Energy xxx (2014) 1e9

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

A numerical and experimental approach to the estimation of borehole

thermal resistance in ground heat exchangers
Seok Yoon 1, Seung-Rae Lee*, Gyu-Hyun Go
Department of Civil and Environmental Engineering, KAIST, Daejeon 305-701, South Korea

a r t i c l e i n f o a b s t r a c t

Article history: This paper presents a numerical and experimental study on the evaluation of borehole thermal resis-
Received 12 February 2014 tance with TRT (thermal response test) and TPT (thermal performance test) results observed in closed-
Received in revised form loop vertical type boreholes with U and W type GHEs (ground heat exchangers). Field TRTs were carried
22 March 2014
out for 48 h on a closed-loop vertical type borehole, and an equivalent ground thermal conductivity was
Accepted 27 April 2014
estimated using the infinite line source model. Closed-loop vertical type boreholes with U and W type
Available online xxx
GHEs and field ground conditions were numerically modeled using a three dimensional finite element
method to estimate borehole thermal resistance and the TRT results were compared. Field TPTs were also
Keywords:
Ground thermal conductivity
conducted for 100 h continuously to calculate the heat exchange rate and borehole thermal resistance.
Borehole thermal resistance The borehole thermal resistance values were compared with various analytical solutions, and the
Thermal response test multipole and EQD (equivalent diameter) method produced results closer to those of the experimental
Thermal performance test and numerical analysis than the SF (shape factor) method.
Multipole method Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction are used most widely. The closed-loop vertical type ground heat
exchanger is composed of a heat exchange pipe, the ground and
Among various renewable energy resources, the use of grout that fill the empty space between the pipes inside the
geothermal energy has been regarded as energy efficient way of borehole.
space heating and cooling [1e3]. Geothermal energy has a great Considering the high initial construction cost, researchers are
potential as a directly usable type of energy, especially in connec- conducting numerous studies on closed-loop vertical type ground
tion with GSHP (ground source heat pump) systems. Hence, GSHP heat exchangers in order to obtain higher thermal efficiencies [7e
systems combined with various types of GHEs (ground heat ex- 11]. The heat transfer between the surrounding ground through
changers) have been widely used since the early 20th century [4e the ground heat exchanger has a close relationship with the heat
6]. Geothermal energy is often called ubiquitous energy because it transfer between the fluid that circulates within the heat exchanger
can be used anytime and anywhere. pipe and the complex medium (grout/ground) surrounding the
The GSHP system is largely composed of a geothermal heat pipe [12e14]. Therefore, the ground thermal conductivity and
pump and a ground heat exchanger. The ground heat exchanger is a borehole thermal resistance are important design parameters that
system that extracts or emits heat using a circulation fluid such as determine the heat performance of GSHP systems [15,16]. The
flowing water or an anti-freezing solution through the heat ground thermal conductivity is almost accurately measured
exchanger installed in the ground. The system uses the heat source through an in-situ TRT and the obtained value is used as a design
of the ground, which maintains a relatively uniform temperature to parameter in GSHP systems. However, there is no clear guideline on
release heat energy in the summer and absorb heat energy in the a method to determine the borehole thermal resistance and not
winter. The ground heat exchanger is an important element that many studies are being conducted in comparison with ground
determines the performance and initial installation fee of the entire thermal conductivity measurement.
system and generally 150e200 m depth closed-loop vertical types This paper presents a numerical and experimental study to
derive the borehole thermal resistance. U type and W type GHEs
were installed in a landfill area at Incheon International Airport in
* Corresponding author. Tel.: þ82 42 350 3617; fax: þ82 42 350 7200. South Korea, and then in-situ TRTs and TPTs were conducted to
E-mail address: srlee@kaist.ac.kr (S.-R. Lee). verify the suitability of the borehole thermal resistance analytical
1
Current address: Department of Civil and Environmental Engineering, KAIST,
291, Gwahakro, Yuseong-gu, Daejeon, 305-701, South Korea.
models. Furthermore, the TRT test and on-site ground conditions

http://dx.doi.org/10.1016/j.energy.2014.04.104
0360-5442/Ó 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Yoon S, et al., A numerical and experimental approach to the estimation of borehole thermal resistance in
ground heat exchangers, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.04.104
2 S. Yoon et al. / Energy xxx (2014) 1e9

The total pipe length of U and W type GHE was 100 m and 200 m,
respectively. Fig. 2 shows the construction process of the vertical
GHEs.
The ground was composed of silt, clay, weathered granite soil
and weathered rock. The ground water level was 3.5 m below the
top of the embedded borehole, and no noticeable flow of ground
water was observed. The SPT (Standard Penetration Test) N value
was 9/30e33/30 in the partially saturated landfill ground, and
weathered rock appeared 30 m below the ground level. The average
void ratio was 0.95 and the water content was 30e35%.

2.2. Theory of TRT analysis

The heat transfer mechanism of the ground heat exchanger in-

volves the process of absorbing and releasing heat to and from the
grout material and the surrounding ground as the heat transfer
fluid flows through the pipe within the borehole, whereas the heat
transfer behavior between the ground heat exchanger and the
surrounding ground involves a complex mechanism, and the heat
transfer to the ground is through conduction [3]. The heat transfer
governing equation from conduction in the ground is as follows:
!
v2 T v2 T v2 T vT
li þ þ þ ri ci þ qinternal ¼ 0 (1)
2
v x 2
v y 2
v z vt

where T is the temperature, l is the thermal conductivity, r is the

density, c is the specific heat capacity, qinternal is the internal heat
generation. The subscript i denote each region of the GHE such that
g and s indicate the grout and soil, respectively.
Fig. 1. Diagram of ground heat exchanger.
Heat transfer in the GHE involves pipe convection, pipe con-
duction, grout conduction in the borehole and ground conduction.
were numerically modeled using the finite element method In order to measure the ground thermal conductivity in the GHE
coupled with a CFD (computational fluid dynamics) analysis. The system outside the borehole, some analytical equations such as
borehole thermal resistance values were calculated by a numerical line source, cylindrical source, and numerical analysis models have
analysis of the TRT and TPT (thermal performance test) results and been used. Among these, the infinite line source model is the most
compared with analytical solutions. widely employed to measure the ground thermal conductivity due
to its simplicity and convenience in analysis, and the analytical
solution for the heat transfer between the buried pipe and the
2. Experimental setup ground can be obtained by the Kelvin theory. As shown in Fig. 3,
the vertical closed-loop ground heat exchanger has a borehole
2.1. Setup of GHE radius (rb) that is much smaller than the borehole length (L), and
hence it can be assumed to be a line source, and the ground is
U and W type GHEs (Fig. 1) were installed in a partially saturated regarded as an infinite and isotropic medium. When the heat
landfilled runway area of Incheon International Airport. The bore- transfer medium surrounding the line source is a different mate-
hole depth was 50 m, and the diameter was 15 cm. The distance rial, such as that of grout and soil, the following solution of the
between each borehole was 6 m to avoid thermal inference. Poly- heat conduction equation can be obtained when considering the
butylene pipes (inner/outer diameter of pipe ¼ 0.016/0.02 m) were thermal resistance between the borehole and soil about the line
used as GHEs, and bentonite grout was poured into the borehole. source [17e19].

Fig. 2. Construction process of GHE.

Please cite this article in press as: Yoon S, et al., A numerical and experimental approach to the estimation of borehole thermal resistance in
ground heat exchangers, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.04.104
 S. Yoon et al. / Energy xxx (2014) 1e9 3

The fluid average temperature (Tf) in Eq. (7) can be expressed as

a linear equation about lnt as given in Eq. (8).

Tf ¼ Ax þ B (8)

with A ¼ ðQ =LÞ=4pl, x ¼ lnt, and B ¼ Aðð4at=rb2 Þ  gÞ þ ðQ =LÞRb þ

Tg Therefore, once A can be solved, the thermal conductivity (l) can
be obtained by

Q =L
l ¼ (9)
4pA
Therefore, as fluid temperatures at the inlet and outlet of GHEs
can be measured with respect to time by the TRT, an effective
thermal conductivity can be obtained with the value of the slope
(A) from the Tf-lnt relationship.

2.3. Principle of TPT

In-situ TPTs were conducted using U and W type GHEs. Prior to

the TPT, in-situ TRTs were also conducted to measure the ground
thermal conductivity. There is a difference between TRT and TPT.
The TRT is used to measure the ground thermal conductivity where
Fig. 3. Temperature variation around borehole. a pre-defined constant heat power is put into the water tank in the
equipment. Then the ground thermal conductivity can be obtained
by Eq. (9). On the other hand, the TPT evaluates the heat exchange
ZN
Q eu rate of the borehole under the condition that the inlet temperature
Tðr; tÞ  To ¼ du (2)
4pLl u be kept constant. The heat exchange rate per length of borehole can
r2
4at
be calculated using Eq. (10). Tin is the inlet temperature of the fluid,
and Tout is the outlet temperature of the fluid, and m is the flow rate
Here, To is the initial ground temperature and a is the thermal of the fluid
diffusivity of the ground. In Eq. (2), (r2/4at) is the integral variable,
and the right side integral can be expressed as an infinite series as Q mcðTin  Tout Þ
¼ (10)
shown below using an exponential integral. L L

ZN  2  2
eu r r2 1 r2
du ¼ g  ln þ  ,,, (3) 3. Numerical analysis
u 4at 4at 4 4at
r2
4at A finite element analysis program coupled with a CFD module in
In Eq. (3), g is the Euler constant with a value of 0.5772. When COMSOL Multiphysics [20] was used to simulate the TRTs con-
the integral variable (r2/4at) in Eq. (3) is very small, Eq. (3) can be ducted in a closed-loop vertical type borehole considering the
expressed in the following manner. configuration of U and W type GHEs. The governing equation of the
numerical model based on the convection current and conduction
  2   is expressed by Eq. (11) [21].
Q r r2 r2
DTðr; tÞ ¼ ln gþ 1 (4)
4plL 4at 4at 16at
vT rA  3
rAc þ rcAu$VT ¼ V$AlVT þ fD u þ Q þ Qwall (11)
When the heat is transferred from the fluid temperature (Tf) of vt 2dh
the circulating fluid to the ground, as shown in Fig. 3,
Here, Q refers to the regular heat injection and Qwall represents
Eq. (5) is applied, which is from the thermal resistance inside the
the overlapped area of temperature between fluid convection and
borehole.
pipe conduction. Qwall represents the overlapped area of temper-
Q Tf  Tb ature between fluid convection and pipe conduction. A is the pipe
¼ (5) cross section area available for the flow, T is the temperature, c
L Rb
represents the specific heat capacity, and r is the density. Further,
Here, the circulating fluid temperature (Tf) is the average tem- dh is the average hydraulic diameter, fD (non-dimensional) refers
perature of the circulating fluid inlet and outlet regions. The to the coefficient of friction, u represents the tangential velocity,
borehole wall surface temperature (Tb) is calculated as shown and l is the thermal conductivity. A CFD analysis was performed
below by substituting r ¼ rb into Eq. (4), which becomes Tb ¼ T(rb, t). with a Newtonian fluid model (Eq. (11)) with the dynamic prop-
( ! !) erties of a certain fluid, after which the result could be coupled
Q rb2 rb2 rb2 with the heat conduction equation of a solid mass through Eq.
Tb  To ¼ ln gþ 1 (6)
4plL 4at 4at 16at (12).
 
Combining Eqs. (5) and (6) and reorganizing about Tf gives Qwall ¼ ðhZÞeff Tp  Tf (12)
Eq. (7).
!
Q Q 4a Q Here, Tp is the temperature of the pipe wall, which comes from
Tf ¼ ln t þ ln 2  g þ R þ To (7) the heat conduction equation of the solid mass, and Tf is the fluid
4plL 4plL rb L b
temperature in the pipe. From Eq. (12), it can enable the exact heat

Please cite this article in press as: Yoon S, et al., A numerical and experimental approach to the estimation of borehole thermal resistance in
ground heat exchangers, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.04.104
4 S. Yoon et al. / Energy xxx (2014) 1e9

hot-wire method [22]. The soil was remolded to have the same void
ratio and water content of the construction site. The thermal con-
ductivity of the silt-clay in the landfill area below the ground water
table was measured as 2.3 W/m K. Since the soil of every layer was
not sampled, the thermal conductivity of the weathered granite soil
was estimated at around 2.4 W/m K from the fitted model sug-
gested by Park et al. [23] for deriving the thermal conductivity of
Korean granite soil.

4. Analytical solution

4.1. Series sum method

In the series-sum model, the borehole thermal resistance is

estimated by summing the convective resistance of the fluid Rfluid
(Eq. (15)), the conductive resistance between the pipe and the grout
Rpipe (Eq. (16)), the thermal resistance of the grout Rgrout, as
depicted in Eq. (14).

Rb ¼ Rfluid þ Rpipe þ Rgrout (14)

1 0:023Re0:8 Pr n lf
Rfluid ¼ ; where hi ¼ (15)
npdi hi di

 
1 de pffiffiffi
Fig. 4. Finite element model for the numerical simulation. Rpipe ¼ ln ; de ¼ ndo (16)
2plp de  ðdo  di Þ
Here, do is the outer diameter of the pipe, di is the inner diameter
transfer with coupled analysis between fluid convection and pipe of the pipe, de is the equivalent diameter of the pipe, lp is the
conduction. Further, (hZ)eff is the effective value of the heat transfer thermal conductivity of the pipe, hi is the convective heat transfer
coefficient, and Z is the wall perimeter of the pipe. For a circular coefficient of the fluid circulating in the pipe, and n is the number of
tube, the effective hZ can be denoted as: pipes (U type, n ¼ 2; W type, n ¼ 4). Re is the Reynolds number of
the circulating fluid, Pr is the Prandtl number, n ¼ 0.4 for heating
2p
ðhZÞeff ¼ 0 1 (13) and n ¼ 0.3 for cooling, and lf is the thermal conductivity of the
P
N ln r rn
fluid.
1 1 B C
r0 hint
þr þ @ ln A n1
The thermal resistance of the grout is the largest factor in the
N hext
n¼1 overall borehole resistance, and an exact calculation of the grout
resistance is very important for a reliable design of the GSHP sys-
where ln is the thermal conductivity of wall n, rn is the outer radius tem. For the calculation of the grout resistance, a few methods have
of wall n, and hint and hext are the film heat transfer coefficients been introduced, such as Eqs. (17) and (18).
inside and outside of the tube. Fig. 4 represents the finite element
model for the thermal response test simulation. 1 dg
Rgrout ¼ ln pffiffiffi (17)
Table 1 shows the thermo-physical properties used in the nu- 2plg do n
merical analysis. For the finite element model, a free tetrahedral
mesh was used. On the other hand, the mesh element of the heat 1
exchanger wall surface was formed using the wall layer function, Rgrout ¼ (18)
lg b0 ðdg =do Þb1
which was built into the COMSOL Pipe module, rather than creating
a direct mesh. The temperature of the circulating water was derived Here, dg is the grout diameter and lg is the thermal conductivity
using the function obtained from the TRT data (see Fig. 5). The of the grout. Eq. (17) is called an EQD (equivalent diameter method)
thermal conductivity of the landfill soil was measured using TP-08 when calculating the grout thermal resistance. It was proposed
device (Hukseflux Thermal Sensors Inc.) based on a transient based on the concentricity assumption of steady-state operation.
One of the pipe legs was assumed to be concentric with the grout
region and the thermal influence from the other legs was estimated
Table 1
Basic thermal properties of materials for numerical simulation. using the principle of superposition [24]. Further, Remund et al.
[25] considered the shank distance between the pipe legs as an
Materials Thermal Specific heat Density
important factor for the estimation of the thermal resistance
conductivity (W/m K) capacity (J/kg K) (kg/m3)
introduced in Eq. (18). They suggested shape factors b0 and b1
Soil1 0.21 800 1600
presented in Table 2, for which the borehole configurations corre-
Soil2 2.30 1300 2100
Soil3 2.40 1280 2140 sponding to cases A, B, and C are shown in Fig. 6. A W type GHE was
Rock 2.50 879 2640 obtained by back analysis of the GLD (ground loop design) [26], a
Bentonite grout 0.9 380 1580 commercial design program. The GLD uses the shape factor method
Polybutylene pipe 0.38 525 955 for calculating the grout resistance. Therefore, Eq. (18) is known as
Circulating water 0.57 4200 1000
the SF (shape factor) method.

Please cite this article in press as: Yoon S, et al., A numerical and experimental approach to the estimation of borehole thermal resistance in
ground heat exchangers, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.04.104
 S. Yoon et al. / Energy xxx (2014) 1e9 5

Fig. 5. Comparison between measured temperatures and predictions.

4.2. Multipole method Eq. (21) represents the thermal resistance between the pipe and
the borehole wall, while Eq. (23) was used to determine the ther-
The multipole method [27,28] considers the conductive heat mal resistance between two pipes. Further, Eq. (23) is the sum of
flow in and between pipes of different measurements of radius and the thermal resistance of the pipe wall and the fluid boundary layer.
asymmetrical location using a multipole algorithm. This model Like the second formula in Eq. (24), the thermal resistance is used
considers the steady state condition of the borehole. In this model, as the dimensionless quantity bm, which takes any non-negative
the tubes are placed in a circular homogeneous medium inside value such that 0  bm  N. Next, the ultimate borehole resis-
another infinite homogeneous medium. The solution of the tance is finally obtained by the superposition of each component.
multipole method was derived from the steady state two- Thus, once the borehole resistance is determined, the fluid
dimensional heat conduction equation. The temperature rise of temperatures can be estimated with a given borehole wall
the ground is assumed to be caused by the heat sources (which are temperature.
the tubes) and the heat sink at the mirror points (Fig. 7). The
temperature rise at position (x, y), DT(x, y) is then calculated by Eq. (   !)
^ 1 rb rb2
(19). R o
¼ bm þ ln þ s,ln 2 m ¼ 1; :::N
m;m
2plg rpm rb  rm2

q (21)
DTðx; yÞ ¼ ReðWn0 Þ (19)
2plg
Here, q is the heat flux per unit length and Re(Wn0) is the real lg  l
s¼ (22)
component of the zero-th order multipole (Wn0). For higher mul- lg þ l
tipoles, derivations are taken of the Wn0 (Eq. (20)).

1 vj
Wnj ¼ , ðWn0 Þ (20)
ðj  1Þ! vrnj

where Wnj is the jth order multipole of the nth line source, Wn0 is
the 0th order multipole of the nth line source, j is the order of
multipole and rn is the location of the nth tube in the polar coor-
dinate. The temperature rise of the borehole wall as compared to
the undisturbed temperature of the ground soil is obtained by
replacing rn ¼ rb (borehole radius). Among simulation models, this
model is assessed as one of the most accurate method which
exactly describes the real configuration of heat exchangers in a
borehole, and it was used in the EED (Earth Energy Designer) [30]
design program.

Table 2
Shape factors for U-type and Double-U type GHEs.

Configuration U-type GHE Double U-type GHE

b0 b1 b0 b1
A (Close together) 20.10 0.9447 27.68 0.9411
B (Average) 17.44 0.6052 21.36 0.6031
C (Along outer wall) 21.91 0.3796 25.52 0.3921
Fig. 6. Location of GHEs in boreholes.

Please cite this article in press as: Yoon S, et al., A numerical and experimental approach to the estimation of borehole thermal resistance in
ground heat exchangers, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.04.104
6 S. Yoon et al. / Energy xxx (2014) 1e9

1 A
IðcÞ ¼
 (28)
2c2 exp c2 B

Here, A ¼ ic8þ8.5733287c6 þ 18.059017c4 þ 8.637609c2

þ 0.2677737, B ¼ c8þ9.5733233c6 þ 25.6329561c4 þ 21.0996531c2
þ 3.9684969. The total resistance can be calculated by summing the
borehole thermal resistance and the thermal resistance of soil as in
Eq. (29).
X
R ¼ Rb þ Rsoil ¼ Rfluid þ Rpipe þ Rgrout þ Rsoil (29)

The heat exchange rate is calculated as the temperature differ-

ence between the fluid average temperature and the initial ground
temperature divided by the total thermal resistance followed by
Eq. (30).

Q ðT þ Tout Þ=2  To
Fig. 7. Source and sink locations of a single pipe (Young, [29]).
¼ in P (30)
L R
The outlet temperature can be obtained with Eqs. (10) and (30)
(   !) after getting the total thermal resistance of Eq. (29), and heat ex-
^ 1 rb rb2 change rate per depth can be calculated.
Ro m;m ¼ ln þ s,ln  
 msn
2plg rmn rb2  zn zm 
5. Results and discussion
m; n ¼ 1; :::N
(23) 5.1. Experimental results

  In this study, first, in-situ TRTs were conducted for 48 h until

1 rpm 1
Rpm ¼ ,ln þ steady state conditions were achieved with U and W type GHEs to
2plp rpm  dpw 2rpm ,hp (24) derive the ground thermal conductivity. Fig. 8 plots the tempera-
bm ¼ 2plg Rpm ture distribution at the inlet and the outlet of the GHE pipe,
respectively. Heat-free water circulation was performed for 30 min
Here, s is a dimensionless parameter for the two thermal con-
to equalize soil and circulating fluid temperatures. The initial
ductivities, lg is the thermal conductivity
^ of the grout and l is the
temperature of the soil from heat free water circulation was
ground thermal conductivity. Ro m;m is the borehole thermal
16.44  C for the U type GHE and 15.8  C for the W type GHE. With
resistance whenJ ¼ 0, Rpm denotes the thermal resistance of pipe
the temperature value and the slope of (DT/Dlnt), the ground
m, and dpw is the thickness of the pipe wall. Further, hp is the
thermal conductivity from Eq. (9) was evaluated as 2.13 W/m K
convective heat transfer coefficient by Rohsenhow et al. [31]. In
with the U type GHE and 2.15 W/m K with the W type GHE,
contrast to the EQD and SF methods, the multipole method can
respectively. The difference of the GHE type could lead to slightly
consider the pipe configuration at any location and ground thermal
different values of the ground thermal conductivity.
conductivity.
All the in-situ TPTs were conducted for 100 h under continuous
operation conditions. Temperatures of the water at the inlet and the
4.3. Thermal resistance of soil and the heat exchange rate outlet were measured during the tests, and the flow rate was also
measured at the outlet. The inlet temperature was 31  C to consider
Eq. (2) can be transformed as below [9]. the cooling operation, and the flow rate was 7e8 lpm. The heat
exchange rates per length of U and W type GHEs were calculated
ZN ZN with the temperature value at the inlet and outlet and the flow rate
eu eu
2
Q Q Q
Tðr; tÞ  To ¼ du ¼ du ¼ IðcÞ using Eq. (10). Fig. 9 shows the heat exchange rate with respect to
4pLl u 2pLl u 2pLl
r2 pr
2 at
ffiffiffi time using U and W type GHEs. The average heat exchange rates for
4at
100 h for U and W type GHEs were 35.71 W/m, 40.76 W/m,
(25)
respectively. The W type GHE had a 10e15% higher heat exchange
When r ¼ rb, the thermal resistance of soil is as follows: rate than the U type GHE. It can be thought that the W type GHEs
    had a relatively larger heat exchange area than the U type GHE.
Tb  To Q r Q 1
Rs ¼ ¼ I pbffiffiffiffiffi ¼ I pffiffiffiffiffi (26)
Q =L 2pLl 2 at 2pLl 2 Fo 5.2. Borehole thermal resistance

Here, when 0 < x  1, the I function can be calculated as in Eq.

Fig. 10 shows the in-situ TRT results with the numerical analysis
(27)
values of the fluid temperature with respect to time. The experi-
 mental values and numerical analysis results are in good agree-
IðcÞ ¼ 0:5  ln c2  0:57721566 þ 0:99999193c2 ment. Furthermore, after acquiring the borehole wall temperature
 0:24991055c4 þ 0:05519968c6  0:00976004c8 value at the quasi-steady state condition through the numerical
 analysis shown in Fig. 11, Eq. (5) was used to calculate the borehole
þ 0:00107857c10 thermal resistance. The values were 0.233 m K/W and 0.209 m K/W
for U and W type GHEs, respectively. Even though these results
(27)
were not obtained from the experiments since temperature sensors
Moreover, when x  1, were not installed on the borehole wall surface, the borehole

Please cite this article in press as: Yoon S, et al., A numerical and experimental approach to the estimation of borehole thermal resistance in
ground heat exchangers, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.04.104
 S. Yoon et al. / Energy xxx (2014) 1e9 7

34 32

32
30
30

Average Fluid Temperature ( C)

28
Fluid Temperature ( C) 28

26 26 y = 1.4203x + 18.99
2
R = 0.9429
24 U type 24

22
22
20 Inlet
U type
20
18 Outlet

16 18

14 16
0 500 1000 1500 2000 2500 3000 2 3 4 5 6 7 8
Elapsed Time (min) ln (time)
45

40
40

Average Fluid Temperature ( C)

35
35
Fluid Temperature ( C)

30 y = 2.2161x + 21.016
30 2
R = 0.9901
W type
25 25

W type
20
Inlet 20
Outlet

15 15

0 500 1000 1500 2000 2500 3000 2 3 4 5 6 7 8

Elapsed Time (min) ln (time)

(a) Temperature distribution vs. time (b) Temperature distribution vs. lnt
Fig. 8. Fluid average temperature distribution during the TRT (Thermal response test).

thermal resistance value obtained through the numerical analysis shown in Fig. 9, the temperature of the circulating fluid reached
using the borehole wall temperature would likely be similar to the almost a steady state after 25 h. After calculating the soil resistance
actual experimental results. A U type GHE was installed like B case from Eq. (26), the average borehole thermal resistance was calcu-
in Fig. 6, and pipe distance from center to center was about lated from Eqs. (29) and (30) under the steady state condition.
6 w 7 cm. However, a W type GHE was installed between the A and The analytically and experimentally determined borehole ther-
B cases in Fig. 6, and pipe distance was about 5 cm. Thus, the co- mal resistance values from the TRT and TPT results are shown in
efficient of the shape factor in the W type GHE was calculated by Table 3. The borehole thermal resistance values from the TRT and
interpolation between the values of A and B cases. Borehole ther- TPT results had similar values. The borehole thermal resistance
mal resistance values were also derived from the TPT results. As value, as well as the heat exchange rate, can be obtained from the
TPT once the ground thermal conductivity is obtained. As there
were some previous studies to derive the borehole thermal resis-
100 tance using an infinite line source model as in Eq. (7) with the TRT
results, the borehole resistance value was also calculated using Eq.
90 (7). However, as the borehole thermal resistance varies greatly
Heat Exchange Rate (W/m)

according to the arrangement and shape of GHEs, it is not adequate

80
to use in the infinite line source model in calculating borehole
70 thermal resistance because it is almost impossible to consider the
GHE arrangement and shape exactly.
U
60 The comparison with analytical solution revealed that the SF
W
method overestimates the borehole thermal resistance and that the
50
EQD and multipole methods produce results similar to the nu-
40 merical results. The multipole method gave more accurate results
than the EQD, and hence it seems that the multipole method is
30 better than the others in calculating the borehole thermal resis-
-- -- 50h -- -- 100h
tance because it considers the pipe configuration for any location
Elapsed Time (hr)
and ground thermal properties. It is known that the borehole
thermal resistance can be varied according to ground thermal
Fig. 9. Heat exchange rate for 100 h. conductivity [32].

Please cite this article in press as: Yoon S, et al., A numerical and experimental approach to the estimation of borehole thermal resistance in
ground heat exchangers, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.04.104
8 S. Yoon et al. / Energy xxx (2014) 1e9

Fig. 10. Temperature variation of experimental results and numerical analysis.

Fig. 11. Temperature variation at the borehole wall.

6. Conclusion assumed that the ground thermal conductivity for the same
ground will be the same, and we found here that the ground
In this paper, U type and W type GHEs were installed in a landfill thermal conductivity was slightly greater when using the W
area. In-situ TRTs were conducted to measure the ground thermal type. This is probably the result of the difference of heat
conductivity and TPTs were conducted to measure heat exchange exchanger type shapes and the accuracy of temperature and
rates using U and W type GHEs. The borehole thermal resistance power readings.
values were also calculated by a numerical analysis of the TRT, TPT 2. In-situ TPTs were also conducted for 100 h under continuous
results and line source theory with TRT data. Then, a comparative operation conditions. Heat exchange rates per length of U and W
analysis with the analytical solution was then conducted, leading to type GHEs were calculated under the cooling operation condi-
the following conclusions. tion. The average heat exchange rates for 100 h for U and W type
GHEs were 35.71 W/m, 40.76 W/m, respectively. The W type
1. The ground thermal conductivity calculated using the infinite GHE had a 10e15% higher heat exchange rate than the U type
line source model after conducting the in-situ thermal response GHE. The W type GHEs have a relatively larger heat exchange
test for a vertical closed-loop GSHP system with U type and W area than the U type GHE.
type heat exchangers was 2.13e2.15 W/m K. Generally, it is 3. With regard to the GSHP system design, the ground thermal
conductivity and borehole thermal resistance are very impor-
tant factors, and hence the borehole thermal resistance value
Table 3
was calculated through a numerical analysis with the TRT re-
Summary of borehole thermal resistance values.
sults. In order to calculate the borehole thermal resistance based
GHE type FEM (based ILSM TPT Multipole Series on the test in this study, numerous temperature sensors have to
on TRT) (TRT) method
SF EQD be installed on the borehole wall surface in the length and
U type 0.233 0.215 0.230 0.233 0.282 0.227 circumference directions. However, installing temperature
W type 0.208 0.190 0.209 0.209 0.258 0.205 sensors on the borehole wall surface was not possible due to the
SF: Shape factor.
site and construction conditions. Therefore, the site and test
EQD: Equivalent diameter. conditions were modeled exactly through a numerical analysis
ILSM: Infinite line source model. to reproduce the site conditions, and the circulating fluid

Please cite this article in press as: Yoon S, et al., A numerical and experimental approach to the estimation of borehole thermal resistance in
ground heat exchangers, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.04.104
 S. Yoon et al. / Energy xxx (2014) 1e9 9

temperature measured through the thermal response test was [3] Li M, Lai ACK. New temperature response functions (G functions) for pile and
borehole ground heat exchangers based on composite-medium line-source
found to be almost the same as the numerical analysis result.
theory. Energy 2012;38:255e63.
Based on this, the calculated borehole thermal resistance was [4] Man Y, Yang H, Diao N, Cji P, Liu L, Fang Z. Development of spiral heat
0.233 m K/W for the U type GHE and 0.209 m K/W for the W source model for novel pile ground heat exchangers. HVAC R Res
type GHE using the borehole wall surface temperature value 2011;17(6):1075e88.
[5] International Energy Agency. Renewable energy essentials: geothermal; 2010.
obtained through the numerical analysis. Borehole thermal [6] Choi JC, Lee SR, Lee DS. Numerical simulation of vertical ground heat ex-
resistance values were also derived from the TPT results. With changers: intermittent operation in unsaturated soil condition. Comput
the heat exchange rate, soil resistance and average fluid tem- Geotech 2011;38:949e58.
[7] Beck M, Bayer P, Paly MC, Hecht-Mendez J, Zell A. Geometric arrangement and
perature, the borehole thermal resistance value could be ob- operation mode adjustment in low-enthalpy geothermal borehole fields for
tained under the steady state condition. Therefore, the borehole heating. Energy 2013;49:434e43.
thermal resistance values from the TRT and TPT results had [8] Gao J, Zhang X, Liu J, Li K, Yang J. Numerical and experimental assessment of
thermal performance of vertical energy piles: an application. Appl Energy
similar values. It can be concluded that the borehole thermal 2008;28:2295e304.
resistance value, as well as the heat exchange rate, can be ob- [9] Jun L, Xu Z, Jun G, Jie Y. Evaluation of heat exchange rate of GHE in geothermal
tained from the TPT once the ground thermal conductivity is heat pump systems. Renew Energy 2009;34:2898e904.
[10] Pahud D, Matthey B. Comparison of the thermal performance of double U-
obtained.
pipe borehole heat exchangers measured in situ. Energy Build 2001;33:
4. The borehole thermal resistance value from the TRT and TPT 503e7.
results was compared with various analytical solutions. The [11] Park H, Lee SR, Yoon S, Choi JC. Evaluation of thermal response and perfor-
mance of PHC energy pile. Appl Energy 2013;103:12e24.
comparison results showed that the SF model overestimates the
[12] Beier RA, Smith MD, Spitler JD. Reference data sets for vertical borehole
borehole thermal resistance while the EQD and multi-pole ground heat exchanger models and thermal response test analysis. Geo-
methods were in better agreement with the numerical anal- thermics 2011;40:79e85.
ysis results. In addition, the borehole thermal resistance values [13] Park H, Lee SR, Yoon S, Shin H, Lee DS. Case study of heat transfer behavior of
helical ground heat exchanger. Energy Build 2012;53:137e44.
calculated using the infinite line source model did not have [14] Roth P, Georgiev A, Busso A, Barraza E. First in situ determination of ground
significant differences from the test results. It has been reported and borehole thermal properties in Latin America. Renew Energy
that the borehole thermal resistance value varies greatly ac- 2004;29(12):1947e63.
[15] Lee CH, Park MS, Kang SH, Sohn BH, Choi HS. Comparison of effective thermal
cording to the arrangement and shape of the GHE, but because it conductivity in closed-loop vertical ground heat exchangers. Appl Therm Eng
is difficult to consider the heat exchange arrangement and shape 2011;31:3669e76.
with the infinite line source model, calculating the borehole [16] Park S, Lee SR, Park H, Yoon S, Chung JH. Characteristics of an analytical so-
lution for a spiral coil type ground heat exchanger. Comput Geotech 2013;49:
thermal resistance using the infinite line source model is not 18e24.
desirable. [17] Carslaw H, Jaeger J. Conduction of heat in solids. Oxford; 1947.
5. In conclusion, this study confirmed that the borehole thermal [18] Ingersoll LR, Adler FT, Plas HJ, Ingersoll AC. Theory of earth heat exchangers
for the heat pump. ASHVE Trans 1950;56:167e88.
resistance value for the vertical closed-loop GSHP system design
[19] Ingersoll L, Zobel O, Ingersoll A. Heat conduction with engineering geological
is accurate when using values obtained with the multi-pole and other applications. New York: McGraw-Hill; 1954.
method or the EQD method. However, unlike the SF and EQD [20] COMSOL. Multiphysics user’s guide. COMSOL Multiphysics version 4.2a; 2011.
[21 Incropera FP, Burkhard S. Fundamentals of heat and mass transfer. 4th ed. John
methods, the multi-pole method can reflect the thermal con-
Wiley and Sons; 1996.
ductivity of the ground, and as such further research on the [22] Yun TS, Santamarina JC. Fundamental study of thermal conduction in dry soils.
effect that ground thermal conductivity has on the borehole Granul Matter 2008;10:197e207.
thermal resistance is necessary for a more accurate analysis of [23] Park H, Park HS, Lee SR, Go GH. Estimation of thermal conductivity of
weathered granite soils. J Korean Soc Civ Engrs 2012;32(2C):69e77 [in
the borehole thermal resistance analytical solution. Korean].
[24] Shonder JA, Beck JV. Field test of a new method for determining soil for-
mation thermal conductivity and borehole resistance. ASHARE Trans
Acknowledgment 1999;106:843e50.
[25] Remund CP. Borehole thermal resistance: laboratory and field studies.
ASHARE Trans 1999;105:439e45.
This research was supported by the basic research project by the
[26] Gaia Geothermal. Ground loop design software. GLD2012.
National Research Foundation of Korea under the Ministry of Sci- [27] Bennet J, Claesson J, Hellstrom G. Multipole method to compute the
ence, ICT and Future Planning, and by the 2011 Construction conductive heat flows to and between pipes in a composite cylinder. Notes on
Technology Innovation Project (11 Technology Innovation E04) heat transfer. Lund, Sweden: Lund University; 1987. p. 3.
[28] Hu P, Yu Z, Zhu N, Lei F, Yuan X. Performance study of a ground heat
under the Korea Agency for Infrastructure Technology exchanger based on the multipole theory heat transfer model. Energy Build
Advancement. 2013;65:231e41.
[29] Young TR. Development, verification and design analysis of the borehole fluid
thermal mass model for approximation short term borehole thermal
References response. Master Thesis, Oklahoma State University.
[30] Earth Energy Designer, Blocon. (www.buildphysics.com).
[1] Florides GA, Pouloupatis PD, Kalogirou S, Messaritis V, Panayides I, Zomeni Z. [31] Rohsenhow WM, Hartner JP, Ganic EN. Handbook of heat transfer funda-
The geothermal characteristics of the ground and the potential of using mental. New York: McGraw-Hill; 1985.
ground coupled heat pumps in Cyprus. Energy 2011;36:5027e36. [32] Go GH, Yoon S, Park DW, Lee SR. Thermal behavior of energy pile considering
[2] Johnston IW, Narsilio GA, Colls S. Emerging geothermal energy technologies. ground thermal conductivity and thermal interference between piles.
KSCE J Civ Engrs 2011;15(4):643e53. J Korean Soc Civ Engrs 2013;33(6):2381e91 [in Korean].

Please cite this article in press as: Yoon S, et al., A numerical and experimental approach to the estimation of borehole thermal resistance in
ground heat exchangers, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.04.104