (m2) Fluid flow rate (kg h1) Ground thermal conductivity (W m1 K1) Grout thermal conductivity (W m1 K1) Thermal conductivity of pipe (W m1 K1)
40 8 115 2 21.32 25 7.68 840 2.23 1.2 0.4
switched OFF otherwise. Tco is the collector outlet temperature ( C) and Tgd_HSB is the ground temperature of HSB at 21 m deep ( C). When pump PP1 is ON, valves V1, V4, V5, V17, V18, V20 are open and all other valves are close. The uncertainties of the measurements are summarised in Table 2. 3. TRNSYS model The experimental set up for heat charging was modelled in TRNSYS-17 [19]. Fig. 2 shows the components used in graphical user interface of the TRNSYS model. HSB is represented by Type 257, a modified version of thermal energy system specialists (TESS) Type 557 [20]. The unglazed solar collector is represented by a modified version of TESS Type 559. The other components such as circulating pump, buffer tank and pipes were used from the standard TRNSYS library. The heat charging experiment was conducted from October 2012 to March 2013. During the experiment, temperatures of the water at the inlets and outlets of the heat storage borehole, solar collectors and the buffer tank were measured (Fig. 1). These measurements enabled validation of the TRNSYS Types for HSB, solar collectors and buffer tank. Although the experiment was conducted and TRNSYS types applied were validated for a period of 180 days, only a week data (17e23 December 2012) was presented in this paper. The validated model will later be used to simulate the performance of the entire system which is not discussed in this paper. 4. Results and discussion Fig. 3 shows the measured short wave- solar radiation on the collector plane and ambient air temperature on site during the period 17e23 December 2012. The undisturbed ground temperature of the HSB was measured to be 17.3 C [18]. The ground temperature near to the surface fluctuates depending on the thermal properties of the ground and the conditions at the surface. This fluctuation disappears below a certain depth and becomes stable. Fig. 4 shows the monthly average ground temperature at 2, 21 and 40 m along with the monthly average ambient air temperature. As evident from the measurements the temperature at 2 m below the ground surface fluctuates with the monthly average ambient air temperature whereas the temperature at 21 m and 40 m depths are nearly constant throughout the year. Below 21 m deep, the ground temperature is higher than the ambient in winter and lower than ambient in summer. This feature of the ground temperature is the reason behind the use of GCHP systems. 4.1. Temperature Each section of the model has input and outlet temperature measured which enable to compare with the simulated values. The
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
167
Fig. 1. Schematic diagram of the system.
comparison was done by using three statistical parameters namely root mean squared error (RMSE) and mean bias error (MBE). RMSE gives overall assessment of the difference between the measured and simulated variables, while MBE shows the bias of the difference [21]. RMSE is defined by Equation (1) and MBE by Equation (2) [21].
RMSE ¼
MBE ¼
X
ðYi Xi Þ2
. 0:5 n
(1)
. X ðYi Xi Þ n
(2)
where Xi is the ith measured value, Yi is the ith simulated value and n is the number of observations. The RMSE and MBE for the heat charging period is shown in Table 3.
Table 2 Uncertainties of measured and calculated parameters. Absolute Instrument/measured variable Power transducer 10 W Flow meter 33.6 Kg h1 Thermistor 0.002 C Pyranometer 10 W m2 Wind speed sensor 1.1 m s1 Temperature (ambient) 0.21 C Calculated Parameter Temperature 0.12 C Heat transfer rate 163 W
Relative 0.2% 2% 5% 4%
0.2% 2.8%
Remarks e e For 1 to 60 C range Whichever is higher Whichever is higher Between 0 and 50 C e e
Figs. 5e7 shows the plot of hourly average simulated and measured inlet and outlet temperatures. As evident from the comparisons, there is a good agreement between the TRNSYS model simulations and the measured data. There is a small but irregular under and over estimation of the both the inlet and outlet temperatures by the TRNSYS model. The discrepancy is more in solar collector outlet temperature as indicated in Fig. 5(b). The plausible reason for this could be the solar collector convective heat loss coefficient in TRNSYS model which was computed from the measured data and correlated with the wind speed. The convective heat loss coefficient derived from the measured data was found to be substantially varied with time. The heat losses from piping network between the components (solar collector, HSB and buffer tank) also affect the inlet and outlet temperatures. The heat losses from the pipes were calculated based on the pipe heat loss coefficients which were not measured but estimated from the thermal properties of the pipes and insulations. Similarly there is both under and over estimation of the HSB inlet and outlet temperatures. The HSB outlet temperature is directly proportional to the heat transfer from the fluid to the ground which in turn mainly depends on the effective thermal conductivity of the ground and the ground temperature. A high effective thermal conductivity results in higher heat transfer rate to the ground with lower corresponding borehole outlet temperature and vice versa. Thus the discrepancy exists between the measured and TRNSYS simulated HSB outlet temperature. In case of buffer tank, there is less discrepancy between the simulated and measured inlet and outlet temperatures as shown in Fig. 7. From the above comparisons, it can be concluded that in general
168
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
Fig. 2. Schematic of TRNSYS model for HSB charging.
system in heat charging mode. The energy charged is more than the energy absorbed by the solar collector due to the energy input from the pump. The pump operated in the heat charging mode even when there was no solar radiation. The source of heat in this case was from the water in the buffer tank which also behaves as a temporary thermal storage. Due to the low thermal conductivity of the ground, not all heat could be charged immediately into the ground. Therefore the remaining heat was being stored in the buffer tank temporarily. 4.3. Borehole thermal performance Fig. 3. Short wave solar radiation on collector plane and ambient air temperature (17e 23 Dec 2012).
there is a good agreement between the TRNSYS model simulations and the measured data. Thus it confirms the ability of the modified unglazed solar collector (Type 559), modified HSB model (Type 257) and the buffer tank model (Type 39) to predict their respective outlet temperatures consistently for the stated parameters. 4.2. Energy balance After 180 days of operation, a total of 11.6 GJ of energy was charged into the borehole at an average of 2.7 MJ h1. On the other hand, the solar collector operated for 1204 h and gained 7.9 GJ at an average rate of 6.6 MJ h1. Fig. 8 shows the energy balance of the
Fig. 9 shows the average daily measured and simulated heat charged into the HSB. The TRNSYS simulation underestimated the heat charging capacity of the system. However the deviation is less than 10% most of the days. So, it can be inferred that the TRNSYS model is able to predict the heat transfer from the fluid in the ground heat exchanger to the ground. Fig. 10 shows the daily average heat transfer rate at HSB while the pump is in operation. The maximum heat transfer rate is 86 W m1 with an average of 57 W m1. Due to the continuous heat charging into the borehole, the heat diffuses from the borehole to the surrounding ground. This is indicated by rise in ground temperatures at the monitoring boreholes located at 1, 2 and 3 m from the centre of HSB. Fig. 11 shows the average daily temperature of the ground at 21 m deep near the HSB rises gradually. The HSB appears to be working as indicated by the change in the surrounding ground temperatures. Whether the system is able to retain the heat is yet to be monitored. However, with the Table 3 Statistical parameters of the TRNSYS types used in the model compare to measured values. Model validated Piping from buffer tank to solar collector Solar collectors (Type 559)
Fig. 4. Average monthly ambient air and ground temperatures.
Result
Collector inlet temperature Collector outlet temperature Piping from solar collector to HSB HSB inlet temperature HSB (Type 257) HSB outlet temperature Piping from HSB to buffer tank Tank inlet temperature Buffer tank (Type 39) Tank outlet temperature
RMSE ( C)
MBE ( C)
1.47
0.25
0.76
0.03
2.65 0.67 1.69 0.62
1.37 0.14 0.86 0.05
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
Fig. 5. Measured and simulated (a) inlet and (b) outlet temperature of the unglazed solar collector during the period 17e23 December 2012.
heat charging period ending in April and heat extraction starting in May, the heat will begin to flow in the reverse direction, i.e. from far field towards the boreholes. This is because as the system begins to extract heat from the ground, the temperature around the borehole will decrease and hence result in reverse direction of heat flow. 4.4. Collector performance The useful heat output of an unglazed solar collector depends on the solar radiation received, ambient air temperature, the long wave radiation exchange, and wind speed. These four environmental factors were measured during the experiment. It should be noted that the pump (PP1) was not running when it was raining. So the effects of rain on the collector performance can be eliminated. Fig.12
169
Fig. 6. Measured and simulated (a) inlet and (b) outlet temperature of the borehole during the period 17e23 December 2012.
shows a comparison of the average daily measured and simulated useful heat output for 180 days. There are discrepancies between the measured and simulated heat gain. The reasons for these discrepancies can be attributed to two factors. The first one being the discrepancies in the collector inlet and outlet temperatures as discussed in Section 4.1. The second factor is the uncertainties in the estimation of heat loss coefficient of the solar collector in the model. The total useful energy gained and efficiency are important performance parameters of the solar collector. The instantaneous efficiency of the collector can be expressed as:
hcol ¼ Qu_col =Ac G
(3)
Qu_col ¼ FR Ac ½ðsaÞG UL ðTci Tamb Þ
(4)
where Ac is the collector area (m2), G is the total irradiation on the surface of the collector (W m2), FR is the collector heat removal
170
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
Fig. 8. Energy balance of the system during heat charging.
heat charging experiment. A linear correlation obtained by robust fit of the measured data shows collector efficiency can be represented as 0.51e8.33Cpf. The efficiency is higher at lower performance coefficient. It should be noted that the affect of wind on the solar collector heat gain was not taken into account. The wind speed was found to vary from 0.5 m s1 to 8.5 m s1 with an average of 2.5 m s1. Fig. 14 shows the hourly average collector efficiency during a day. The collector efficiency peaks around noon and gradually decreases. 4.5. System efficiency The efficiency of the system in heat charging mode was calculated based on the total energy that was charged into the ground and the total solar energy available at the collector plus the electricity input to the circulating pump (Equation (6)). This efficiency is different from the efficiency of the solar collector. The system efficiency is an important parameter to evaluate the performance of
Fig. 7. Measured and simulated (a) inlet and (b) outlet temperature of the buffer tank during the period 17e23 December 2012.
factor, s is the transmissivity of collector cover (for unglazed collector s ¼ 1), a is the absorptance of collector (), UL is the total collector heat loss coefficient (W m2 K1), Tci is the collector inlet temperature (K) and Tamb is the ambient temperature (K). Substituting Equation (4) in (3), the instantaneous collector efficiency can be expressed as:
hcol ¼ FR ½a ðUL ðTci Tamb Þ=GÞ ¼ FR a FR UL Cpf
(5)
Collector performance coefficient, Cpf ¼ (TciTamb)/G. As FR, UL and a are assumed to be constant, instantaneous collector efficiency is indirectly proportional to (TciTamb)/G. The efficiency of the unglazed collectors varied between 0.21 and 0.55 with a mean value of 0.31. Fig. 13 shows the collector efficiency verses the collector performance coefficient for the duration of the
Simulated heat chaeged (W)
1800
1200
600
0 0
600
1200
Measure heat charged (W) Fig. 9. Average daily heat charged for 180 days.
1800
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
171
Rate of heat transfer (W m-1)
100
75
50
25
0 2-10-12 1-11-12 1-12-12 31-12-12 30-1-13
1-3-13
31-3-13
Fig. 10. HSB heat transfer rate.
Fig. 13. Variation of hourly average collector efficiency with performance coefficient during the heat charging period.
Fig. 11. Ground temperature recorded in the monitoring boreholes at 21 m depth.
the system in heat charging mode. This efficiency accounts for all energy inputs, outputs and losses of the system.
hhs ¼ Qchg
. Qsa þ Qpi
(6)
where Qchg is the total energy charged into the boreholes (MJ), Qsa is the amount of solar radiation available at the collector surface (MJ) and Qpi is the pump electricity input (MJ). Fig. 15 shows the average
daily system efficiency. The system in this study is able to achieve a maximum daily efficiency of 72% with an average of 38%. 5. Conclusions The application of an unglazed solar collector for heat charging of boreholes has been presented. The unglazed solar collectors are able to collect heat for charging the storage borehole in the ground at an average of 5.7 MJ day1 m2 during the test period. The average heat transfer rate of borehole was found to be 57 W m1. After 180 days of operation, the average temperatures of the ground surrounding the HSB have increased by 2.5 C. Thus there is a potential for such system to be used for heat charging of the
1000
800
600 0.8
System efficiency (-)
Simulated heat gain (W)
Fig. 14. Average hourly solar collector efficiency on 18 December 2012.
400
200
0 0
200
400
600
Measure heat gain (W) Fig. 12. Average daily heat gain for 180 days.
800
1000
0.6 0.4 0.2 0.0 3-10-12
3-11-12
3-12-12
3-1-13
3-2-13
3-3-13
Fig. 15. Average daily system efficiency during heat charging.
3-4-13
172
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
boreholes. A comparison of the TRNSYS simulations with the measured data showed a good agreement confirming the ability of the TRNSYS model to predict their respective outlet temperatures. The heat charging performance strongly depends on the solar radiation available at the surface of the solar collector. The mean unglazed solar collector efficiency is 30%. The system is able to achieve mean efficiency of 38%. The heating and cooling experiments have been conducted and the results will be reported elsewhere. Acknowledgements We would like to thank Dr Simon Chapuis and Professor Michel Bernier, Department of Mechanical Engineering, Ecole Polytechnique de Montreal, Canada for providing the source code of the modified version of TESS Type 557 for simulations of two U-tubes boreholes. Tshewang Lhendup would also like to thank AusAID for providing a scholarship for the study. Nomenclature Symbols Ac Cpf FR G n Q T UL Xi Yi
collector area (m2) collector performance factor (K m2 W1) collector heat removal factor () total irradiation on the surface of the collector (W m2) number of data points () energy (MJ) temperature (K) total collector heat loss coefficient (W m2 K1) ith measured value ith simulated value
Greek symbols a absorptivity of collector () s transmissivity of collector cover () Subscripts amb ambient chg charged ci collector inlet co collector outlet gd ground pi pump input sa solar radiation at the collector surface Abbreviations SCOP system coefficient of performance CSB coolth storage borehole GCHP ground-coupled heat pump HSB heat storage borehole
MBE RMSE
mean bias error root mean square error
References [1] IEA. Energy efficiency requirements in building codes, energy efficiency policies for new buildings. Paris, France: International Energy Agency; 2008. [2] Stark C, Penney K, Feng A. 2012 Australian energy update. Canberra: Bureau of Resources and Energy Economics; 2012. [3] Che N, Pham P. Economic analysis of end-use energy intensity in Australia. Canberra: Bureau of Resources and Energy Economics; 2012. [4] Rad FM, Fung AS, Leong WH. Feasibility of combined solar thermal and ground-source heat pump systems in cold climate, Canada. Energy Build 2013;61:224e32. [5] Cauret O, Kummert M. Potential of solar assisted ground-source heat pumps with unglazed solar collectors for French office buildings. Fed Eur Heat Vent Air Cond Assoc (REHVA) J August 2011:16e21. [6] Stojanovic B, Akander J. Build-up and long-term performance test of a fullscale solar-assisted heat pump system for residential heating in Nordic climatic conditions. Appl Therm Eng 2010;30:188e95. [7] Bakirci K, Ozyurt O, Comakli K, Comakli O. Energy analysis of a solar-ground source heat pump system with vertical closed-loop for heating applications. Energy 2011;36:3224e32. [8] Mempouo B. Investigations of novel heat pump systems for low carbon homes. United Kingdom: Institute of Sustainable Energy Technology and Institute of Building Technology, Department of Architecture and The Built Environment, The Faculty of Engineering, The university of Nottingham; 2011. p. 308. [9] Eslami-Nejad P, Bernier M. Coupling of geothermal heat pumps with thermal solar collectors using double U-tube boreholes with two independent circuits. Appl Therm Eng 2011;31:3066e77. [10] Han Z, Zheng M, Kong F, Wang F, Li Z, Bai T. Numerical simulation of solar assisted ground-source heat pump heating system with latent heat energy storage in severely cold area. Appl Therm Eng 2008;28:1427e36. [11] Nordell B, Hellström G. High temperature solar heated seasonal storage system for low temperature heating of buildings. Sol Energy 2000;69:511e23. [12] Chiasson AD, Yavuzturk C. Assessment of the viability of hybrid geothermal heat pump systems with solar thermal collectors. ASHRAE Trans 2003;109: 487e500. [13] Trillat-Berdal V, Souyri B, Achard G. Coupling of geothermal heat pumps with thermal solar collectors. Appl Therm Eng Heat Trans Sustain Energy Technol 2007;27:1750e5. [14] Wang H, Qi C. Performance study of underground thermal storage in a solarground coupled heat pump system for residential buildings. Energy Build 2008;40:1278e86. [15] Xi C, Lin L, Hongxing Y. Long term operation of a solar assisted ground coupled heat pump system for space heating and domestic hot water. Energy Build 2011;43:1835e44. [16] Wang E, Fung AS, Qi C, Leong WH. Performance prediction of a hybrid solar ground-source heat pump system. Energy Build 2012;47:600e11. [17] Eslami-Nejad P, Bernier M. Simulations of a new double U-tube borehole configuration with solar heat injection and ground freezing. In: eSim 2012: the Canadian conference on building simulation, Halifax, Nova Scotia 2012. p. 257e68. [18] Lhendup T, Aye L, Fuller RJ. Experimental study of coolth charging of an interseasonal underground thermal storage system. In: Solar 2012, 50th annual AuSES conference. Hawthorn Campus, Melbourne, Australia: Swinburne University of Technology; 2012. [19] Klein SA, Beckman WA, Mitchell JW, Duffe JA, Duffe NA, Freeman TL, et al. TRNSYS 17: a transient system simulation program. WI: University of Wisconsin-Madison; 2010. [20] Chapuis S. Stockage Thermique Saisonnier Dans Un Champ De Puits Geothermiques Verticaux En Boucle Fermee. Montreal: Department of Mechanical Engineering Polytechnic School of Montreal; 2009. [21] Iqbal M. An introduction to solar radiation. Canada: Academic Press; 1983.
Contents lists available at ScienceDirect
Renewable Energy journal homepage: www.elsevier.com/locate/renene
Thermal charging of boreholes Tshewang Lhendup, Lu Aye*, Robert James Fuller Renewable Energy and Energy Efficiency Group, Department of Infrastructure Engineering, Melbourne School of Engineering, The University of Melbourne, Victoria 3010, Australia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 October 2013 Accepted 14 November 2013 Available online 8 December 2013
This paper presents experimental study of thermal charging the boreholes that are used for interseasonal thermal storage of heat and coolth integrated with ground-coupled heat pump and unglazed solar collectors. After 180 days of thermal charging, it was observed that the temperature of the ground at 21 m depth and 1 m distance from the borehole had increased by 2.5 C. The unglazed collectors are able to collect heat for charging the heat storage borehole at an average of 43.9 MJ day1. The system is able to charge borehole at an average heat transfer rate of 57 W m1. A comparison of the experimental results with the simulated results of a TRNSYS model of the system showed a good agreement. The mean efficiency of the unglazed solar collector during 180 days charging operation was found to be 30% and the mean efficiency of the system was found to be 38%. Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved.
Keywords: Thermal charging Unglazed solar collector Inter-seasonal thermal storage
1. Introduction Globally 40% of the total energy consumption is by buildings [1]. In Australia, household energy consumption accounted for 12% of the total energy end use during the year 2010e2011 [2], out of which more than 35% was used for space heating [3]. The demand for the space heating and cooling energy is increasing due to increasing number of buildings associated with increase in population. This demand is expected to increase over the years. Therefore there is a need to look for alternative technologies which have lower energy consumption for space heating and cooling. While renewable energy is an alternative technology which is environmentally friendly, the problem is its intermittent availability. Most of the renewable energy sources are not available continuously. Building heating and cooling loads vary continuously. Moreover the peak loads and the availability of the renewable energy sources may not match. Therefore storage of energy is necessary to enable the system to match the demand. Inter-seasonal thermal storage integrated with a heat pump and solar collector has been gaining attention in the past [4e10]. Inter-seasonal thermal storage is defined as the storing of heat/coolth from one season to another season. The accumulated heat/coolth is extracted by the heat pump for space heating in winter and cooling in summer respectively. The heat/coolth charging is expected to enhance the performance of the heat pump. Several experimental and theoretical studies have been
* Corresponding author. E-mail address: [email protected] (L. Aye).
conducted in the past regarding the performance of thermal storage with heat pump and solar collectors. Nordell and Hellstrom [11] found that 60% of the total heat demand (both for space heating and domestic hot water) of 9000 m2 building floor area can be supplied by 3000 m2 glazed flat plate solar collectors mounted on roof and 60 000 m3 of borehole storage system. The mean temperature of the borehole storage varied from 30 to 45 C and a heat pump is not required to upgrade the heat delivered. The storage loss was estimated to be about 40% of the total energy charged by the solar collectors. Chiasson and Yavuzturk [12] studied the performance of hybrid geothermal heat pump systems with solar thermal collectors in six cities in USA. They applied the same school building parameters for all locations but different climatic data sets. From their findings the borehole length could be reduced from 16% to 33% when a fixed solar collector was used compared to a system without solar thermal collectors. Trillat-Berdal et al. [13] performed simulation using a validated model to study the performance of a glazed solar assisted ground-coupled heat pump (SAGCHP) consisting of two boreholes of 90 m deep and 12 m2 roof top glazed solar thermal collectors for space heating and domestic hot water production for a residential house (180 m2 floor area). They concluded that the proposed system could help to reduce the borehole length and initial cost of the system. Wang and Qi [14] studied the performance of borehole thermal storage and GCHP system for the heating and cooling of residential building. The system comprised of four boreholes of 50 m deep and 220 mm diameter and 25 m2 glazed flat plate collector for a 120 m2 floor area residential house. They reported that out of 42 GJ of useful energy delivered at the collector outlet, only 30 GJ (71%) could be charged to the boreholes and the rest was lost.
0960-1481/$ e see front matter Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.renene.2013.11.045
166
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
Xi et al. [15] simulated a SAGCHP system for space heating and hot water supply over 20 years for Beijing climatic conditions. The system consisted of 270 m borehole, 45 m2 glazed flat plate collector and 1.8 m3 water storage tank. They concluded that the SAGCHP system coefficient of performance (SCOP) is 26% higher than the SCOP of conventional GCHP without solar collector and electricity consumption for heating could be reduced by 1.6 MWh annually. Moreover, SAGCHP could save borehole length up to 3.6 m per m2 of solar collector area compared to the conventional GCHP system. Wang et al. [16] investigated the relative performance of a conventional GCHP and a SAGCHP. The conventional GCHP comprised of 66 boreholes of 120 m deep and 4 m apart while the SAGCHP comprised of 25 boreholes of 50 deep and 2.5 m apart and 280 m2 flat plate collector. The load for the system was a four storey building having a floor area of 4953.4 m2. The study was based on a multi-year simulation using TRNSYS. They found the simulated system SCOP varied from 3.42 to 3.17 and 2.99 to 2.95 for SAGCHP and conventional GCHP respectively at the end of 25 years of operation. A comparative study on performance of SAGCHP in three cities in Canada (Edmonton, Montreal and Vancouver) by Eslami-Nejad and Bernier [17] found that by charging solar energy collected into the boreholes using double U-tubes, the design borehole length could be reduced in the respective locations by 13%, 12% and 18% with a corresponding reduction of annual heat pump electricity consumption by 6%, 5.3% and 6.3%. Mempouo [8] also found that borehole length could be reduced up to 60% by using a SAGCHP system compared to system without solar collectors. A recent study by Rad et al. [4] found that by charging heat from 6.81 m2 solar collector into four boreholes of 55 m deep each, the borehole length could be reduced by 15% compared to the system without thermal charging. The system was used for heating a two-storey detached residential house having 498 m2 floor area with design heating and cooling demand of 11.5 kW and 9.5 kW respectively. The presented past studies indicated that the solar thermal charging of the boreholes provides two potential benefits, improvement of COP of the system or reduction in design borehole length compared to the system without solar collectors with same or slight increase in COP. The above studies are limited to heating application only, whereas in most parts of Australia, there is demand for both heating and cooling. Therefore an inter-seasonal heat and coolth storage system is being investigated at the University of Melbourne. The system consists of a separate heat and coolth storage boreholes integrated with a GCHP and unglazed solar collectors. The unglazed solar collectors allow charging of both heat and coolth. The heat collected during the day is charged into the heat storage borehole (HSB). During the night the heat extracted from the coolth storage borehole (CSB) is dissipated from the unglazed solar collector surface to the ambient. The aim of this paper is to present the potential of heat charging using unglazed solar collector conducted experimentally on a small scale set up in Melbourne. The coolth charging experiment was presented in Lhendup et al. [18]. 2. Experimental set up The experimental set up consists of two 40 m deep boreholes, two 3.84 m2 unglazed solar collectors connected in parallel, two circulating pumps, piping networks and six 21 m deep monitoring boreholes. One 40 m deep borehole is used as HSB and another 40 m deep borehole as a CSB. Each borehole has two U-tubes which enable independent charging and discharging operations. For detailed descriptions of the experimental set up, refer to Lhendup et al. [18]. The design parameters used in this study are summarised in Table 1. Fig. 1 shows the schematic diagram of the experimental set up. Water pump PP1 is switched ON when (TcoeTgd_HSB > 1) and
Table 1 Design parameters used in the study. Parameter
Value
Borehole depth (m) Distance between the boreholes (m) Diameter of borehole (mm) No of U-tube (HDPE pipe) U-tube pipe inside diameter (mm) U-tube pipe outside diameter (mm) Total solar collector area (m2) Fluid flow rate (kg h1) Ground thermal conductivity (W m1 K1) Grout thermal conductivity (W m1 K1) Thermal conductivity of pipe (W m1 K1)
40 8 115 2 21.32 25 7.68 840 2.23 1.2 0.4
switched OFF otherwise. Tco is the collector outlet temperature ( C) and Tgd_HSB is the ground temperature of HSB at 21 m deep ( C). When pump PP1 is ON, valves V1, V4, V5, V17, V18, V20 are open and all other valves are close. The uncertainties of the measurements are summarised in Table 2. 3. TRNSYS model The experimental set up for heat charging was modelled in TRNSYS-17 [19]. Fig. 2 shows the components used in graphical user interface of the TRNSYS model. HSB is represented by Type 257, a modified version of thermal energy system specialists (TESS) Type 557 [20]. The unglazed solar collector is represented by a modified version of TESS Type 559. The other components such as circulating pump, buffer tank and pipes were used from the standard TRNSYS library. The heat charging experiment was conducted from October 2012 to March 2013. During the experiment, temperatures of the water at the inlets and outlets of the heat storage borehole, solar collectors and the buffer tank were measured (Fig. 1). These measurements enabled validation of the TRNSYS Types for HSB, solar collectors and buffer tank. Although the experiment was conducted and TRNSYS types applied were validated for a period of 180 days, only a week data (17e23 December 2012) was presented in this paper. The validated model will later be used to simulate the performance of the entire system which is not discussed in this paper. 4. Results and discussion Fig. 3 shows the measured short wave- solar radiation on the collector plane and ambient air temperature on site during the period 17e23 December 2012. The undisturbed ground temperature of the HSB was measured to be 17.3 C [18]. The ground temperature near to the surface fluctuates depending on the thermal properties of the ground and the conditions at the surface. This fluctuation disappears below a certain depth and becomes stable. Fig. 4 shows the monthly average ground temperature at 2, 21 and 40 m along with the monthly average ambient air temperature. As evident from the measurements the temperature at 2 m below the ground surface fluctuates with the monthly average ambient air temperature whereas the temperature at 21 m and 40 m depths are nearly constant throughout the year. Below 21 m deep, the ground temperature is higher than the ambient in winter and lower than ambient in summer. This feature of the ground temperature is the reason behind the use of GCHP systems. 4.1. Temperature Each section of the model has input and outlet temperature measured which enable to compare with the simulated values. The
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
167
Fig. 1. Schematic diagram of the system.
comparison was done by using three statistical parameters namely root mean squared error (RMSE) and mean bias error (MBE). RMSE gives overall assessment of the difference between the measured and simulated variables, while MBE shows the bias of the difference [21]. RMSE is defined by Equation (1) and MBE by Equation (2) [21].
RMSE ¼
MBE ¼
X
ðYi Xi Þ2
. 0:5 n
(1)
. X ðYi Xi Þ n
(2)
where Xi is the ith measured value, Yi is the ith simulated value and n is the number of observations. The RMSE and MBE for the heat charging period is shown in Table 3.
Table 2 Uncertainties of measured and calculated parameters. Absolute Instrument/measured variable Power transducer 10 W Flow meter 33.6 Kg h1 Thermistor 0.002 C Pyranometer 10 W m2 Wind speed sensor 1.1 m s1 Temperature (ambient) 0.21 C Calculated Parameter Temperature 0.12 C Heat transfer rate 163 W
Relative 0.2% 2% 5% 4%
0.2% 2.8%
Remarks e e For 1 to 60 C range Whichever is higher Whichever is higher Between 0 and 50 C e e
Figs. 5e7 shows the plot of hourly average simulated and measured inlet and outlet temperatures. As evident from the comparisons, there is a good agreement between the TRNSYS model simulations and the measured data. There is a small but irregular under and over estimation of the both the inlet and outlet temperatures by the TRNSYS model. The discrepancy is more in solar collector outlet temperature as indicated in Fig. 5(b). The plausible reason for this could be the solar collector convective heat loss coefficient in TRNSYS model which was computed from the measured data and correlated with the wind speed. The convective heat loss coefficient derived from the measured data was found to be substantially varied with time. The heat losses from piping network between the components (solar collector, HSB and buffer tank) also affect the inlet and outlet temperatures. The heat losses from the pipes were calculated based on the pipe heat loss coefficients which were not measured but estimated from the thermal properties of the pipes and insulations. Similarly there is both under and over estimation of the HSB inlet and outlet temperatures. The HSB outlet temperature is directly proportional to the heat transfer from the fluid to the ground which in turn mainly depends on the effective thermal conductivity of the ground and the ground temperature. A high effective thermal conductivity results in higher heat transfer rate to the ground with lower corresponding borehole outlet temperature and vice versa. Thus the discrepancy exists between the measured and TRNSYS simulated HSB outlet temperature. In case of buffer tank, there is less discrepancy between the simulated and measured inlet and outlet temperatures as shown in Fig. 7. From the above comparisons, it can be concluded that in general
168
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
Fig. 2. Schematic of TRNSYS model for HSB charging.
system in heat charging mode. The energy charged is more than the energy absorbed by the solar collector due to the energy input from the pump. The pump operated in the heat charging mode even when there was no solar radiation. The source of heat in this case was from the water in the buffer tank which also behaves as a temporary thermal storage. Due to the low thermal conductivity of the ground, not all heat could be charged immediately into the ground. Therefore the remaining heat was being stored in the buffer tank temporarily. 4.3. Borehole thermal performance Fig. 3. Short wave solar radiation on collector plane and ambient air temperature (17e 23 Dec 2012).
there is a good agreement between the TRNSYS model simulations and the measured data. Thus it confirms the ability of the modified unglazed solar collector (Type 559), modified HSB model (Type 257) and the buffer tank model (Type 39) to predict their respective outlet temperatures consistently for the stated parameters. 4.2. Energy balance After 180 days of operation, a total of 11.6 GJ of energy was charged into the borehole at an average of 2.7 MJ h1. On the other hand, the solar collector operated for 1204 h and gained 7.9 GJ at an average rate of 6.6 MJ h1. Fig. 8 shows the energy balance of the
Fig. 9 shows the average daily measured and simulated heat charged into the HSB. The TRNSYS simulation underestimated the heat charging capacity of the system. However the deviation is less than 10% most of the days. So, it can be inferred that the TRNSYS model is able to predict the heat transfer from the fluid in the ground heat exchanger to the ground. Fig. 10 shows the daily average heat transfer rate at HSB while the pump is in operation. The maximum heat transfer rate is 86 W m1 with an average of 57 W m1. Due to the continuous heat charging into the borehole, the heat diffuses from the borehole to the surrounding ground. This is indicated by rise in ground temperatures at the monitoring boreholes located at 1, 2 and 3 m from the centre of HSB. Fig. 11 shows the average daily temperature of the ground at 21 m deep near the HSB rises gradually. The HSB appears to be working as indicated by the change in the surrounding ground temperatures. Whether the system is able to retain the heat is yet to be monitored. However, with the Table 3 Statistical parameters of the TRNSYS types used in the model compare to measured values. Model validated Piping from buffer tank to solar collector Solar collectors (Type 559)
Fig. 4. Average monthly ambient air and ground temperatures.
Result
Collector inlet temperature Collector outlet temperature Piping from solar collector to HSB HSB inlet temperature HSB (Type 257) HSB outlet temperature Piping from HSB to buffer tank Tank inlet temperature Buffer tank (Type 39) Tank outlet temperature
RMSE ( C)
MBE ( C)
1.47
0.25
0.76
0.03
2.65 0.67 1.69 0.62
1.37 0.14 0.86 0.05
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
Fig. 5. Measured and simulated (a) inlet and (b) outlet temperature of the unglazed solar collector during the period 17e23 December 2012.
heat charging period ending in April and heat extraction starting in May, the heat will begin to flow in the reverse direction, i.e. from far field towards the boreholes. This is because as the system begins to extract heat from the ground, the temperature around the borehole will decrease and hence result in reverse direction of heat flow. 4.4. Collector performance The useful heat output of an unglazed solar collector depends on the solar radiation received, ambient air temperature, the long wave radiation exchange, and wind speed. These four environmental factors were measured during the experiment. It should be noted that the pump (PP1) was not running when it was raining. So the effects of rain on the collector performance can be eliminated. Fig.12
169
Fig. 6. Measured and simulated (a) inlet and (b) outlet temperature of the borehole during the period 17e23 December 2012.
shows a comparison of the average daily measured and simulated useful heat output for 180 days. There are discrepancies between the measured and simulated heat gain. The reasons for these discrepancies can be attributed to two factors. The first one being the discrepancies in the collector inlet and outlet temperatures as discussed in Section 4.1. The second factor is the uncertainties in the estimation of heat loss coefficient of the solar collector in the model. The total useful energy gained and efficiency are important performance parameters of the solar collector. The instantaneous efficiency of the collector can be expressed as:
hcol ¼ Qu_col =Ac G
(3)
Qu_col ¼ FR Ac ½ðsaÞG UL ðTci Tamb Þ
(4)
where Ac is the collector area (m2), G is the total irradiation on the surface of the collector (W m2), FR is the collector heat removal
170
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
Fig. 8. Energy balance of the system during heat charging.
heat charging experiment. A linear correlation obtained by robust fit of the measured data shows collector efficiency can be represented as 0.51e8.33Cpf. The efficiency is higher at lower performance coefficient. It should be noted that the affect of wind on the solar collector heat gain was not taken into account. The wind speed was found to vary from 0.5 m s1 to 8.5 m s1 with an average of 2.5 m s1. Fig. 14 shows the hourly average collector efficiency during a day. The collector efficiency peaks around noon and gradually decreases. 4.5. System efficiency The efficiency of the system in heat charging mode was calculated based on the total energy that was charged into the ground and the total solar energy available at the collector plus the electricity input to the circulating pump (Equation (6)). This efficiency is different from the efficiency of the solar collector. The system efficiency is an important parameter to evaluate the performance of
Fig. 7. Measured and simulated (a) inlet and (b) outlet temperature of the buffer tank during the period 17e23 December 2012.
factor, s is the transmissivity of collector cover (for unglazed collector s ¼ 1), a is the absorptance of collector (), UL is the total collector heat loss coefficient (W m2 K1), Tci is the collector inlet temperature (K) and Tamb is the ambient temperature (K). Substituting Equation (4) in (3), the instantaneous collector efficiency can be expressed as:
hcol ¼ FR ½a ðUL ðTci Tamb Þ=GÞ ¼ FR a FR UL Cpf
(5)
Collector performance coefficient, Cpf ¼ (TciTamb)/G. As FR, UL and a are assumed to be constant, instantaneous collector efficiency is indirectly proportional to (TciTamb)/G. The efficiency of the unglazed collectors varied between 0.21 and 0.55 with a mean value of 0.31. Fig. 13 shows the collector efficiency verses the collector performance coefficient for the duration of the
Simulated heat chaeged (W)
1800
1200
600
0 0
600
1200
Measure heat charged (W) Fig. 9. Average daily heat charged for 180 days.
1800
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
171
Rate of heat transfer (W m-1)
100
75
50
25
0 2-10-12 1-11-12 1-12-12 31-12-12 30-1-13
1-3-13
31-3-13
Fig. 10. HSB heat transfer rate.
Fig. 13. Variation of hourly average collector efficiency with performance coefficient during the heat charging period.
Fig. 11. Ground temperature recorded in the monitoring boreholes at 21 m depth.
the system in heat charging mode. This efficiency accounts for all energy inputs, outputs and losses of the system.
hhs ¼ Qchg
. Qsa þ Qpi
(6)
where Qchg is the total energy charged into the boreholes (MJ), Qsa is the amount of solar radiation available at the collector surface (MJ) and Qpi is the pump electricity input (MJ). Fig. 15 shows the average
daily system efficiency. The system in this study is able to achieve a maximum daily efficiency of 72% with an average of 38%. 5. Conclusions The application of an unglazed solar collector for heat charging of boreholes has been presented. The unglazed solar collectors are able to collect heat for charging the storage borehole in the ground at an average of 5.7 MJ day1 m2 during the test period. The average heat transfer rate of borehole was found to be 57 W m1. After 180 days of operation, the average temperatures of the ground surrounding the HSB have increased by 2.5 C. Thus there is a potential for such system to be used for heat charging of the
1000
800
600 0.8
System efficiency (-)
Simulated heat gain (W)
Fig. 14. Average hourly solar collector efficiency on 18 December 2012.
400
200
0 0
200
400
600
Measure heat gain (W) Fig. 12. Average daily heat gain for 180 days.
800
1000
0.6 0.4 0.2 0.0 3-10-12
3-11-12
3-12-12
3-1-13
3-2-13
3-3-13
Fig. 15. Average daily system efficiency during heat charging.
3-4-13
172
T. Lhendup et al. / Renewable Energy 67 (2014) 165e172
boreholes. A comparison of the TRNSYS simulations with the measured data showed a good agreement confirming the ability of the TRNSYS model to predict their respective outlet temperatures. The heat charging performance strongly depends on the solar radiation available at the surface of the solar collector. The mean unglazed solar collector efficiency is 30%. The system is able to achieve mean efficiency of 38%. The heating and cooling experiments have been conducted and the results will be reported elsewhere. Acknowledgements We would like to thank Dr Simon Chapuis and Professor Michel Bernier, Department of Mechanical Engineering, Ecole Polytechnique de Montreal, Canada for providing the source code of the modified version of TESS Type 557 for simulations of two U-tubes boreholes. Tshewang Lhendup would also like to thank AusAID for providing a scholarship for the study. Nomenclature Symbols Ac Cpf FR G n Q T UL Xi Yi
collector area (m2) collector performance factor (K m2 W1) collector heat removal factor () total irradiation on the surface of the collector (W m2) number of data points () energy (MJ) temperature (K) total collector heat loss coefficient (W m2 K1) ith measured value ith simulated value
Greek symbols a absorptivity of collector () s transmissivity of collector cover () Subscripts amb ambient chg charged ci collector inlet co collector outlet gd ground pi pump input sa solar radiation at the collector surface Abbreviations SCOP system coefficient of performance CSB coolth storage borehole GCHP ground-coupled heat pump HSB heat storage borehole
MBE RMSE
mean bias error root mean square error
References [1] IEA. Energy efficiency requirements in building codes, energy efficiency policies for new buildings. Paris, France: International Energy Agency; 2008. [2] Stark C, Penney K, Feng A. 2012 Australian energy update. Canberra: Bureau of Resources and Energy Economics; 2012. [3] Che N, Pham P. Economic analysis of end-use energy intensity in Australia. Canberra: Bureau of Resources and Energy Economics; 2012. [4] Rad FM, Fung AS, Leong WH. Feasibility of combined solar thermal and ground-source heat pump systems in cold climate, Canada. Energy Build 2013;61:224e32. [5] Cauret O, Kummert M. Potential of solar assisted ground-source heat pumps with unglazed solar collectors for French office buildings. Fed Eur Heat Vent Air Cond Assoc (REHVA) J August 2011:16e21. [6] Stojanovic B, Akander J. Build-up and long-term performance test of a fullscale solar-assisted heat pump system for residential heating in Nordic climatic conditions. Appl Therm Eng 2010;30:188e95. [7] Bakirci K, Ozyurt O, Comakli K, Comakli O. Energy analysis of a solar-ground source heat pump system with vertical closed-loop for heating applications. Energy 2011;36:3224e32. [8] Mempouo B. Investigations of novel heat pump systems for low carbon homes. United Kingdom: Institute of Sustainable Energy Technology and Institute of Building Technology, Department of Architecture and The Built Environment, The Faculty of Engineering, The university of Nottingham; 2011. p. 308. [9] Eslami-Nejad P, Bernier M. Coupling of geothermal heat pumps with thermal solar collectors using double U-tube boreholes with two independent circuits. Appl Therm Eng 2011;31:3066e77. [10] Han Z, Zheng M, Kong F, Wang F, Li Z, Bai T. Numerical simulation of solar assisted ground-source heat pump heating system with latent heat energy storage in severely cold area. Appl Therm Eng 2008;28:1427e36. [11] Nordell B, Hellström G. High temperature solar heated seasonal storage system for low temperature heating of buildings. Sol Energy 2000;69:511e23. [12] Chiasson AD, Yavuzturk C. Assessment of the viability of hybrid geothermal heat pump systems with solar thermal collectors. ASHRAE Trans 2003;109: 487e500. [13] Trillat-Berdal V, Souyri B, Achard G. Coupling of geothermal heat pumps with thermal solar collectors. Appl Therm Eng Heat Trans Sustain Energy Technol 2007;27:1750e5. [14] Wang H, Qi C. Performance study of underground thermal storage in a solarground coupled heat pump system for residential buildings. Energy Build 2008;40:1278e86. [15] Xi C, Lin L, Hongxing Y. Long term operation of a solar assisted ground coupled heat pump system for space heating and domestic hot water. Energy Build 2011;43:1835e44. [16] Wang E, Fung AS, Qi C, Leong WH. Performance prediction of a hybrid solar ground-source heat pump system. Energy Build 2012;47:600e11. [17] Eslami-Nejad P, Bernier M. Simulations of a new double U-tube borehole configuration with solar heat injection and ground freezing. In: eSim 2012: the Canadian conference on building simulation, Halifax, Nova Scotia 2012. p. 257e68. [18] Lhendup T, Aye L, Fuller RJ. Experimental study of coolth charging of an interseasonal underground thermal storage system. In: Solar 2012, 50th annual AuSES conference. Hawthorn Campus, Melbourne, Australia: Swinburne University of Technology; 2012. [19] Klein SA, Beckman WA, Mitchell JW, Duffe JA, Duffe NA, Freeman TL, et al. TRNSYS 17: a transient system simulation program. WI: University of Wisconsin-Madison; 2010. [20] Chapuis S. Stockage Thermique Saisonnier Dans Un Champ De Puits Geothermiques Verticaux En Boucle Fermee. Montreal: Department of Mechanical Engineering Polytechnic School of Montreal; 2009. [21] Iqbal M. An introduction to solar radiation. Canada: Academic Press; 1983.