ASHRAE LV-11-C002-2011 An Experimental Investigation of the Accuracy of Thermal Response Tests Used to Measure Ground Thermal Properties.pdf

1、6DTLE-DYHG is a graduate student at Chalmers University of Technology, Sweden. -HIIUH6SLWOHUis a professor at Oklahoma State University, Stillwater. 3HU)DKOpQ is a professor at Chalmers University of Technology, Sweden. $Q(SHULPHQWDO,QYHVWLJDWLRQRIWKH$FFXUDFRI7KHUPDO5HVSRQVH7HVWV8VHGWR0HDVXUH*URXQG7

2、KHUPDO3URSHUWLHV6DTLE-DYHG3( -HIIUH6SLWOHU3K3( 3HU)DKOpQ3KStudent Member ASHRAE Fellow ASHRAE Member ASHRAE $%675$ Smith and Perry, 1999). In contrast, ASHRAE (2007) recommends minimum test durations of 36-48 hours. Moreover, Spitler et al. (1999) and Gehlin (2002) emphasize minimum test durations o

3、f 50 and 60 hours, respectively. This paper uses a 270-hour test to analyze the effects of different test lengths on the ground thermal conductivity and the borehole thermal resistance estimations. The paper also addresses the issue of uncertainties caused by using different heat injection rates in

4、groundwater-filled boreholes. Gehlin (2002) and Gustafsson and Westerlund (2010) have shown that, for groundwater-filled boreholes, the choice of heat injection rate can significantly influence the ground thermal conductivity and the borehole thermal resistance estimations. This paper investigates t

5、he effects of different heat injection rates on the ground thermal conductivity, and the borehole thermal resistance estimations, by retesting boreholes with injection rates between 25-150 W/m (8-46 W/ft). (;3(5,0(17$/)$&,/,7$10(7+22/2*A new ground source heat pump test facility (Javed and Fahln, 20

6、10) has been developed at Chalmers University of Technology, Sweden. This new test facility provides a unique opportunity to study thermal properties, including undisturbed ground temperature, ground thermal conductivity and borehole thermal resistance of nine boreholes in close proximity. The labor

7、atorys borehole system consists of nine groundwater-filled boreholes, each about 80 m (262 ft) deep. The boreholes of the new test-site are drilled in a 3 x 3 rectangular configuration. The horizontal cross-section of an individual borehole and the layout of the whole borehole system are shown in Fi

8、gure 1. (a) (b) Figure 1 Geometry and layout of the laboratory boreholes. RockU-tube GroundwaterBrine110 mm (4.3 in)40 mm(1.6 in)LaboratoryBuildingBH-1 BH-2 BH-3BH-4 BH-5 BH-6BH-7 BH-8 BH-910 m (33 ft)8 m (26 ft)N14 ASHRAE TransactionsThe thermal response setup of the laboratory includes a variable

9、capacity electric heater, variable speed circulation pumps and temperature and flow sensors. An electric resistance heater is used to conduct TRTs in heat injection mode. It is also possible to conduct tests in heat extraction mode using a heat pump. All the TRTs reported in this paper were conducte

10、d in the heat injection mode. Before conducting TRTs of laboratory boreholes, undisturbed ground temperature measurements were taken for all nine boreholes. Following the undisturbed ground temperature measurements, TRTs were conducted for nine boreholes. Similar heat injection and flow rates were u

11、sed for all tests. The input power was monitored and kept steady. The chosen heat injection rate matched the expected peak loads on the boreholes. The flow from the variable circulation pumps ensured turbulent regime in the ground loop. The tests were conducted for a minimum of 48 hours. The ground

12、thermal conductivity values were estimated using the method proposed by Gehlin (2002). The method, which is based on the line source approximation, uses the slope of the borehole mean fluid temperature plotted against the logarithmic time to estimate the ground thermal conductivity value. The method

13、 suggested by Beier and Smith (2002) was used to determine the borehole thermal resistance values. This method utilizes the temperature difference between the experimentally measured mean fluid temperature and the borehole wall temperature calculated from the line source approximation. The borehole

14、thermal resistance is then calculated as the ratio of this temperature difference to the heat transfer rate per unit length of the borehole. 81,6785%(*52817(03(5$785(0($685(0(176The undisturbed ground temperature for each borehole was determined using two different approaches. In the first approach,

15、 the fluid was circulated through the undisturbed borehole for a minimum of 30 minutes. The inlet and outlet fluid temperatures were recorded at intervals of 10 seconds. The fluid temperature stabilized after approximately 30 minutes of circulation. The stabilized mean fluid temperature was taken as

16、 a measure of the undisturbed ground temperature. One of the problems with this approach is that, for longer times, the undisturbed ground temperature measurements are affected by the heat gains from the circulation pump. However, this problem was avoided by the use of highly efficient custom-made p

17、umps for borehole applications. The measurements of the undisturbed ground temperature calculated by this approach vary between 8.1 and 9.2 C (46.6 and 48.6 F). One possible explanation of the variations in undisturbed ground temperature measurements is the ambient coupling of the circulating fluid

18、temperatures. With the water table for the laboratory boreholes as high as the ground level, the top of the groundwater-filled boreholes is affected by the ambient temperature changes. The second approach used to measure the undisturbed ground temperature was to monitor the start-up exit fluid tempe

19、ratures from the U-tube. If the fluid is kept long enough in the U-tube, it reaches equilibrium with the surrounding ground. The undisturbed ground temperature can then be determined by taking the average temperature of the fluid present in the U-tube. This approach gave a consistent estimation of 8

20、.3 C (46.9 F) for all boreholes. The undisturbed ground temperatures, calculated using the start-up exit fluid temperature approach, have been used for results reported in this paper. 7575(68/76The TRT of the nine laboratory boreholes were conducted over a period of four months. The duration of most

21、 of the TRTs was between 68 to 98 hours, but tests as short as 48 hours, and as long as 267 hours, were also conducted. The results of ground thermal conductivity and borehole thermal resistance estimations for the nine laboratory boreholes are summarized in Table 1. The ground thermal conductivity

22、estimations for the nine boreholes vary between the extreme values of 2.81 and 3.2 W/mK (1.62 and 1.85 Btu/hftF), whereas the estimated values of borehole thermal resistance vary between the extreme values of 0.049 and 0.074 mK/W (0.085 and 0.128 hftF/Btu). The ground thermal conductivity and boreho

23、le thermal resistance estimations have noticeable random variations. The ground thermal conductivity estimations have a mean value of 3.01 W/mK (1.74 Btu/hftF). The estimated values for all nine boreholes lie within 7 % of the mean value. The estimated values are within commonly assumed uncertaintie

24、s of 10 % in TRT measurements (Witte et al., 2002). On the other hand, the estimated borehole thermal resistance values exhibit larger variations. The borehole thermal resistance values of nine laboratory boreholes lay in a range of 0.062 0.012 mK/W (0.107 0.021 hftF/Btu). 2011 ASHRAE 157DEOH*URXQG7

25、KHUPDO&RQGXFWLYLWDQG%RUHKROH7KHUPDO5HVLVWDQFH(VWLPDWLRQVIRU/DERUDWRU%RUHKROHVBorehole Duration, Hours Ground Thermal Conductivity, W/mK (Btu/hftF) Borehole Thermal Resistance, mK/W (hftF/Btu) 1 75 2.88 (1.66) 0.059 (0.102) 2 54 3.06 (1.77) 0.064 (0.111) 3 267 3.04 (1.76) 0.074 (0.128) 4 48 2.81 (1.6

26、2) 0.049 (0.085) 5 68 2.98 (1.72) 0.064 (0.111) 6 91 2.89 (1.67) 0.063 (0.109) 7 48 3.19 (1.84) 0.064 (0.111) 8 69 3.20 (1.85) 0.065 (0.112) 9 98 3.12 (1.80) 0.069 (0.119) 6(16,7,9,7$1$/6(686,1*&$6(678,(6As seen in the previous section, there exist random variations in the ground thermal conductivit

27、y and borehole thermal resistance estimations for nine laboratory boreholes. This section analyzes the effect of these variations on the design of the borehole systems using three case studies. The first case to be discussed involves the Astronomy-House building at Lund University, Sweden. The build

28、ing has a gross floor area of approximately 5,300 m2(57,050 ft2). It contains offices, a large lecture hall, a library and laboratories. The heating and cooling requirements of the building are met by a borehole system consisting of twenty, 200 m (656 ft) deep boreholes. The borehole system provides

29、 475 MWh (1,620 x 106Btu) of heating and 155 MWh (530 x 106Btu) of free cooling, details of which are given in Table 2. The Astronomy-House building is essentially a heating-dominated building with some cooling requirements. The borehole system of the building uses a 4 x 5 rectangular configuration

30、to store some thermal energy in the ground at a time of energy surplus (i.e. summer) and to extract it in winter. The ground thermal conductivity and the borehole thermal resistance values used to design the borehole system of Astronomy-House were 2.8 W/mK (1.62 Btu/hftF) and 0.07 mK/W (0.121 hftF/B

31、tu), respectively. 7DEOH0RQWKO+HDWLQJDQG&RROLQJHPDQGVRIWKH&DVH6WXG%XLOGLQJVMonth Lund Tulsa Burlington Heating, MWh (106 Btu) Cooling, MWh (106 Btu) Heating, MWh (106 Btu) Cooling, MWh (106 Btu) Heating, MWh (106 Btu) Cooling, MWh (106 Btu) Jan 97.9 (334) - 16.3 (56) - 36.4 (124) - Feb 89.3 (305) -

32、5.0 (17) 1.8 (6) 30.4 (104) - Mar 69.8 (238) 3.4 (12) 1.6 (5) 9.7 (33) 18.3 (62) 0.1 (1) Apr 40.9 (140) 7.3 (25) 0.4 (1) 21.4 (73) 4.5 (16) 5.7 (19) May 20.9 (71) 15.0 (51) - 54.3 (185) 0.5 (2) 23.4 (80) Jun - 25.7 (88) - 103.5 (353) - 37.0 (126) Jul - 33.2 (113) - 127.9 (436) - 63.0 (215) Aug - 31.

33、3 (107) - 128.2 (437) - 54.5 (186) Sep - 19.2 (66) - 54.1 (185) 0.4 (1) 18.7 (64) Oct 31.4 (107) 13.3 (45) 0.3 (1) 31.0 (106) 1.8 (6) - Nov 47.5 (162) 6.4 (22) 1.7 (6) 4.0 (14) 7.6 (26) - Dec 77 (263) - 6.9 (24) - 23.4 (80) - Year 475 (1620) 155 (530) 32 (110) 536 (1830) 123 (420) 202 (690) 16 ASHRA

34、E TransactionsFor case studies 2 and 3, a hypothetical office building, based on three floors of an actual office building in Tulsa, Oklahoma, is used. The building has a footprint of approximately 49 m x 49 m (161 ft x 161 ft), and is 9 m (30 ft) high. The building faade is approximately 60% covere

35、d by double-pane glass windows. The building has high occupancy 1 person per 5 m2(54 ft2) and high lighting and equipment heat gains combined 23.1 W/m2(2.1 W/ft2) with office-appropriate schedules. The building is described more fully by Gentry (2007). The hourly heating and cooling loads for this o

36、ffice building have been determined for very different climates conditions of Tulsa (warm-humid) and Burlington, Vermont (cold-humid) using building energy simulation software. As seen from Table 2, the heating and cooling requirements of the same building, located in Tulsa and in Burlington, are co

37、nsiderably different. For the case of Tulsa, the building has predominant cooling requirements of 536 MWh (1830 x 106Btu) and heating requirements of just 32 MWh (110 x 106Btu). To meet these requirements, a borehole system is designed using a commercially available software. Ideally, the borehole s

38、ystem of the Tulsa building should maximize the heat transfer between the borehole system and the surrounding ground and, hence, should have an open configuration. However, because the cooling requirements of the building are quite high, using an open configuration, like a Line or a U configuration,

39、 will result in a very large and impractical borehole field. Therefore, a rectangular configuration of 9 x 25 was chosen for the borehole field of Tulsa. For the Burlington building, the heating and cooling demands are 123 and 202 MWh (420 x 106and 690 x 106Btu), respectively. As this building has f

40、airly balanced demands, a 7 x 10 rectangular configuration is chosen to exploit the seasonal heat storage ability of the ground. The three cases presented above have been used to analyze the effects of random variations between test results of nine boreholes on the design of borehole fields of these

41、 cases. This is done by calculating the required length of the borehole field, for all three cases, using ground thermal conductivity and borehole thermal resistance values estimated for each of the nine laboratory boreholes. The calculations are made for minimum and maximum heat pump entering fluid

42、 temperatures of -5 and 35 C (23 and 95F) in heating and cooling modes, respectively. The results are summarized in Table 3. In the case of the Lund building, the random uncertainties in ground thermal conductivity and borehole thermal resistance values for nine boreholes result in total borehole le

43、ngth varying between extremes of 3,770 m (12,370 ft) and 4,020 m (13,190 ft). The difference between these two lengths is 250 m (820 ft), approximately equivalent to one and a quarter boreholes out of 20. For the Tulsa building, 20,870 m (68,470 ft) and 22,615 m (74,195 ft) are respectively the smal

44、lest and largest required borehole lengths. The difference between these two lengths is 1,745 m (5725 ft), approximately equivalent to 17 boreholes out of 225. For the Burlington building, the 640 m (2,100 ft) difference between the extreme lengths of 6,860 m (22,505 ft) and 7,500 m (24,605 ft) corr

45、espond to approximately 6 out of 70 boreholes. For all three test cases, the random uncertainties between TRTs moderately affect the total length requirements of the borehole field. These uncertainties change the total borehole length requirements by 6-10 %. If these uncertainties are not accounted

46、for in the design with a factor of safety, the resulting borehole systems can be moderately under-sized. 7DEOH(IIHFWRI5DQGRP9DULDWLRQVLQ*URXQG7KHUPDO&RQGXFWLYLWDQG%RUHKROH7KHUPDO5HVLVWDQFHRQWKH6LHRI%RUHKROH)LHOGTRT Lund Tulsa Burlington Total Length, m (ft) Individual Borehole Depth, m (ft) Total Le

47、ngth, m (ft) Individual Borehole Depth, m (ft) Total Length, m (ft) Individual Borehole Depth, m (ft) 1 4,005 (13,140) 200.2 (656.8) 22,410 (73,525) 99.6 (326.8) 7,120 (23,360) 101.7 (333.7) 2 3,900 (12,795) 195.0 (639.8) 21,600 (70,865) 96.0 (315.0) 7,120 (23,360) 101.7 (333.7) 3 3,990 (13,090) 199.4 (654.