Photovoltaic and solar-assisted ground-source heat pump systems

Engineering Sustainability Volume 166 Issue ES1 Photovoltaic and solar-assisted groundsource heat pump systems Varney and Vahdati ice | proceedings Proceedings of the Institution of Civil Engineers Engineering Sustainability 166 February 2013 Issue ES1 Pages 32–45 http://dx.doi.org/10.1680/ensu.11.00006 Paper 1100006 Received 15/02/2011 Accepted 26/09/2011 Published online 31/07/2012 Keywords: energy conservation/geotechnical engineering/ renewable energy ICE Publishing: All rights reserved Photovoltaic and solar-assisted ground-source heat pump systems Kevin E. Varney MSc, MBCS Research student, University of Reading, Berkshire, UK Maria M. Vahdati PhD, CEng, MIChemE Lecturer, University of Reading, Berkshire, UK This review investigates the performance of photovoltaic and solar-assisted ground-source heat pumps in which solar heat is transferred to the ground to improve the coefficient of performance. A number of studies indicate that, for systems with adequately sized ground heat exchangers, the effect on system efficiency is small: about 1% improvement if the heat source is photovoltaic, a 1–2% decline if the source is solar thermal. With possible exceptions for systems in which the ground heat exchanger is undersized, or natural recharge from ground water is insufficient, solar thermal energy is better used for domestic hot water than to recharge ground heat. This appears particularly true outside the heating season, as although much of the heat extracted from the ground can be replaced, it seems to have little effect on the coefficient of performance. Any savings in electrical consumption that do result from an improved coefficient can easily be outweighed by an inefficient control system for the circulation pumps. 1. Introduction This is a review of hybrid photovoltaic (PV) and solar-assisted ground-source heat pump (GSHP) systems in which the heat generated as a by-product of a PV panel, or from solar collectors, could be used to improve the coefficient of performance (COP) of a GSHP. Mono- and polycrystalline silicon PV panels have negative temperature coefficients, meaning their power output decreases the hotter they get (the performance of amorphous silicon PV cells is not so affected by temperature increases). Typical temperature coefficients range from 0?3 to 0?5% per degree Celsius for crystalline silicon cells. While PV generates electricity with heat as a by-product, GSHPs consume electricity to concentrate heat. Both technologies are applicable to buildings, but at higher latitudes, they are most effective at different times of the year. PV is more productive in spring and summer, while GSHPs would typically be used for heating during autumn and winter. GSHPs extract heat energy from the ground, which stays reasonably constant throughout the year. However, too high a concentration of GSHP systems, or too low a surface area of ground heat exchangers could lead to a decline in COP as the ground cools. Therefore, dumping excess heat from the solar panels into the ground is conceivably a useful way of keeping up ground temperature. Solar-assisted GSHPs transfer heat to a brine solution partly from the ground and partly from solar thermal means. Excess heat produced during the sunnier parts of the year could conceivably be used to recharge ground heat before the start of the next heating season. PV-assisted GSHPs could 32 improve the performance of the PV cells by transferring heat away from the PV cells. However, although cooling PV cells should enable them to generate more electricity, while increasing ground heat should reduce a GSHP’s electrical consumption, these electrical savings would have to be measured against the increased electrical consumption of the circulation pumps. 2. Reviewed PV and solar-assisted GSHP systems For an overview of the research studies reviewed herein, see Table 1. 2.1 A simulation of a photovoltaic/thermal (PV/T) assisted GSHP for space and water heating The computer model of the system described by Bakker et al. (2005) was the only one involving PV and GSHP technology in which heat from the PV was transferred to the ground. The PV cells were part of a combined PV/T panel used to heat domestic hot water (DHW), as well as to provide electricity (as shown in Figure 1). Only excess heat not required for DHW was diverted to the ground. This was a system envisaged for a house in the Netherlands. The simulation indicated that the system was capable of covering 100% of the home’s heat demand, including space heating and hot water, and most of its net electricity demand. Average long-term ground temperature was kept constant. A cost comparison was made with a reference system, using separate PV panels and solar collectors. The reference system was 6% cheaper, but took up 32% more roof space. Engineering Sustainability Volume 166 Issue ES1 Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati Solar Heat pump technology source Space conditioning Hot water technology Ground recharge The Netherlands PV/T panel Ground Under-floor heating Yes Los Angeles and Montreal PV panels Ground Air heating and cooling House Switzerland Ground House France Solar collectors Solar collectors Under-floor heating Under-floor heating Combined PV/T and GSHP (a) electrical heating (b) desuperheater (c) solar thermal (d) heat pump Solar thermal and GSHP Solar thermal with electrical back-up Solar thermal and GSHP Study Method Building Bakker et al. (2005) Computer House model Biaou and Computer House Bernier (2008) models Pahud and Lachal (2005) Trillat-Berdal et al. (2006a) Equipped building Equipped building Kjellsson et al. Computer Family (2010) model dwelling Site Sweden Ground Solar collectors Ground Space heating No Yes Yes Yes Table 1. Summary of studies reviewed 2.2 Simulations of a zero net energy home incorporating PV panels, GSHP and various water-heating technologies Biaou and Bernier (2008) created a series of computer models of a zero net energy home, in which all the electricity consumed by the GSHP for space heating, DHW and other household electrical requirements was generated by rooftop PV panels (as PVT panels shown in Figure 2). The house was grid connected, so electrical consumption could exceed production at times, provided the deficit was made up. The models shared the same GSHP system for space heating, but used different technologies for DHW. The four subsystems were modelled at two locations: Los Angeles and Montreal. Space cooling was a major consideration in this study. To grid Storage vessel Cold water Heat pump Floor heating Ground loop heat exchangers Figure 1. Schematic diagram of a PV/T-assisted GSHP for space and water heating, reproduced from Bakker et al. (2005) 33 Engineering Sustainability Volume 166 Issue ES1 Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati PV arrays Inverter Air Lights Air Power meters Appliances DHW Energy distribution Heat pump Ground heat exchanger Figure 2. Schematic diagram of a zero net energy home incorporating PV panels, a GSHP and water heating technology, reproduced from Biaou and Bernier (2008) Of the four DHW heating systems, solar heating was found to require the least electricity at both locations. The heat pump water heater used only slightly more electricity at the Los Angeles site, where it assisted with space cooling. The desuperheater option had the third highest electrical consumption, while the electrical water heater option resulted in greatest electricity use. 2.3 An experimental solar-assisted GSHP home heating system Pahud and Lachal (2005) reported on a 2-year, collaborative project between four academic institutions, which monitored a solar-assisted GSHP installed at a house in Switzerland (as shown in Figure 3). Solar thermal was used primarily for DHW, but any excess heat was stored in the ground. The system worked well over a period of 2 years and suffered no drop in performance. It was discovered that solar heat injected into the boreholes did not improve performance overall as the 34 extra energy consumed by the pumps was not compensated by greater ground temperature. However, they thought it was a useful way of stopping the solar collectors from overheating. 2.4 Geosol solar-assisted GSHP space and water heating system Trillat-Berdal et al. (2006a) introduced a solar-assisted GSHP system, termed Geosol, for space heating and DHW (as shown Figure 4). Solar thermal energy was again used primarily for DHW, but excess heat was transferred to the borehole. This too was an experimental system installed at a house, although it was also modelled using TRNSys (transient system simulation program) software. A focus of this study was the control system, which cut down electricity consumption in the circulation pumps. They concluded that for installations without ground water flow, solar thermal recharge was potentially a useful method of maintaining ground temperatures from year to year. Engineering Sustainability Volume 166 Issue ES1 Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati Ambient temperature Hot water store V4 Hot water Solar collectors V1 V2 7·8 m2 P1 V3 Store stopper Heat pump D Heating D P2 P3 Temperature gauge CE D Heating: 14 kW at B0W35 CE Flow meter Electricity meter Circulation pump Three-way valve Boreholes Immersed heat exchanger 3 80 m Electrical resistance Figure 3. Solar-assisted GSHP, reproduced from Pahud and Lachal (2005) 2.5 Simulations of a GSHP with solar borehole recharge or water heating Kjellsson et al. (2010) used the computer modelling package TRNSys to investigate a series of domestic space heating and DHW systems incorporating a GSHP and solar collectors for a family home in Sweden. The major focus in this study was how best to use the heat from the solar collectors. Six systems were considered initially, in which solar heat is was well sized. They found that borehole recharge was more effective for undersized boreholes, and thought that it could be useful where the density of boreholes in an area was high. Another conclusion was that it was easy to wipe out the savings in electricity consumption by the inefficient use of the extra circulation pumps required to transfer the brine around the system. 3. (a) (b) (c) (d) (e) (f) not included (this is the base case system, system 1) sent to the borehole by way of the heat pump (system 2) sent directly to the borehole avoiding the heat pump (system 3) used to heat DHW (system 4) used to heat DHW in the summer but sent to the borehole by way of the heat pump in winter (system 5) transferred to a brine-filled buffer tank acting as a thermal store for the heat pump (system 6). Of these, systems 3 and 6 did not initially appear promising so only the remaining four were investigated further. The best performing system overall was found to be system 5 (as shown in Figure 5). They concluded that storing thermal heat in the ground during summer was not very effective if the borehole Ground recharge Many of the systems studied had some type of ground recharge. Trillat-Berdal et al. (2006a) injected 2121 kWh of solar heat into the ground over an 11-month period, equivalent to 34% of the heat extracted over the same time. The solar collector area was deliberately oversized for hot water requirements, so that excess heat would be available. The improvement to the COP was only evident in May when there was both heating demand and significant amounts of solar heat available. Otherwise, the main benefit was the restoration of the ground temperature by the start of the next heating season. Ground temperatures were measured at 5 and 45 m depth. After the GSHP was installed, the 45 m ground temperature fell from 11?6 to 11?3 ˚C by June the next year, but rose to 12˚C by the end of summer, thanks in part to the heat injected back into the ground. It was suspected that ground temperature was only this high close to the 35 Engineering Sustainability Volume 166 Issue ES1 Solar collector Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati TSC1 THW0 P1 Combined solar/electrial hot water tank T5 P3 Borehole heat exchanger TBHE2 2 TGD5 2 TGD45 2 TGD85 Heat pump T1 T3 T2 T4 P4 PRESSURE RELIEF TANK No 2 P2 PRESSURE RELIEF TANK No 1 TBHE1 Heating/cooling floor P5 Figure 4. Schematic diagram of the Geosol process, reproduced from Trillat-Berdal et al. (2006a) boreholes. Heat injection rates averaged 39?5 W/m, similar to the rate of extraction. Trillat-Berdal et al. (2006b) also predicted that over a 20-year period ground temperature would fall by 1?2˚C for a system with thermal recharge, compared to 3˚C for a system without. Over 10 years, ground temperature would fall by 2˚C without thermal recharge. This drop in ground temperature would only equate to a drop in COP of 10%. They observed that, in reality, many ground heat exchange loops are oversized, so the drop in COP would not be so high. If, however, the heat exchange loop is undersized, there is a risk that the ground close to the borehole, or the filling materials, might freeze. This could cause the ground to expand, possibly damaging the ground heat exchanger. In addition, ice does not provide good thermal contact. They observed that, taking into account the energy taken by the pump, it would not be worthwhile to inject heat to the ground for normal family houses. However, for buildings such as large blocks of flats, in which there are a large number of boreholes in reasonably close proximity, ground recharge would be easier to justify. Bakker et al. (2005) tested their computer model against low-, medium- and high-quality ground types, and found that the type of ground did not greatly affect the performance of their model. However, the model they used to evaluate ground heat exchange was limited in that it did not consider heat transfer by way of ground water. They stated that for dense ground types such as 36 rock, clay and silt, the omission was justifiable, but that for more porous soils, such as gravel and sand, the omission would lead to an underestimate of the regenerative capacity of the ground of between 5% for dense soils and 25% for porous soils. Ground temperature fluctuations were simulated over a period of 10 years. At 10 m depth, temperatures ranged from about 15 to 5 ˚C in close proximity to the heat exchanger pipes. At a distance of 2 m from the pipes, the fluctuations were much less, to within 0?5 ˚C of average. They reported that ground temperatures are influenced up to 3 m horizontally and 5 m vertically from the GSHP borehole. Bakker et al. (2005) found that heat recharge from their PV/T panels would recharge 83% of ground heat extracted, which would ensure, along with natural recharge, that the ground temperature would not decrease over time. Pahud and Lachal (2005) reported on a house equipped with a GSHP and solar collectors. Solar heat not required for DHW was injected into the ground. They found the improvement in the COP of the GSHP in the following heating season was so slight that the extra electrical consumption of the circulation pumps exceeded the energy savings at the heat pump. The addition of solar heat raised the temperature of the glycol solution to 20˚C, but this only corresponded to a heat injection rate of 20 W/m, only about half that reported by Trillat-Berdal et al. (2006a). Kjellsson et al. (2010) examined four systems in detail. Two of these stored thermal energy in the ground by way of a heat Engineering Sustainability Volume 166 Issue ES1 Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati Warm water Cold water VXS Po Pv VX2 VX1 Pb E V Heating system K D Figure 5. Solar-assisted GSHP space and water heating system in which solar heat may be directed towards DHW, the heat pump evaporator or the borehole, reproduced from Kjellsson et al. (2010) pump, one for the whole year round and the other only in winter. They found that the most efficient was the system that used solar heat for DHW in the summer, but to recharge the borehole, or to raise the temperature of the evaporator during the winter. They found that transferring solar heat to the borehole in summer had little long-term effect on the COP of a system. Natural recharge was sufficient, and the heat dissipated too quickly to be used. In winter the heat collected by the solar collectors was at lower temperature, and could be more effectively used to recharge the borehole than to heat water. Comparing the electrical consumption of the entire heating systems, including auxiliary heating, ground recharge systems seem to improve performance significantly where boreholes were undersized. However, with adequately sized boreholes total electrical consumption for all four systems was similar. From these studies, it would appear that the benefit of ground recharge in terms of improved GSHP performance is marginal. Although ground recharge can be effective at replacing much of the heat extracted during the heating season, it appeared not to make a very great difference to the efficiency of the heating system. Ground recharge would appear to be most beneficial where the ground heat exchanger is too short, where natural recharge from ground water is low, or where neighbouring GSHPs impact the ground heat available. The cost of including ground recharge in a system would have to be compared to the cost of providing a longer ground heat exchanger. In cases where available ground area is limited, the cost comparison may be between a horizontal ground heat exchanger with ground recharge, or a borehole system. 4. Heat Values relating to the heat extracted from the ground, and of heat provided to the buildings are presented in Table 2. A solar-assisted GSHP system, named Geosol, was measured by Trillat-Berdal et al. (2006a) over the course of 12 months, from October 2004 to September 2005. By May 2004, it had 37 Engineering Sustainability Volume 166 Issue ES1 Authors System Trillat-Berdal et al. (2006a) Pahud and Lachal (2005) Geosol process Solar-assisted GSHP space and DHW heating system Bakker et al. Low-quality soil (2005) Medium-quality soil High-quality soil Montreal Biaou and (space heating only) Bernier (2008) GSHP with 100 m Kjellsson et al. (2010) borehole but no solar collectors GSHP with 100 m borehole. All solar heat recharges the borehole by way of the heat pump GSHP with 100 m borehole. All solar heat used for DHW GSHP with 100 m borehole. Solar heat used for DHW in summer but borehole recharge in winter Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati Solar area: m2 Ground recharge: kWhth Solar hot water: kWhth GSHP output: kWhth GSHP input: kWhth Roof m2 Roof m2 8527 6253 2121 177 1600 138 Period COPHP (av.) 12 12 months 3?75 7?8 4?1 30 400 23 550 4660 597 n/k n/k 4?2 32 400 24 700 5620 721 n/k n/k 0 12 months (year 1) 12 months (year 2) 1 year 1 year 1 year 6220 h 2?59 2?66 2?71 3?3 4451 4451 4451 11 340 2732 2778 2809 7904 2306 2306 2306 — 502 502 502 — 916 916 916 — 200 200 200 — 0 1 year 3?73 27 326 20 000 0 0 0 0 10 1 year 3?92 27 326 20 335 8855 886 0 0 10 1 year 3?70 24 667 18 000 0 0 600 60 10 1 year 3?79 24 750 18 220 1220 122 557 56 4?59 Table 2. Heat derived from heat pumps (figures in bold type have been reproduced, those in italics were calculated from other values, while those in plain type were measured from graphs) extracted 5987 kWh of thermal energy from the ground in 838 h of operation. By the end of the 12-month period 6253 kWh of heat had been extracted. Over the same period 2121 kWh of heat was injected back in. The heating system described by Pahud and Lachal (2005) extracted approximately 24 000 kWh of thermal energy from the ground per year, with an additional 150–300 kWh of heat provided directly from the solar collectors. There seems to be a discrepancy in the GSHP energy flows for year 1. Subtracting the heat input from the heat output leaves a GSHP electrical consumption about 7% smaller than is stated, which would increase the COPHP from 4?1 to 4?4. The same calculation for 38 year 2 leads to an electrical consumption 1% smaller than stated. There are also some slight discrepancies between tables, of 50 kWh, for electrical consumption of the circulation pumps. The heat inputs are larger than the other studies involving houses. However, this house had 250 m2 of floor space, compared to 180 m2 for the house studied by TrillatBerdal et al. (2006a), 156 m2 for the house modelled by Biaou and Bernier (2008), and 132 m2 for the house modelled by Bakker et al. (2005). The house was built in the 1980s. Bakker et al. (2005) modelled their PV/T–GSHP system to provide 100% of space heating and DHW demand. Space heating requirements amounted to approximately 8?9 GJ Engineering Sustainability Volume 166 Issue ES1 Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati (2472 kWh). The DHW requirements amounted to approximately 10?3 GJ (2861 kWh), of which about 3?3 GJ (916 kWh) was provided by the PV/T panels, while the remaining 7 GJ (1944 kWh) was provided by the GSHP. The number of hours the GSHP was in operation during the heating season was not stated, although the duration of the heating season was given as 28 May to 10 April, about 225 days. The PV cells appear to be modelled on multi-crystalline silicon technology. The 7?3 GJ (2028 kWh) generated annually by the panels was enough to cover 96% of electricity consumption. The electrical consumption was reduced by a further 0?8 GJ (222 kWh) by the integration of a thermal collector onto the PV panel, as although an extra circulation pump was required, less electricity was required to heat DHW. Working backwards from the quoted COP for medium ground thermal characteristics (see Figure 6), it seems that approximately 2028 kWh of electricity was consumed by the GSHP, of which 931 kWh was for space heating and 1097 kWh for DHW. The remainder of the energy required for DHW was provided by the solar collector part of the PV/T panels. 1?2 GJ (344 kWh) of electricity was saved annually, while at the same time the electrical yield of the PV/T panels was improved (by cooling them). As annual consumption was about 7?6 GW (2111 kWh), it would seem about 416 kWh was used for other purposes. Presumably, this value relates to the electrical consumption of other components of the heating system, such as the electrical heater and circulation pumps. Kjellsson et al. (2010) assumed a total annual heat demand of 29 400 kWh and stated that the GSHP would cover 85–95% of demand. The building’s overall conductance was specified as 250 W/K. The results of the simulations were presented in a series of graphs. However, it was difficult to extract which proportions of the space heating and DHW demand were satisfied by the GSHP, the solar collectors or by auxiliary heating. A particular problem was the ground heat extraction graph. Injected ground heat appears to have been subtracted from heat extracted for two of the systems, but this seems misleading as much of the injected heat would have dissipated before it could be reused. In the mixed mode system it is very difficult to disentangle the effect of borehole recharge from the effect of the solar hot water system. Using values from the graphs, it appears the solar collectors, if used exclusively for DHW, could provide from 1650 to 2000 kWh of the annual demand. Therefore it seems unlikely that the mixed mode system, with only about 6% of the annual insolation to recharge the borehole, could enable as much extra heat to be extracted by the GSHP as is indicated. Notwithstanding the greater size of the house measured by Pahud and Lachal (2005), its heating demand seems large compared to the other houses studied. For example the house modelled by Bakker et al. (2005) required only 26% as much heat per square metre of floor area. This would seem to underline the importance of building to more energy-efficient standards, although it may also reflect the difference between a computer model and reality. 5. 6. Domestic hot water Biaou and Bernier (2008) examined four ways of providing DHW, of which two employed heat pumps: the desuperheater option and the heat pump water heater (HPWH). The desuperheater takes some of the thermal energy from the GSHP to heat water, while the HPWH is a separate air source heat pump. During the heating season, a proportion of the thermal energy going into the HPWH would originally have been provided by the GSHP, but in hot weather this system has the advantage of helping to cool the building. An important point raised by Biaou and Bernier (2008) is that the temperature of mains water will change from place to place, and from month to month. Mains water temperature ranges from 24 to 2 ˚C in Montreal, and from 25 to 17 ˚C in Los Angeles. Therefore, it would require far more energy to heat DHW in Montreal during the winter than Los Angeles during the summer. Photovoltaics and electricity A comparison of the electricity produced and consumed by each of the solar or PV-assisted heat pump systems is presented in Table 3. Biaou and Bernier’s (2008) eight computer models were each designed to import zero net electricity. The lowest amount of roof area was 29?30 m2, generating 8003 kWh of electricity annually, for the system with solar water heating based in Los Angeles. The largest area of PV was 73?20 m2, yielding 13 914 kWh, for a system with electrical water heating based in Montreal. The PV material is mono-crystalline silicon. The model designed by Bakker et al. (2005) was endowed with 25 m2 of hybrid PV/T panels for electricity and water heating. Bakker et al. (2005) use the same GSHP for heating water and for space heating. The water is heated to 55 ˚C, which is fine for hot water, but too hot for an efficient space heating system. Therefore the hot water is mixed with water returning from the heating system, to bring it to a temperature of 30 ˚C, which is appropriate for the heating system. The family dwelling studied by Kjellsson et al. (2010) had a DHW demand of 3400 kWh/year, which was about 12% of the total heat demand. A GSHP was designed to supply 85–95% of the total heat, but was backed up by auxiliary electric heaters. It was observed that one benefit of using solar heat for DHW was potentially to extend the life of the GSHP by avoiding intermittent use. From the graphs it would seem up to 39 Authors System Bakker et al. (2005) Soil conductivity Biaou and Bernier (2008) Montreal Los Angeles Geosol process Solar + GSHP Roof Per m2 25 25 25 73?2 2027 2027 2027 14368 81?1 81?1 81?1 196?3 8133 65?9 12932 196?2 431 6205 56?1 11016 196?4 2147 453 7214 61?0 11974 196?3 2424 3638 264 6326 39?0 11364 291?4 0 2385 2547 188 5120 34?2 9943 290?1 3?36 0 2475 600 269 3344 29?3 8523 290?9 2?98 0 1689 1689 392 3770 29?3 8523 290?9 3?35 2274 — — 271 2545 — — — 3?7 3?8 3?3 3?4 7390 7790 6479 6165 — — — — — — 847 806 570 610 955 1066 8000 8400 8281 8037 — — — — — — — — — — — — 3?3 3?4 6380 6235 — — 287 296 808 749 7474 7281 — — — — — — COPsys (av.) Heating Cooling DHW Other Total 2?07 2?11 2?16 1?92 955 929 912 3580 — — — 642 761 741 728 4605 437 441 416 428 2153 2111 2056 9255 2?21 3996 716 2940 481 2?90 3593 644 1537 2?49 3913 701 1?78 0 2?19 Low Medium High DHW by electric heater DHW by desuperheater DHW by solar collectors DHW by heat pump DHW by electric heater DHW by desuperheater DHW by solar collectors DHW by heat pump Year 1 Year 2 No solar collectors Solar heat recharges borehole by way of the heat pump Solar heat used for DHW Solar heat used for DHW in summer but borehole in winter Table 3. Electricity consumption (figures in bold type have been reproduced, those in italics were calculated from other values, while those in plain type were measured from graphs) Electricity supplied by PV: kWh Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati Trillat-Berdal et al. (2006a) Pahud and Lachal (2005) Kjellson et al. (2010) PV area: m2 Engineering Sustainability Volume 166 Issue ES1 40 Electricity consumption for space conditioning and hot water: kWh Engineering Sustainability Volume 166 Issue ES1 8·5 Electrical yield and use: GJ 8·0 Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati Total electricity use PV/T electrical yield 7·5 7·0 6·5 0·5 Bi lt D e Bi lt D e 61 –1 97 0 Tr y 19 10 re ge PV n. /T ,n W o ith re ge PV n. /T ,w ith re ge n. W ith PV /T ,n o o N Lo w -q ua lit M y ed so iu il m -q ua lit y so H il ig hqu al ity so il 0·0 Figure 6. PV/T yield for different soil properties, reproduced from Bakker et al. (2005). De Bilt, the Netherlands, location of the Royal Dutch Meteorological Institute (KNMI); Try, test reference year 2000 kWh of the DHW demand could be provided by means of solar water heating. The systems studied by Pahud and Lachal (2005) and TrillatBerdal et al. (2006a) use solar heat primarily to heat DHW with only the excess going to recharge ground heat, while the system modelled by Bakker et al. (2005) also heated DHW partly by a PV/T solar panel. The system studied by Trillat-Berdal et al. (2006a) only directed heat to the ground once water temperature had reached 72 ˚C. Both systems had auxiliary electric heating and neither used the GSHP to heat water. The assumption that DHW is the best way to use heat from the solar collectors was supported by the finding by Biaou and Bernier (2008) that heating DHW with solar collectors was the most efficient of the four DHW systems they modelled, as well as the modelling of Kjellsson et al. (2010). 7. Heat pump performance Most of the reviewed studies report the power of heat pump they have available, or which they are modelling (see Table 4). Table 4 also lists the COP values for the heat pump systems. Bakker et al. (2005) used the equation . COP~2:4551z0:0706Tevap to calculate the COP of their system, where Tevap is in degrees centigrade. The condenser temperature was fixed at 55 ˚C so only the evaporator temperature had any bearing on the COP value. They calculated that ground recharge improved COP by about 0?06.They admitted that the COPs used in their simulation were rather low for a modern GSHP. They found that the COP varied only slightly (,5%) depending on the thermal qualities of the three types of ground that they simulated. Authors Year Building Requirement Trillat-Berdal et al. Bakker et al. 2006 2005 House House Biaou and Bernier 2008 House Pahud and Lachal Kjellsson et al. 2005 2010 House Family dwelling Space heating Space heating and DHW Space heating and cooling Space heating Space heating and DHW Powerelec: kW Borehole depth: m 15?8 — 3 6 80 2 6 35 8?75 14 7 COP 3?75 2?59 to 2?71 100 3–6 2 6 90 60–160 3?89 (average) 3?68–4?15 Table 4. Heat pumps 41 Engineering Sustainability Volume 166 Issue ES1 Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati System: no solar collectors, short borehole System: no solar collectors, adequate borehole Borehole size Coefficient of performance (COP) Seasonal performance factor (SPF) Total heat demand: kWh Heat input to GSHP: kWh Heat output by GSHP 5 Qin/(12(1/COP)): kWh Heat deficit according to COP calculation: kWh Electrical consumption of heat pump: kWh Electrical consumption with auxiliary heating but excluding circulation pumps: kWh System electrical consumption: kWh Heat output 5 SPF 6 electrical consumption: kWh Heat deficit according to SPF calculation: kWh Total electricity consumption with ohmic heating (assuming previous figure excludes auxiliary heating): kWh Electricity consumption of circulation pumps: kWh 100 m 3?75 0?75 29 400 8000 10 909 2 909 21 400 Coefficient of performance (COP) Seasonal performance factor (SPF) Total heat demand: kWh Heat input to GSHP: kWh Heat output by GSHP 5 Qin/(12(1/COP)) Heat deficit: kWh Electrical consumption of heat pump: kWh Electrical consumption excluding circulation pumps: kWh Calculated SPF excluding circulation pumps: kWh 18 000 13 500 Table 6. Example of SPF anomaly for a system with an adequately sized borehole 18 491 3?13 15 900 33 900 15 091 Table 5. Example of SPF, COP, electrical and heat anomalies of a base sytem with a short borehole Biaou and Bernier (2008) configured their computer simulations to provide set requirements for space heating, cooling, DHW and appliances for similar houses located in Los Angeles and Montreal. However, dividing the space conditioning requirements by the electrical consumption of the GSHP leads to different COP figures than are stated. For example, an average COP of 3?9 is quoted for the GSHP space conditioning system for the house in Montreal, which requires 11 340 kWh of heating and 2031 kWh of cooling in the summer. The electrical consumption for space conditioning (for the alternative with electrical water heating) is given as 4222 kWh, which would appear to equate to a COP of 3?2. For the Los Angeles location, with no heating load but a cooling load of 7600 kWh and electrical consumption of 2424 kWh for space conditioning, the COP would appear to equate to 3?1 instead of the average value of 4?7 stated. Kjellsson et al. (2010) calculated COPs for the four systems they investigated. However, they appeared to consider the seasonal performance factor (SPF) to be a more important measure, as this included the electrical consumption for the entire heating system, including the auxiliary heating and the circulation pumps. This was a different definition of the SPF to Pahud and Lachal (2005). The COP of the GSHP did not vary very greatly with borehole depth for any of the systems, staying 42 29 20 27 2 7 9 3?73 3?30 400 000 326 074 326 400 in the region of 3?7 to 4?1, but this was because the auxiliary heating system was used more often to prevent the ground from over-cooling. The SPF, however, varied much more, approaching unity at 60 m and levelling off at about 3?5 at 100 m depth. With shorter borehole depths, the simulation showed that the SPF was considerably improved by the use of thermal recharge. However, their values for SPF seem incorrect. Table 5 outlines some discrepancies in heat flows and electricity use for a system with an undersized borehole. It was not very clear whether the figure for system electrical consumption included auxiliary heating. If it did then the system failed to provide two-thirds of the heat demand, but if not then electrical consumption of the circulation pumps seems to be implausibly high (assuming that the auxiliary heating is ohmic, in which 1 kW of electricity converts to 1 kW of heat). Even for systems with adequately sized boreholes the SPF and electrical consumption figures do not appear to be consistent with the COP and extracted ground heat figures, as Table 6 indicates. 8. Costs Table 7 outlines the three studies that discussed the investment costs of their systems. Bakker et al. (2005) itemised the costs of their PV/T-assisted GSHP and compared it to the cost of a similar system, optimised for cost, with separate PV panels and solar collectors. The costs were fairly similar at around J27 000, of which about J8000 were fixed costs common to both systems. The system with the combined PV/T units was about J1600 more expensive but took up less roof space. Pahud and Lachal (2005), by contrast, reported the GSHP and associated plumbing costs as more than the installation of the Bakker et al. (2005) 132 Biaou and Bernier (2008) 156 Space conditioning demand: kWhth 2472 13 371 Pahud and Lachal (2005) 250 30 000 2917 4605 3638 n/k Heating system Place PV and solar-assisted The Netherlands GSHP for space heating and DHW, incorporating ground thermal recharge Zero net energy home Los Angeles equipped with GSHP, for space heating and cooling. PV panels cover all domestic electrical consumption. Montreal No ground thermal recharge Solar-assisted GSHP Table 7. Investment costs: n/k, not known; 1USD 5 0.98 CHF Switzerland Design variant Separate PV and solar collectors Combined PV/T units DHW by electric heater DHW by desuperheater DHW by solar collectors Los Angeles DHW by HPWH DHW by electric heater DHW by desuperheater DHW by solar collectors DHW by HPWH GSHP only with no solar assist Solar thermal collectors supply DHW and ground recharge GSHP + other capital costs PV panel costs Solar collector costs J8080 J15 458 J8080 J19 750 n/k n/k n/k $44 850 $39 330 $33 695 n/k $33 695 n/k n/k n/k $84 180 $75 785 $64 515 $6720 n/k CHF 30 000 $70 150 0 0 CHF 45 600 0 CHF 2500 J2695 $3360 Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati 7600 Hot water demand: kWhth Engineering Sustainability Volume 166 Issue ES1 Authors Floor space: m2 43 Engineering Sustainability Volume 166 Issue ES1 Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati solar collectors and heat transfer system. Nevertheless, the cost of this equipment would be very high for ground thermal recharge alone. A large part of this cost was attributable to the amount of on-site work and the use of non-standard parts, which could be brought down. The primary purpose of the solar collectors was to provide DHW, which in this system, the GSHP did not provide. However, DHW demand was not reported, nor was there an estimate of what a solar hot water system would cost on its own. (2010) remarked that ground thermal recharge could be helpful in areas with a high density of GSHPs. Biaou and Bernier (2008) provided cost data for the PV and solar collector arrangements of their eight system scenarios, but not the balance of costs for the GSHP, ground loop and other components. The costs of the solar panels were expensive when compared to the systems in the other studies, reflecting the large areas of PV panels needed to become a zero net energy home. Their least expensive scenario was a system that used solar collectors to provide DHW. 9. Discussion Biaou and Bernier (2008) modelled four variants of a PVassisted GSHP system for hot water and space conditioning, but did not examine transferring heat from PV panels. Their space conditioning systems would still have transferred heat to the ground, but only as a means of cooling the house. All the other studies did investigate the potential for transferring excess heat to the ground to improve the COP of the GSHP. Pahud and Lachal (2005) concluded that the system efficiency of the GSHP decreased slightly, because the improved COP did not compensate for the additional electrical consumption of the circulation pumps. Bakker et al. (2005) concluded that there was a slight reduction in electrical consumption, but this was a combined effect of an increased PV yield resulting from cooling the panels, in addition to an improved GSHP COP. Both the Pahud and Lachal (2005) and Bakker et al. (2005) studies indicated that ground recharge had a very small effect on the overall electrical consumption of a heating system. Trillat-Berdal et al. (2006a) found that the injection of solar heat into the ground was insufficient to improve the COP of the GSHP during the majority of the heating season, although they did consider it a way of helping the ground to recover heat before the start of the next heating season, especially in places where underground water flow was absent. Kjellsson et al. (2010) found that, during winter, borehole recharge or evaporator heating was potentially useful for systems with undersized boreholes. For the rest of the year, solar energy was better used for hot water. All the reviewed GSHP pumps used vertical ground heat exchangers in boreholes. It is possible that thermal recharge could have more potential for use with horizontal ground heat exchangers. Both Pahud and Lachal (2005) and Kjellsson et al. 44 The systems studied by Bakker et al. (2005), Pahud and Lachal (2005) and Trillat-Berdal et al. (2006a) used solar energy primarily for DHW before diverting heat to the ground. The system studied by Bakker et al. (2005) used hybrid PV/T for water heating and generating electricity. It seemed to be assumed that this is the best way to use solar thermal energy, and this would appear to be backed up by both the study by Biaou and Bernier (2008) and that of Kjellsson et al. (2010). Trillat-Berdal et al. (2006a) asserted the importance of implementing a well-designed control system in order to make best use of solar and ground thermal resources. This was backed up by Kjellsson et al. (2010), who concluded that running the circulation pumps whenever there was solar heat available to recharge the borehole could lead to greater electricity consumption. Using solar thermal energy for DHW is more effective than ground thermal recharge. However, for a roof covered with PV panels, ground thermal recharge could be economically justifiable. This is partly because the solar area is greater than would be required for DHW in summer. In addition, the temperature required for DHW is higher than is ideal for PV panels. Many GSHP systems are required to provide both space and water heating, so some of the plumbing costs associated with solar hot water could be avoided. There appear not to be many studies of systems incorporating heat transfer from PV panels to GSHP ground loops. Interestingly, the Schwedisches Wärmepumpencenter claims a maximum COP of 7 for a solar-assisted GSHP produced by the Swedish company Evi Heat (see www.swc-cottbus.de/produkte/erdwaermepumpen/evi-heat-split-sun). This high value occurs when the temperature of the brine entering the heat pump is at its maximum design temperature of 20 ˚C. An average COP of 5?5 is claimed. 10. Conclusions From the reviewed studies, it would seem that solar to ground thermal recharge only has a slight effect on the performance of a GSHP during the heating season. It would seem that solar energy, from solar collectors, if not PV panels, would be more profitably used for solar hot water. Solar ground recharge would appear to be most beneficial for GSHP systems with undersized ground loops, in locations with high densities of GSHPs or locations with little ground water flow. In any solar ground recharge system, the control system should be carefully designed to prevent electrical consumption by the circulation pumps from undermining any savings in the rest of the system. Notwithstanding these conclusions, it is noted that there are Engineering Sustainability Volume 166 Issue ES1 Photovoltaic and solar-assisted ground-source heat pump systems Varney and Vahdati solar-assisted heat pumps on the market, at least one of which quotes a high average COP. Géothermique à Lugano. Office féréral de l’énergie (OFEN), Switzerland. Trillat-Berdal V, Souyri B and Fraisse G (2006a) Experimental study of a ground-coupled heat pump combined with thermal solar collectors. Energy and Buildings 38(12): 1477–1484. Trillat-Berdal V, Achard G and Souyri B (2006b) Les systèmes solaires combines avec des pompes à chaleur géothermiques à échangeurs enterrés verticaux: étude de l’injection d’énergie solaire dans le sol. Proceedings of IBPSA France 2006, La Réunion, 2–3 November 2006. Kjellsson E, Hellström G and Perers B (2010) Optimization of systems with the combination of ground-source heat pump and solar collectors in dwellings. Energy 35(6): 2667– 2673. REFERENCES Bakker M, Zondag HA, Elswijk MJ, Strootman KJ and Jong MJM (2005) Performance and costs of a roof-sized PV/thermal array combined with a ground coupled heat pump. Solar Energy 78(2): 331–339. Biaou A and Bernier M (2008) Achieving total domestic hot water production with renewable energy. Building and Environment 43(4): 651–660. Pahud D and Lachal B (2005) Mesure des Performances Thermiques d’une Pompe à Chaleur Couplée sur des Sondes WHAT DO YOU THINK? To discuss this paper, please email up to 500 words to the editor at [email protected]. Your contribution will be forwarded to the author(s) for a reply and, if considered appropriate by the editorial panel, will be published as discussion in a future issue of the journal. Proceedings journals rely entirely on contributions sent in by civil engineering professionals, academics and students. Papers should be 2000–5000 words long (briefing papers should be 1000–2000 words long), with adequate illustrations and references. You can submit your paper online via www.icevirtuallibrary.com/content/journals, where you will also find detailed author guidelines. 45 Copyright of Engineering Sustainability is the property of Thomas Telford Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.