Vertical-borehole ground-coupled heat pumps: A review of models and systems

H. Yang
a,
*
, P. Cui
a
, Z. Fang
b
a
Renewable Energy Research Group, Department of Building Services Engineering, The Hong Kong Polytechnic University, Hong Kong, China
b
Ground Source Heat Pump Research Center, Shandong Jianzhu University, Jinan, China
a r t i c l e i n f o
Article history:
Received 13 March 2009
Received in revised form 28 April 2009
Accepted 28 April 2009
Available online 29 May 2009
Keywords:
Ground-coupled heat pump
Ground heat exchanger
Hybrid ground-coupled heat pump
Heat transfer
a b s t r a c t
A large number of ground-coupled heat pump (GCHP) systems have been used in residential and com-
mercial buildings throughout the world due to the attractive advantages of high efciency and environ-
mental friendliness. This paper gives a detailed literature review of the research and developments of the
vertical-borehole GCHP technology for applications in air-conditioning. A general introduction on the
ground source heat pump system and its development is briey presented rst. Then, the most typical
simulation models of the vertical ground heat exchangers currently available are summarized in detail
including the heat transfer processes outside and inside the boreholes. The various design/simulation
programs for vertical GCHP systems primarily based on the typical simulation models are also reviewed
in this paper. Finally, the various hybrid GCHP systems for cooling or heating-dominated buildings are
well described. It is found that the GCHP technology can be used both in cold and hot weather areas
and the energy saving potential is signicant.
2009 Elsevier Ltd. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2. Simulation models of vertical GHEs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.1. Heat conduction outside borehole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.1.1. Kelvins line source. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.1.2. Cylindrical source model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.1.3. Eskilsons model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.1.4. Finite line-source solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.1.5. Short time-step model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.1.6. Other typical numerical models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2. Heat transfer inside borehole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2.1. One-dimensional model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2.2. Two-dimensional model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2.3. Quasi-three-dimensional model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3. Comparisons of the analytical and numerical models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3. Computer programs for GCHP design/simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.1. IGSHPA approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.2. Design tools based on the line-source model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.2.1. The Lund programs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.2.2. The GLHEPRO program. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.2.3. The GeoStar program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.2.4. The building simulation programs integrated with GHE models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.3. The GchpCalc program based on cylindrical source model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.4. Numerical simulation programs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4. Hybrid GCHP systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1. HGCHP systems with supplemental heat rejecters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
0306-2619/$ - see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apenergy.2009.04.038
* Corresponding author. Tel.: +852 2766 7801; fax: +852 2774 6146.
E-mail addresses: behxyang@polyu.edu.hk (H. Yang), sdcuiping@gmail.com (P. Cui), fangzh@sdjzu.edu.cn (Z. Fang).
Applied Energy 87 (2010) 1627
Contents lists available at ScienceDirect
Applied Energy
j our nal homepage: www. el sevi er. com/ l ocat e/ apenergy
4.2. HGCHP systems with hot water supply. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.3. HGCHP systems with solar collectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5. Conclusions and recommendations for future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
1. Introduction
Ground source heat pump (GSHP) systems use the ground as a
heat source/sink to provide space heating and cooling as well as
domestic hot water. The GSHP technology can offer higher energy
efciency for air-conditioning compared to conventional air condi-
tioning (A/C) systems because the underground environment pro-
vides lower temperature for cooling and higher temperature for
heating and experiences less temperature uctuation than ambient
air temperature change.
The rst known record of the concept of using the ground as
heat source for a heat pump was found in a Swiss patent issued
in 1912 [1]. Thus, the research associated with the GSHP systems
has been undertaken for nearly a century. The rst surge of interest
in the GSHP technology began in both North America and Europe
after World War Two and lasted until the early 1950s when gas
and oil became widely used as heating fuels. At that time, the basic
analytical theory for the heat conduction of the GSHP system was
proposed by Ingersoll and Plass [2], which served as a basis for
development of some of the later design programs.
The next period of intense activity on the GSHPs started in
North America and Europe in 1970s after the rst oil crisis, with
an emphasis on experimental investigation. During this time peri-
od, the research was focused on the development of the vertical-
borehole system due to the advantage of less land area require-
ment for borehole installation. In the ensuing two decades, consid-
erable efforts were made to establish the installation standard and
develop design methods [36].
To date, the GSHP systems have been widely used in both resi-
dential and commercial buildings. It is estimated that the GSHP
system installations have grown continuously on a global basis
with the range from 10% to 30% annually in recent years [7].
The GSHPs comprise a wide variety of systems that may use
ground water, ground, or surface water as heat sources or sinks.
These systems have been basically grouped into three categories
by ASHRAE [8], i.e. (1) ground water heat pump (GWHP) systems,
(2) surface water heat pump (SWHP) systems and (3) ground-cou-
pled heat pump (GCHP) systems. The schematics of these different
systems are shown in Fig. 1. The GWHP system, which utilizes
ground water as heat source or heat sink, has some marked advan-
tages including lowinitial cost and minimal requirement for ground
surface area over other GSHP systems [9]. However, a number of fac-
tors seriously restrict the wide application of the GWHP systems,
such as the limited availability of ground water and the high main-
tenance cost due to fouling corrosion in pipelines and equipment. In
addition, many legal issues have arisen over ground water with-
drawal and re-injection in some regions, which also restrict the
GWHP applications to a large extent. In a SWHP system, heat rejec-
tion/extraction is accomplished by the circulating working uid
through high-density polyethylene (HDPE) pipes positioned at an
adequate depthwithina lake, pond, reservoir, or other suitable open
channels. Natural convection becomes the primary role in the heat
exchangers of the SWHP system rather than heat conduction in the
heat transfer process in a GCHP system, which tends to have higher
heat exchange capability than a GCHP system. The major disadvan-
tage of the system is that the surface water temperature is more af-
fected by weather condition, especially in winter.
In a GCHP system, heat is extracted from or rejected to the
ground via a closed loop, i.e. ground heat exchanger (GHE), through
which pure water or antifreeze uid circulates. The GHEs com-
monly used in the GCHP systems typically consist of HDPE pipes
which are installed in either vertical boreholes (called vertical
GHE) or horizontal trenches (horizontal GHE). In the horizontal
GCHP systems, the GHEs typically consist of a series of parallel pipe
arrangements laid out in dug trenches approximately 12 m below
the ground surface. A major disadvantage is that the horizontal
systems are more affected by ambient air temperature uctuations
because of their proximity to the ground surface. Another disad-
vantage is that the installation of the horizontal systems needs
much more ground area than vertical systems.
In the vertical GCHP systems, the GHE congurations may in-
clude one, tens, or even hundreds of boreholes, each containing
one or double U-tubes through which heat exchange uid is circu-
lated. Typical U-tubes have a diameter in the range of 1938 mm
and each borehole is normally 20200 m deep with a diameter
ranging from 100 mm to 200 mm. The borehole annulus is
Fig. 1. Schematics of different ground source heat pumps.
H. Yang et al. / Applied Energy 87 (2010) 1627 17
generally backlled with some special material (named as grout)
that can prevent contamination of ground water. A typical bore-
hole with a single U-tube is illustrated in Fig. 2.
The worldwide growing energy shortage and increasing energy
demand have recently driven a great incentive of the GSHP appli-
cations in air conditioning eld. Among the various GSHP systems,
the vertical GCHP system has attracted the greatest interest in re-
search eld and practical engineering as well, owing to its advan-
tages of less land area requirement and wide range of
applicability. During the past few decades, a considerable number
of studies have been carried out to investigate the development
and applications of the GCHP systems with various types of GHEs
and addressed their individual advantages and disadvantages in
detail. Furthermore, various hybrid GCHP systems which couple
the conventional GCHP equipment with a supplemental heat rejec-
tion/generation device have been recently developed in order to
improve the economics of the GCHP systems for unbalanced cli-
mates. Several literature reviews on the GCHP technology have
been reported [7,1012]. This paper mainly presents a detailed lit-
erature review of the vertical-borehole GCHP systems, primarily
related to the typical heat transfer models of the GHEs and the rep-
resentative design/simulation programs as well as advanced engi-
neering applications of hybrid GCHP systems.
2. Simulation models of vertical GHEs
The major difference between the GCHP system and a conven-
tional A/C system is the use of a special heat exchanger (i.e. GHE)
instead of a cooling tower. The construction costs of the GHEs
are critical for the economical competitiveness of a GCHP system
for a heating or an A/C system. On the other hand, the GHE size also
plays a decisive role on the operation performance of the GCHP
system. Thus, it is of great importance to work out sophisticated
and validated tools by which the thermal behavior of any GCHP
system can be assessed and then, optimized in technical and eco-
nomical aspects.
The main objective of the GHE thermal analysis is to determine
the temperature of the heat carrier uid, which is circulated in the
U-tubes and the heat pump, under certain operating conditions. A
design goal is then to control the temperature rise of the ground
and the circulating uid within acceptable limits over the system
lifespan.
The rule of thumb approximation method was in fashion for a
long time, which was discussed by Ball et al. [1]. Rules of thumb
can serve well for specic localities where soil and weather condi-
tions are fairly uniform because design specications are primarily
based on the experience with related installations. However, some
systems have suffered from the inability of the rule of thumb
designers to properly assess the effect of varied design parameters,
such as shallower burial depth, lower shank spacing between U-
tube legs and larger borehole space in ground surface.
In addition to the rule of thumb method, several models with
different complexity have been developed for the design and per-
formance prediction of the GHEs in engineering applications. Actu-
ally, the heat transfer process in a GHE involves a number of
uncertain factors, such as the ground thermal properties, the
ground water ow and building loads over a long lifespan of sev-
eral or even tens of years. In this case, the heat transfer process
is rather complicated and must be treated, on the whole, as a tran-
sient one. In view of the complication of this problem and its long
time scale, the heat transfer process may usually be analyzed in
two separated regions. One is the solid soil/rock outside the bore-
hole, where the heat conduction must be treated as a transient pro-
cess. With the knowledge of the temperature response in the Fig. 2. Schematic of a vertical grouted borehole.
Nomenclatures
a ground thermal diffusivity (m
2
/s)
c water heat capacity (kJ/(kgK))
D uppermost part of the borehole (m)
F run fraction
H borehole length (m)
k ground thermal conductivity (W/m K)
M uid mass ow rate per borehole (kg/s)
q
l
heating rate per borehole length (W/m)
r
b
borehole radius (m)
R thermal resistance (m
2
K/W)
T or t temperature (C)
t
0
ground initial temperature (C)
z variable of borehole depth (m)
e heat transfer efciency of a borehole
s time (s)
Subscripts
b borehole wall
C cooling mode
f uid
H heating mode
p pipe of the U-tube
s soil or ground
Abbreviation
GLHEPRO ground loop heat exchange program
18 H. Yang et al. / Applied Energy 87 (2010) 1627
ground, the temperature on the borehole wall can then be deter-
mined for any instant on specied operational conditions. Another
sector often segregated for analysis is the region inside the bore-
hole, including the grout, the U-tube pipes and the circulating uid
inside the pipes. This region is sometimes analyzed as being stea-
dy-state or quasi-steady-state and sometimes analyzed as being
transient. The analyzes on the two spatial regions are interlinked
on the borehole wall. The heat transfer models for the two separate
regions are as follows.
2.1. Heat conduction outside borehole
A number of simulation models for the heat transfer outside the
borehole have been recently reported, most of which were based
on either analytical methodologies or numerical methods. A few
models were developed based on the incorporation of the analyti-
cal and numerical solutions, such as Eskilsons model [6].
2.1.1. Kelvins line source
The earliest approach to calculating the thermal transport
around a heat exchange pipe in the ground is the Kelvin line-source
theory, i.e. the innite line source [2,13]. In the Kelvins line-source
theory, the ground is regarded as an innite medium with an initial
uniform temperature, in which the borehole is assumed as an in-
nite line source. The heat transfer in the direction of the borehole
axis, including the heat ux across the ground surface and down
the bottom of the borehole, is neglected. The heat conduction pro-
cess in the ground is, therefore, simplied as one-dimensional one.
According to the Kelvins line-source theory, the temperature re-
sponse in the ground due to a constant heat rate is given by:
tr; s t
0

q
l
4pk

1
r
2
4as
e
u
u
du 1
where r is the distance from the line-source and s the time since
start of the operation; t the temperature of the ground at distance
r and time s; t
0
the initial temperature of the ground; q
l
the heating
rate per length of the line source; and k and a are the thermal con-
ductivity and diffusivity of the ground.
The solution to the integral term in Eq. (1) can be found from
the related references [5,14,15]. Although it is characterized by
the simplicity and less computation time, this model can only be
applied to small pipes within a narrow range of a few hours to
months because of the assumption of the innite line source
[6,16]. It was estimated that using the Kelvins line source may
cause a noticeable error when
as
r
2
b
< 20 [14].
This approach has been widely utilized in some analytical de-
sign methods that are currently used to analyze the heat transfer
of GHEs [5,15,17]. A number of improvements for this approach
have been proposed to account for some complicated factors so
that the accuracy can be comparable to that of the numerical
methods. Of all these methods employing Kelvins line-source the-
ory, the Hart and Couvillion method may be more accurate than
others [15].
2.1.2. Cylindrical source model
The cylindrical source solution for a constant heat transfer rate
was rst developed by Carslaw and Jaeger [18], then rened by
Ingersoll et al. [14], and later employed in a number of research
studies [1921]. It is actually an exact solution for a buried cylin-
drical pipe with innite length under the boundary condition of
either a constant pipe surface temperature or a constant heat
transfer rate between the buried pipe and the soil. In the cylindri-
cal source model, the borehole is assumed as an innite cylinder
surrounded by homogeneous medium with constant properties,
i.e. the ground. It also assumes that the heat transfer between
the borehole and soil with perfect contact is of pure heat
conduction.
Based on the governing equation of the transient heat conduc-
tion along with the given boundary and initial conditions, the tem-
perature distribution of the ground can be easily given in the
cylindrical coordinate:
@
2
t
@r
2

1
r
@t
@r

1
a
@t
@s
r
b
< r < 1
2pr
b
k
@t
@r
q
l
r r
b
; s > 0
t t
0
0 s 0; r > r
b

2
where r
b
is the borehole radius.
The cylindrical source solution is given as follows:
t t
0

q
l
k
Gz; p 3
where z
as
r
b
; p
r
r
b
.
As dened by Carslaw and Jaeger [18], the expression G(z, p) is
only a function of time and distance from the borehole center.
The temperature on the borehole wall, where r = r
b
, i.e. p = 1, is
of interest as it is the representative temperature in the design of
GHEs. However, the expression G(z, p) is relatively complex and in-
volves integration from zero to innity of a complicated function,
which includes some Bessel functions. Fortunately, some graphical
results and tabulated values for the G(z, p) function at p = 1 are
available in some related references [14,19]. An approximate
method for G was proposed by Hellstrom [22] and presented by
Liu et al. [23].
2.1.3. Eskilsons model
Both the one-dimensional model of the Kelvins theory and the
cylindrical source model neglect the axial heat ow along the bore-
hole depth; therefore they are inadequate for the long-term oper-
ation of the GCHP systems. A major progress was made by Eskilson
(1987) to account for the nite length of the borehole [6]. In Eskil-
sons model, the ground is assumed to be homogeneous with con-
stant initial and boundary temperatures, and the thermal
capacitance of the borehole elements such as the pipe wall and
the grout are neglected. The basic formulation of the ground tem-
perature is governed by the heat conduction equation in cylindrical
coordinates:
@
2
t
@r
2

1
r
@t
@r

@
2
t
@z
2

1
a
@t
@s
tr; 0; s t
0
tr; z; 0 t
0
q
l
s
1
H

DH
D
2prk
@t
@r
j
rr
b
dz

4
where H is the borehole length; D means the uppermost part of the
borehole, which can be thermally neglected in engineering practice.
In Eskilsons model, the numerical nite-difference method is
used on a radialaxial coordinate system to obtain the temperature
distribution of a single borehole with nite length. The nal
expression of the temperature response at the borehole wall to a
unit step heat pulse is a function of s=s
s
and r
b
=H only:
t
b
t
0

q
l
2pk
gs=s
s
; r
b
=H 5
where s
s
H
2
=9a means the steady-state time. The g-function is
essentially the dimensionless temperature response at the borehole
wall, which was computed numerically.
Another important achievement of Eskilsons model is that the
special superimposition was employed to account for the temper-
ature responses for multiple boreholes. In addition, the sequential
temporal superimposition was used to calculate the temperature
response (i.e. g-functions) to any arbitrary heat rejection/extrac-
tion which can be decomposed into a set of single pulses. In other
H. Yang et al. / Applied Energy 87 (2010) 1627 19
words, the overall temperature response of the GHE to any heat
rejection/extraction at any time can be determined by the special
and temporal superimpositions.
The disadvantage of this approach, however, is time-consum-
ing, and it can hardly be incorporated directly into a design and en-
ergy analysis program for practical applications, since the g-
functions of the GHEs with different congurations have to be
pre-computed and stored in the program as a massive database.
The interpolation function is also needed in using the database
which may lead to some computing errors.
2.1.4. Finite line-source solution
Based on the Eskilsons model, an analytical solution to the -
nite line source has been developed by a research group which
considers the inuences of the nite length of the borehole and
the ground surface as a boundary [24]. Some necessary assump-
tions are taken in the analytical model in order to derive an analyt-
ical solution:
The ground is regarded as a homogeneous semi-innite medium
with constant thermophysical properties.
The boundary of the medium, i.e. the ground surface, keeps a
constant temperature (t
0
), same as its initial one throughout
the time period concerned.
The radial dimension of the borehole is neglected so that it may
be approximated as a line-source stretching from the boundary
to a certain depth, H.
As a basic case of study, the heating rate per length of the source,
q
l
, is constant since the starting instant, s = 0.
The computation results from the analytical solution were com-
pared with the data from numerical solutions in references [6,24],
and they agreed with each other perfectly when as=r
2
b
P5: The
solution of the temperature excess was given by Zeng et al. [24]:
t r; z; s t
0

q
l
4kp

H
0
erfc

r
2
zh
2
p
2

as
p

r
2
z h
2

erfc

r
2
zh
2
p
2

as
p

r
2
z h
2

dh
6
It can be seen from Eq. (6) that the temperature on the bore-
hole wall, where r = r
b
, varies with time and borehole depth. The
temperature at the middle of the borehole depth (z = 0.5H) is
usually chosen as its representative temperature. An alternative
is the integral mean temperature along the borehole depth,
which may be determined by numerical integration of Eq. (6).
For the convenience of applications, the former is usually ac-
cepted as the representative temperature in the design and
analysis program. It is obvious that the integral of Eq. (6) can
be computed much faster than the numerical solution of the
same heat conduction problem in the semi-innite domain with
long duration. The methodology has been complied in the later
design and simulation software developed by other researchers
[25,26].
As discussed in the aforementioned section, when time tends to
innity, the temperature rise of the Kelvins theory tends to inn-
ity, whereas the temperature from the nite line-source model ap-
proaches steady state, which corresponds to the actual heat
transfer mechanism. With respect to long duration, substantial dis-
crepancy between the Kelvins model and the nite line source
may yield [27].
2.1.5. Short time-step model
Since both Eskilsons model and the nite line-source model ne-
glect the effect of the thermal capacity of the borehole including
the U-tubes, circulating uid and the grout, the dimensionless tem-
perature responses on the borehole wall are only valid approxi-
mately for the time greater than
5r
2
b
a
, estimated by Eskilson [6].
For a typical borehole with a radius of 55 mm, the required time
may be any time period between 2 and 6 h.
Yavuzturk and Spitler presented a short time-step model for the
simulation of the transient heat transfer in vertical GHEs, which
can be accurate down to an hour and below [28,29]. This short
time-step model is based on the two-dimensional, fully implicit -
nite volume formulation and utilizes an automated parametric grid
generation algorithm for different pipe sizes, shank spacing and
borehole geometry. The numerical results are expressed in terms
of a short time-step response factor (i.e. g-function), which is a
very useful extension of the long time-step response factor devel-
oped by Eskilson [6]. The numerical model was cast as a compo-
nent model for TRNSYS developed by Klein et al. [30]. The
authors stated that it can evaluate the energy consumption and
electrical demand of the GCHP system in hourly or shorter time
intervals.
2.1.6. Other typical numerical models
Hellstrom (1989, 1991) and Thornton et al. (1997) proposed a
simulation model for ground heat stores, which are densely packed
ground loop heat exchangers used for seasonal thermal energy
storage [22,31,32]. This type of system may be directly used to heat
buildings with or without a heat pump. The duct storage model
(named as DST) divides the ground storage volume with multiple
boreholes into two regions: one is the volume that surrounds a sin-
gle borehole, described as the local region; the other is called
global region, which denotes the ground volume between the
bulk of the heat store volume and the far eld. A two-dimensional
nite difference scheme is used to solve the ground temperature in
the global region while the one-dimensional numerical method
is employed to calculate the temperature in the local region.
Since the ground heat storage is mainly used to provide a heating
function, the boreholes are generally spaced in a quite dense eld,
which may be not suitable for some buildings with a considerable
amount of cooling loads [33].
Muraya, et al. (1996) developed a transient nite-element mod-
el of the heat transfer around a vertical U-tube heat exchanger for a
GCHP system to study the thermal interference that occurred be-
tween the adjacent legs of the U-tube [34]. The thermal interfer-
ence was quantied by dening a heat exchanger effectiveness.
The impacts of the separation distance, leg temperatures, different
ambient soil temperatures and backlls were all investigated.
Rottmayer, et al. (1997) presented a nite difference model that
simulated the heat transfer process of a U-tube heat exchanger
[35]. A geometric factor was introduced to account for the noncir-
cular geometry used to represent the pipes in the borehole. The
model was validated for simple conditions and compared with an
existing model, resulting in good agreement.
A three-dimensional unstructured nite volume model for the
vertical GHEs was developed by Li and Zheng [36]. The model uses
Delaunay triangulation method to mesh the cross-section domain
of the borehole eld, and consequently retains the geometric struc-
ture in the borehole. The surrounding soil is divided into many lay-
ers in the vertical direction in order to account for the effect of
changing uid temperature with depth on the thermal process in
the bore eld. Finally, a comparison of the model predictions and
experimental data shows that the model has good prediction
accuracy.
2.2. Heat transfer inside borehole
The thermal resistance inside the borehole, which is primarily
determined by the thermal properties of the grouting materials
20 H. Yang et al. / Applied Energy 87 (2010) 1627
and the arrangement of ow channels of the borehole, has a signif-
icant impact on the GHE performance. The main objective of this
analysis is to determine the entering and leaving temperatures of
the circulating uid in the borehole according to the borehole wall
temperature, its heat ow and the thermal resistance. A few mod-
els with varying degrees of complexity have been established to
describe the heat transfer inside the GHE boreholes.
2.2.1. One-dimensional model
A simplied one-dimensional model has been recommended for
GHE design, which considers the U-tube as a single equivalent
pipe [5,37]. In this model, both the thermal capacitance of the
borehole and the axial heat ow in the grout and pipe walls are
negligible as the borehole dimensional scale is much smaller com-
pared with the innite ground outside the borehole. Thus, the heat
transfer in this region is approximated as a steady-state one-
dimensional process. The authors stated that the simplied one-
dimensional model was appropriate and convenient for most engi-
neering practices except for the analyzes to deal with dynamic re-
sponses within a few hours. However, this oversimplied model
seems inadequate and unsatisfactory because it is incapable of
evaluating the impact of the thermal short circuiting between
the U-tube legs on the performance of the GHEs.
2.2.2. Two-dimensional model
Hellstrom (1991) derived the analytical two-dimensional solu-
tions of the thermal resistances among pipes in the cross-section
perpendicular to the borehole axis, which is superior to empirical
expressions and one-dimensional model [22]. In the two-dimen-
sional, the temperature of the uid in the U-tubes is expressed as
a superposition of the two separate temperature responses caused
by the heat uxes per unit length, q
1
and q
2
, from the two pipes of
the U-tube, as shown in Fig. 3. If the temperature on the borehole
wall, t
b
, which is also considered as uniform along the borehole
depth, is taken as a reference of the temperature excess, the uid
temperatures in the U-tubes can be obtained from the following
equations:
t
f1
t
b
R
11
q
1
R
12
q
2
t
f2
t
b
R
12
q
1
R
22
q
2

7
where R
11
and R
22
are the thermal resistances between the circulat-
ing uid in each pipe and the borehole wall, and R
12
is the resistance
between the two pipes. A linear transformation of Eq. (7) leads to:
q
1

t
f1
t
b
R
D
1

t
f1
t
f2
R
D
12
q
2

t
f2
t
b
R
D
2

t
f2
t
f1
R
D
12

8
where R
D
1

R
11
R
22
R
2
12
R
22
R
12
; R
D
2

R
11
R
22
R
2
12
R
11
R
12
; and R
D
12

R
11
R
22
R
2
12
R
12
: For the
instance of the symmetric disposal of the U-tube inside the bore-
hole (i.e. R
11
= R
22
), these resistances can be deduced as:
R
D
1
R
D
2
R
11
R
12
; and R
D
12
R
2
11
R
2
12
=R
12
: The steady-state
heat conduction problem in the cross-section of a borehole was
analyzed in detail with the line-source and multipole approxima-
tion by Hellstrom [22].
It is noticeable that there is no distinction between the entering
and exiting pipes since this model does not take into account the
heat transmission on the axial ow of the circulating uid. In this
case, Eskilson made the following assumptions to simplify the
problem [6]: t
f1
t
f2
t
f
and q
1
q
2
q
l
=2. Therefore, the ther-
mal resistance between the uid and borehole wall can be deter-
mined by:
R
b2
R
11
R
12
=2 9
With the aid of these assumptions the temperatures of the uid
entering and exiting the GHE can be calculated. Being superior to
the model of an equivalent pipe, this two-dimensional model pre-
sented quantitative expressions of the thermal resistance in the
cross-section, and provided a basis for discussing the impact of
the U-tube disposal on the heat conduction. However, the temper-
atures of the uid circulating through different legs of the U-tubes
are, in fact, different. As a result, the thermal interference or ther-
mal short circuiting between the U-tube legs is inevitable, which
degrades the effective heat transfer in the GHEs. With the assump-
tion of identical temperature of all the pipes, it is impossible for the
two-dimensional model to reveal the impact of this thermal inter-
ference on the GHE performance.
2.2.3. Quasi-three-dimensional model
On the basis of the two-dimensional model aforementioned, a
quasi-three-dimensional model was proposed by Zeng et al. [38],
which takes account of the uid temperature variation along the
borehole depth. Being minor in the order, the conductive heat ow
in the grout in axial direction, however, is still neglected so as to
keep the model concise and analytically manageable. The energy
equilibrium equations can be written for up-ow and down-ow
of the circulating uid:
Mc
dt
f1
dz

t
f1
t
b

R
D
1

t
f1
t
f2

R
D
12
Mc
dt
f2
dz

t
f2
t
b

R
D
2

t
f2
t
f1

R
D
12

0 6 z 6 H 10
Two conditions are necessary to complete the solution:
z 0; t
f1
t
0
f
z H; t
f1
t
f2

11
The general solution to this problemis derived by Laplace trans-
formation, which is slightly complicated in form. For the instance
of the symmetric placement of the U-tube inside the borehole,
the temperature proles in the two pipes were illustrated by Diao
et al. [27]. For the purpose of practical applications an alternative
parameter e t
0
f
t
00
f
=t
00
f
t
b
is derived from the temperature
proles, which is named as the heat transfer efciency of the bore-
hole. It should be noticed that t
0
f
and t
00
f
are the entering/exiting
uid temperatures to/from the U-tube. From the derived tempera-
ture prole a more accurate heat conduction resistance between
the uid inside the U-tube and the borehole wall can be calculated
by
R
b3

H
Mc
1
e

1
2

12
Fig. 3. Conguration of a U-tube in a borehole.
H. Yang et al. / Applied Energy 87 (2010) 1627 21
The authors validated that the quasi-three-dimensional model
was more accurate than the other current models and recom-
mended it for the design and thermal analysis of the GHEs.
2.3. Comparisons of the analytical and numerical models
Although the numerical models can offer a high degree of exi-
bility and accuracy (especially on short-termscales) compared with
the analytical models, most of them using polar or cylindrical grids
may be computationally inefcient due to a large number of com-
plex grids. Besides, the numerical models are inconvenient to be
incorporated directly into a design and energy analysis program,
unless the simulated data are pre-computed and stored in a massive
database.
The analytical models are usually found based on a number of
assumptions and simplications in order to solve the complicated
mathematical algorithms; therefore, the accuracy of analytical re-
sults is slightly reduced due to the assumption of the line source
at the center of the borehole, which neglects the physical size of
the U-tube in the borehole [39]. However, the required computa-
tion time of the analytical model is much less compared with the
numerical models. Another advantage is that the straightforward
algorithm deduced from the analytical models can be readily inte-
grated into a design/simulation program. A summary of the charac-
teristics of the numerical and analytical models of the GHEs
reviewed is given in Table 1.
It can be found from the literature review that a great number
of studies on the heat transfer analysis of the vertical GHEs have
been carried out, which can provide a valuable basis for the de-
sign/simulation of the GCHP systems in engineering applications.
3. Computer programs for GCHP design/simulation
The reliability and stability of a GHE design mainly depends on
its ability to reject or extract heat to/from ground over a long-term
period and avoidance of excessive heat buildup or heat loss in the
ground. A good design program for the GCHPs should have high
computational efciency, which allows the calculation of the tran-
sient effects over long time periods. Actually, there are numerous
uncertain factors which affect to some extent the nal sizing of a
GHE, such as the employed mathematical methodology, the al-
lowed minimum/maximum temperatures of the uid entering to
the heat pump, the properties of the ground, GHE layout, borehole
conguration and net annual energy transfer to the ground. How-
ever, the mathematical methodology or the heat transfer model of
the GHEs is the crucial part for a design program. A number of de-
sign tools for vertical GHEs based on some typical heat transfer
models have been developed in the last two decades, most of
which have been discussed in aforementioned sections. A short
overview of some typical design tools currently available for verti-
cal GHEs is presented in the following section.
3.1. IGSHPA approach
The International Ground-Source Heat Pump Association
(IGSHPA) is one of the earliest groups which are involved in the
development of GHE design methods [5]. The IGSHPA modeling
procedure is based on the Kelvins line-source theory with a num-
ber of simplifying assumptions. It can only estimate the GHE length
for the coldest and the hottest month of a year using the following
two simple formulas. For heating,
L
H

Capacity
H
COP
H
1
COP
H

R
p
R
s
F
H

T
s;m
T
min
13
and for cooling,
L
C

Capacity
C
COP
C
1
COP
C

R
p
R
s
F
C

T
max
T
s;m
14
where R
s
is the soil resistance of a single vertical heat exchanger ob-
tained by the Kelvins line-source theory; R
p
the thermal resistance
of the U-tube which is assumed to be an equivalent diameter
pipe; F is the run fraction; T
s,m
the mean soil temperature; and T
min
and T
max
are the design heat pump minimum and maximum enter-
ing uid temperatures, respectively.
Obviously, the simplied method does not account for the tran-
sient effects of the long-term operation and the variations of build-
ing loads, which may cause a signicant deviation from practical
conditions.
3.2. Design tools based on the line-source model
3.2.1. The Lund programs
The early PC-programs for sizing vertical GHEs have been pre-
sented by a group of researchers from University of Lund, Sweden
[22,40,41]. The algorithms of the programs were developed based
on the Eskilsons approach (also refereed as nite line source)
where the temperature response of the borehole eld is converted
to a set of non-dimensional temperature response factors, called g-
functions [6,40]. Those g-functions depend on the spacing between
the boreholes on the ground surface and the borehole depth. The g-
function values obtained from the numerical simulations have
been stored in a data le, which is accessed for rapid retrieval of
data by the PC-programs. A major drawback of the programs for
engineering applications is the input interface. Users need to have
a good knowledge of the input parameters and to do some calcula-
tion in advance.
To make the Lund-programs use easier, a more user-friendly
program, called the Earth Energy Designer (EED), has been devel-
oped on the same basis as the previous PC-programs [42,43]. In
the new program, EED, the uid temperature of the GHEs is
Table 1
Comparison of the current models of GHEs.
Model Method Thermal interference between
boreholes
Boundary effects
Outside borehole Kelvins line source Innite line source Yes No
Cylindrical source Innite cylindrical source Yes No
Eskilions model Combination of numerical and analytical methods Yes Yes
Finite line- source solution Analytical method Yes Yes
Short time-step model Numerical methods Yes Yes
Model Method Thermal interference between
U-tube pipes
Heat ux along depth
Inside borehole One-dimensional model (equivalent pipe) No No
Two-dimensional model Yes No
Quasi-three-dimensional model Yes Yes
22 H. Yang et al. / Applied Energy 87 (2010) 1627
calculated according to the monthly heating/cooling loads and the
borehole thermal resistance. The thermal properties of the ground
as well as properties of pipe materials and heat carrier uids are
saved in a database in the program. However, for the cases with
changing borehole depth or borehole distance, the program needs
to interpolate between suitable g-functions as the g-functions
stored in the database are calculated as a function of a ratio of
borehole distance over borehole depth (i.e. B/H). The interpolation
process causes inevitably computing errors.
3.2.2. The GLHEPRO program
The GLHEPRO is developed primarily for designing vertical
ground loop heat exchangers used in commercial/institutional
buildings [44]. The Eskilsons approach is the basis for the GLHE-
PRO program. It was developed in order to make the Swedish
methodology developed by Eskilson tractable for American users.
The design methodology is based on a simulation that predicts
the temperature response of the ground loop heat exchangers to
monthly heating and cooling loads and monthly peak heating
and cooling demands over a number of years. The temperature of
the uid inside the pipes in the borehole is determined using a
one-dimensional steady-state borehole thermal resistance. The de-
sign procedure involves the automatic adjustment of the ground
loop heat exchanger size in order to meet user-specied minimum
or maximum heat pump entering uid temperatures.
3.2.3. The GeoStar program
A software package named GeoStar has been developed for the
design and simulation of the GHEs by a research group in China
[25,26]. This software package is able to size GHEs to meet the
user-specied minimum and maximum entering uid tempera-
tures to a heat pump for a given set of design conditions, such as
building load, ground thermal properties, borehole conguration
and heat pump operating characteristics. The GHE heat transfer
models employed in the software consist of two sectors: one is
the heat conduction process of the solid soil/rock outside the bore-
hole and the other is the region inside the borehole. For the rst
sector, an explicit analytical solution of the nite line source in a
semi-nite medium is derived for convenient calculation of the
thermal resistance outside the borehole for long time periods
[24]. With the assumption of the same heat transfer rate per unit
length of each borehole, the borehole wall temperature of each
individual borehole in a GHE can be obtained by means of analyt-
ical solution. For the heat transfer insider the borehole, the quasi-
three-dimensional model which takes into account the uid
temperature variation along the borehole depth is used to calculate
the uid temperatures of the up-ow and down-ow channels
[38]. The analysis on the two spatial regions is interlinked on the
borehole wall. In addition, the modeling procedure uses spatial
superimposition for multiple boreholes and sequential temporal
superimposition to determine the arbitrary heating or cooling
loads of the systems.
3.2.4. The building simulation programs integrated with GHE models
The EnergyPlus program, which is a popular building energy
simulation program, was extended to allow GCHP system simula-
tions [45]. Models of a water-source heat pump and a vertical
GHE were implemented in the Energy-Plus. The GHE model also
uses Eskilsons g-functions to model the response to time-vary-
ing heat uxes and has been extended to include a computationally
efcient variable time-step load aggregation scheme.
Another building energy analysis program eQUEST and its sim-
ulation engine DOE-2.2 were enhanced to facilitate the design and
energy analysis of the GCHP systems [46]. A vertical GHE model,
based upon the g-function algorithm, was implemented in the
Building Creation Wizard of eQUEST. The author claimed from
the validation results that the enhanced eQUEST/DOE-2.2 is a very
useful tool for GCHP system design and energy analysis.
The GHE model which is also based on the g-function was inte-
grated in the HVACSIM+ modeling environment which is capable
of modeling HVAC systems, HVAC controls, building energy man-
agement systems and other thermal systems [47].
All the three programs employ an effective steady-state bore-
hole thermal resistance to calculate the actual heat transfer inside
the borehole. In general, the three similar models of GHEs, which
take advantage of the wide functionality of the building energy
simulation programs, can conveniently calculate the building heat-
ing and cooling loads and facilitate the design and energy analysis
of the GCHP A/C systems. They also have the exibility to compare
the energy consumption between the GCHP A/C system and a con-
ventional HVAC system.
3.3. The GchpCalc program based on cylindrical source model
The GchpCalc is a program to help engineers in the design of
vertical GCHP systems. The detailed fundamental concepts of this
program can be found in Kavanaugh and Rafferty [4]. The method
is based upon the solution of the cylindrical source model which
was developed and evaluated by Carslaw and Jaeger [18]. This
method uses a simple steady-state heat transfer equation to solve
the required borehole length, which considers three different
pulses of heat to account for long-term heat imbalances, average
monthly heat rates during the design month, and maximum heat
rates for a short-term period during a design day. The thermal
resistance of the ground responding to each pulse is calculated
by means of modied solution of Garslaw and Jaeger. This method
has been implemented in the software program, the GchpCalc, and
has been used widely within the United States for design of vertical
ground coupled systems [4].
3.4. Numerical simulation programs
Some numerical simulation codes, mainly based on the nite-
difference method, have been developed in the GCHP eld [43].
Among them, the most representative numerical program is the
TRNSYS with DST-module developed by Pahud and Hellstrom
[48]. The TRNSYS is a modular system simulation package where
users can describe the components that compose the system and
the manner in which these components are interconnected. Be-
cause the program is modular, the duct ground storage model
(DST) for the vertical GHEs is easily added to the existing compo-
nent libraries. As discussed in section 2, the numerical programs
are inconvenient to users unless the simulated data are pre-com-
puted and stored in a massive database.
4. Hybrid GCHP systems
It is well known that the GCHP systems can achieve better en-
ergy performance in specic locations where building heating
and cooling loads are well balanced all the year round because of
the long-term transient heat transfer in the GHEs. However, most
buildings in warm-climate or cold-climate areas have unbalanced
loads, dominated by either cooling loads or heating loads. When
the GCHP systems are used in the cooling-dominated buildings in
warm climates, more heat will be rejected to the ground than that
extracted from the ground on an annual basis. The heat buildup
within the ground will denitely increase the ground temperature,
which can consequently deteriorate the system performance over
time. To maintain a high operating performance, the cooling-dom-
inated buildings require a much larger GHE size compared to the
buildings with balanced loads. Similarly, when the GCHP systems
H. Yang et al. / Applied Energy 87 (2010) 1627 23
are applied to heating-dominated buildings in cold climates, they
will need extremely large land area to satisfy the higher heating
load requirement. However, the high initial cost and large land
area required for the GHE installation restrict to a large extent
the wider application of the GCHP technology in the buildings with
unbalanced loads.
An alternative to decrease the initial cost of the GCHP system
and, at the same time, to improve the system performance is to
employ a supplemental heat rejecter or heat absorber, which is
called the hybrid GCHP (HGCHP) system. Incorporating a supple-
mental heat rejecter/absorber can reduce a fair amount of heat re-
jected/extracted into/from the ground and then effectively balance
the ground thermal loads, which can consequently reduce the ini-
tial cost of the system and improve the operating performance.
4.1. HGCHP systems with supplemental heat rejecters
In recent years, a considerable amount of work has been carried
out on the development and application of various HGCHP systems
with a uid cooler, a cooling tower or surface heat rejecters in cool-
ing-dominated buildings. Fig. 4 depicts the operation principle of
the HGCHP system with a cooling tower, where the cooling tower
is connected in series with the GHE loop and is isolated from the
building and ground loops with a plate heat exchanger.
The ASHRAE manual (1995) discussed the advantages of the
HGSHP applications for cooling-dominated buildings considering
initial costs and available surface area limitations [9]. A series of
general guidelines for adding a supplementary heat rejecter are
also given. The capacity of the supplemental heat rejecters is deter-
mined based on the difference between the monthly average cool-
ing and heating loads of a given building. The ground loop is sized
to meet the building heating load, while the cooling load in excess
of the heating load is met through the supplemental heat rejection.
Kavanaugh and Rafferty (1997) have discussed the possibility of
the HGSHP system with a uid cooler as a favorable alternative to
lower the initial cost of the GHE installation [4]. They recom-
mended that the hybrid system be sized based on the peak build-
ing load at the design condition and the capacity of the uid cooler
be calculated according to the difference between the GHE lengths
required for cooling and heating loads. Some recommendations
were made for the integration of the supplemental heat rejecters
into the GCHP piping system.
Kavanaugh (1998) has proposed a revised design method for
sizing uid coolers and cooling towers for hybrid system on the ba-
sis of the design procedures by ASHRAE (1995) and Kavanaugh and
Rafferty (1997) [49]. In addition to sizing the GHE and the cooler,
this revision also provides a method for balancing the heat ow
into the ground on an annual basis. The revised design procedure
also includes a method to determine the annual required operating
hours of the supplemental heat rejecter to balance the annual ther-
mal loads of GHEs. The author designed a HGCHP system for an of-
ce building under three different climates by means of the design
procedure. The results indicate that the economic value of hybrid
systems is most apparent in warm and hot climates where the
building cooling loads are much greater than heating loads.
Yavuzturk and Spitler (2000) have investigated the advantages
and disadvantages of various control strategies for the operation
of a HGSHP system with a cooling tower under different climatic
conditions [50]. The investigated control strategies are broadly cat-
egorized into three groups: (1) set point control for the heat pump
entering or exiting uid temperatures to activate the cooling
tower; (2) differential temperature control to operate the cooling
tower when the difference between the heat pump entering or
exiting temperature and the ambient wet-bulb temperature is
greater than a set value and (3) scheduled control to operate the
cooling tower during the night to accomplish the cool storage in
the ground and avoid a long-term temperature rise. A life-cycle
cost analysis was conducted to compare each operating control
strategy based on a 20-year period. The simulation results for a
small building indicate that the hybrid application appears to have
signicant economic benet compared to the conventional system
and the differential temperature control scheme is found to be the
most benecial choice.
A practical hourly simulation model of the HGCHP system with
a cooling tower was developed with the aim of analyzing and mod-
eling the heat transfer process of its main components on an hour-
by-hour basis by Man Yi et al. [51]. The hourly operation data of
the HGCHP system are calculated by the developed computer pro-
gram based on the hourly simulation model. The impacts of four
different control strategies on performances of two different
HGCHP systems designed for a sample cooling-dominated building
are compared, and an analysis of the system investment consider-
ing the initial and operating cost is conducted based on the hourly
calculation results.
A design and simulation tool for modeling the performance of a
shallow pond as a supplemental heat rejecter with GCHP systems
was developed in detail by Chiasson et al. [52]. On the basis of
the simulation model of the shallow pond, Ramamoorthy et al.
developed a system simulation approach to determine the opti-
mum size of a hybrid system with a cooling pond [53].
4.2. HGCHP systems with hot water supply
From another point of view, domestic hot water (DHW) is nec-
essary for daily life in residential and some commercial buildings.
Traditional hot water supply is usually produced by fossil fuel-red
or electrical boilers, which not only consume a great deal of energy
but emit substantial volume of poisonous gases and greenhouse
gases into the atmosphere. Therefore, to reduce the high initial cost
of the GCHP system and simultaneously decrease the energy con-
sumption for DHW heating, the hybrid GCHP systems with domes-
tic hot water heating becomes another favorable alternative for
some cooling-dominated buildings, especially for the residential
or some commercial buildings with equivalent requirements for
cooling and heating including space heating and DHW heating. A
small, auxiliary heat exchanger, which is called as a desuperheater,
uses superheated gas from the compressor (i.e. the excess conden-
sation heat) to preheat/heat a portion of DHW. In cooling seasons,
the desuperheater uses excess heat that would otherwise be ex-
pelled to the ground to heat domestic water virtually for free. In
heating seasons, more heat can be extracted from the ground to
simultaneously provide space heating and DHW heating, which
can further balance the GHE cooling and heating loads in the
ground. A schematic diagram is shown in Fig. 5. Fig. 4. Schematic diagram of a HGCHP with cooling tower.
24 H. Yang et al. / Applied Energy 87 (2010) 1627
Fanney and Dougherty (1992) rst discussed the performance of
a residential earth-coupled heat pump (i.e. GCHP) with an integral
desuperheater water-heatingcircuit [54]. Therecordeddatashowed
that the desuperheater contributed to an average of 27% of the total
energy supplied for heating water through the 24-month monitor-
ing period. Kavanaugh (1992) reported a similar project of the GCHP
system with a desuperheater and concluded that the cost savings
were considerable based on the utility bill [55]. After that, some
qualitative introduction and general concept on the HGCHP with
hot water supply system were presented by ASHRAE [9].
Cui et al. (2008) developed a hybrid GCHP (HGCHP) with DHW
supply system for space cooling/heating and DHW supply for res-
idential buildings in hot-climate areas [56]. A simulation model
for this hybrid system is established within the HVACSIM+ envi-
ronment. A sample system, applied for a small residential apart-
ment located in Hong Kong, is hourly simulated in a typical
meteorological year. The conventional GCHP systemand an electric
heater for DHW supply are also modeled and simulated on an
hourly basis within the HVACSIM+ for comparison purpose. The re-
sults obtained from this case study show that the HGCHP system
can effectively alleviate the unbalanced loads of the GHE and can
offer almost 95% DHW demand. The energy saving for DHW heat-
ing is about 70% compared with an electric heater. This proposed
scheme, i.e. the HGCHP with DHW supply, is suitable to residential
buildings in hot-climate areas.
4.3. HGCHP systems with solar collectors
In heating-dominated climates, the single GCHP system may
cause a thermal heat depletion of the ground, which progressively
decreases the heat pumps entering uid temperature. As a result,
the system performance becomes less efcient. Similar to the cases
of cooling-dominated buildings, the use of a supplemental heat
supply device, such as a solar thermal collector, can signicantly
reduce the GHE size and the borehole installation cost and, there-
fore, can make the GCHP systems economically attractive. Basi-
cally, the GHE is sized to meet the cooling load and the
supplemental heat supply device is sized to meet the excess heat-
ing load that is unmet by the GHE. Fig. 6 depicts the basic operating
principle of the hybrid GCHP system with a solar collector.
The idea to couple a solar collector to the coil of pipes buried in
the ground, by means of which solar energy can be stored in the
ground, was rst proposed by Penrod in 1956. After a few decades,
the solar ground-source heat-pump (SGSHP) system was recom-
mended by Metz [57]. Recently, a number of efforts have been
made to investigate the performance and applications of the so-
lar-assisted GCHP systems. Chiasson and Yavuzturk presented a
system simulation approach to assess the feasibility of the hybrid
GCHP systems with solar thermal collectors in heating-dominated
buildings [58]. The system simulation approach developed in the
TRNSYS environment was used to model 20-year performance of
the hybrid GCHP as well as the conventional systems for different
cases with different climate conditions. The xed and azimuth-
tracking solar collectors were modeled in this study. Finally, the
20-year life-cycle cost analysis indicates that the hybrid solar-as-
sisted GCHP system is a viable choice for space-conditioning of
heating-dominated buildings. Bi et al. conducted the experimental
studies of a solar-ground heat pump system, where the heating
mode is alternated between a solar energy-source heat pump
and a ground-source heat pump with a vertical double-spiral coil
GHE [59]. Ozgener and Hepbasli experimentally investigated the
performance characteristics of a solar-assisted GCHP system for
greenhouse heating with a vertical GHE [60]. A at-type solar col-
lector was directly installed in series into the ground-coupled loop.
Trillat-Berdal et al. presented a GCHP system coupled with ther-
mal solar collectors, which was designed to provide DHW and
space heating for a private residence [61]. In this system, the solar
heat is used in priority to heat DHW and is injected into the ground
Fig. 5. Schematic diagram of the HGCHP with DHW heating system.
H. Yang et al. / Applied Energy 87 (2010) 1627 25
via the boreholes only when the DHW requirement is met. The
experimental study for the operation performance of the system
indicated that the heat injected into the ground by the solar collec-
tor had an average value of 39.5 W/m, approximately the same as
the average heat extracted fromthe ground by the GCHP (40 W/m).
As a consequence, the heat pump COP was signicantly increased
after the completion of the thermal ground heat recovery by the
injection of solar heat, which demonstrates that recharging the
ground with solar heat is a feasible way to balance the ground
loads in cases of heating-dominated buildings.
A solar-assisted GCHP heating system with latent heat energy
storage tank (LHEST) was investigated by Han et al. [62]. The hy-
brid heating system can implement eight different operation
modes according to the outdoor weather conditions by means of
alternative heat source changes among the solar energy, ground
heat and the latent heat energy storage tank. The operation charac-
teristic of the complex system in heating mode was analyzed using
the numerical simulation method. The simulation results showed
that the GCHP system has an improved COP since the ground tem-
perature can obtain an effective recovery due to the addition of so-
lar energy, while the GCHP can compensate the intermittence and
instability of solar energy. Finally, the authors claimed that the
LHEST can improve the solar fraction of the system, and thus the
COP of the heating system can be increased.
5. Conclusions and recommendations for future work
During the past few decades, a large number of GCHP systems
have been widely applied in various buildings around the world
due to the attractive advantages of high efciency and environ-
mental friendliness. Most typical heat transfer simulation models
currently available for vertical GHEs have been described in detail
in this work, which include two separate regions: one is the heat
transfer process outside the borehole and the other is the heat
transfer inside the borehole. A comparison of these models has
been conducted in terms of their advantages and disadvantages.
The design/simulation methods and programs which are intended
for sizing vertical GHEs and energy analysis of GCHP systems have
been comprehensively reviewed as well. Finally, the various hybrid
GCHP systems for cooling or heating-dominated buildings have
been discussed.
Although much effort has been focused on the development and
application of the GCHP systems, there are still a few areas that
need to be done for future study in order to further broaden the
applicability of the GCHP technology:
Most currently available simulation models can not take into
account the thermal effect of the ground water ow through
the boreholes on the performance of the GCHP systems, where
the thermal interaction is highly sensitive to the sizing of the
GHEs. More effort should be focused on this issue.
The various design computer programs should be further vali-
dated by measured eld data and a comprehensive comparison
is intensively needed to investigate their accuracy and applica-
bility for engineering practices.
An optimal design methodology for the hybrid GCHP systems is
highly needed, which can solve the simultaneous interactions
between the building systems, supplemental heat rejecters/
absorbers and GHEs.
More cost-effective drilling technology for boreholes is needed
for lowering initial cost and improving the life-cycle perfor-
mance of the GCHP systems.
Acknowledgements
The work described in this paper is supported by a grant from
the Sun Hung Kai Properties Group (Project No. ZZ1T) and a grant
from the Research Grants Council of the Hong Kong Special Admin-
istrative Region, China (Project No. PolyU 5332/08E).
References
[1] Ball DA, Fischer RD, Hodgett DL. Design methods for ground-source heat
pumps. ASHRAE Trans 1983;89(2B):41640.
[2] Ingersoll LR, Plass HJ. Theory of the ground pipe source for the heat pump.
ASHVE Trans 1948;54:33948.
[3] IGSHPA. Closed-loop/ground-source heat pump systems Installation guide;
1988.
[4] Kavanaugh SP, Rafferty K. Ground-source heat pumps, design of geothermal
systems for commercial and institutional buildings. Atlanta: ASHRAE; 1997.
[5] Bose JE, Parker JD, McQuiston FC. Design/data manual for closed-loop ground
coupled heat pump systems. Oklahoma State Univ for ASHRAE; 1985.
[6] Eskilson P. Thermal analysis of heat extraction boreholes. Ph.D. thesis. Sweden:
University of Lund; 1987.
[7] Bose JE, Smith MD, Spitler JD. Advances in ground source heat pump systems
an international overview. In: Proceedings of the seventh international energy
agency heat pump conference, Beijing; 2002: 1. p. 31324.
[8] ASHRAE. ASHRAE handbook, HVAC applications. Atlanta: American Society of
Heating Refrigeration and Air-Conditioning Engineers Inc.; 1999.
[9] ASHRAE. Commercial/institutional ground-source heat pump engineering
manual. Atlanta: American Society of Heating, Refrigeration and Air-
Conditioning Engineers. Inc.; 1995.
[10] Spitler JD. Ground-source heat pump system research past, present and future.
HVAC R Res 2005;11(2):1657.
[11] Rawlings RHD, Sykulski JR. Ground source heat pumps: a technology review.
Build Serv Eng Res Technol 1999;20(3):11929.
[12] Georgios Floridesa, Soteris Kalogirou. Ground heat exchangersA review of
systems, models and applications. Renew Energy 2007;32(15):246178.
[13] Ingersoll LR, Adler FT, Plass HJ, Ingersoll AC. Theory of earth heat exchangers
for the heat pump. ASHVE Trans 1950;56:16788.
[14] Ingersoll LR, Zobel OJ, Ingersoll AC. Heat conduction with engineering,
geological, and other applications. New York: McGraw-Hill; 1954.
[15] Hart DP, Couvillion R. Earth coupled heat transfer. Publication of the National
Water Well Association; 1986.
[16] Fang ZH, Diao NR, Cui P. Discontinuous operation of geothermal heat
exchangers. Tsinghua Sci Technol 2002;7(2):1947.
[17] Hackner RJ, Hughes PJ, ONeil RA. Design of ECHP Systems in northern
climates. ASHRAE Trans 1987:93.
[18] Carslaw HS, Jaeger JC. Conduction of heat in solids. Oxford UK: Claremore
Press; 1946.
[19] Kavanaugh SP. Simulation and experimental verication of vertical
groundcoupled heat pump systems. Ph. D dissertation. Stillwater, Oklahoma:
Oklahoma State University; 1985.
[20] Deerman JD, Kavanaugh SP. Simulation of vertical U-tube ground coupled heat
pump systems using the cylindrical heat source solution. ASHRAE Trans
1991;97(1):28795.
[21] Bernier MA. Ground-coupled heat pump system simulation. In: ASHRAE
winter meeting CD, technical and symposium papers. ASHRAE; 2001. p. 739
50.
[22] Hellstrom G. Ground heat storage: Thermal analyses of duct storage
systems. Sweden: Department of Mathematical Physics University of Lund;
1991.
[23] Liu XL, Wang DL, Fang ZH. Modeling of heat transfer of a vertical bore in
ground-source heat pumps. J Shandong Inst Architect Eng 2001(1):4751.
[24] Zeng HY, Diao NR, Fang ZH. A nite line-source model for boreholes in
geothermal heat exchangers. Heat Transfer Asian Res 2002;31(7):55867.
Fig. 6. Schematic diagram of a HGCHP system with solar collector.
26 H. Yang et al. / Applied Energy 87 (2010) 1627
[25] Yu MZ, Diao NR, Su DC, Fang ZH. A pilot project of the closed-loop ground-
source heat pump system in China. In: Proceeding of IEA 7th heat pump
conference. Beijing; 2002. p. 35664.
[26] Cui P, Yang HX, Fang ZH. The simulation model and design optimization of
ground source heat pump systems. HKIE Trans 2007;14(1):15.
[27] Diao NR, Zeng HY, Fang ZH. Improvement in modeling of heat transfer in
vertical ground heat exchangers. HVAC & R Res 2004;10(4):45970.
[28] Yavuzturk C, Spitler JD, Rees SJ. A Transient two-dimensional nite volume
model for the simulation of vertical U-tube ground heat exchangers. ASHRAE
Trans 1999;105(A):46574.
[29] Yavuzturk C, Spitler JD. Field validation of a short time step model for vertical
ground-loop heat exchangers. ASHRAE Trans 2001;107(1):61725.
[30] Klein SA et al. TRNSYS Manual, a transient simulation program. Madison: Solar
Engineering Laboratory, University of Wisconsin-Madison; 1996.
[31] Hellstrom G. Duct ground heat storage model manual for computer
code. Sweden: Department of Mathematical Physics, University of Lund; 1989.
[32] Thornton JW, McDowell TP, Shonder JA, Hughes PJ, Pahud D, Hellstrom G.
Residential vertical geothermal heat pump system models: calibration to data.
ASHRAE Trans 1997;103(2):66074.
[33] Yavuzturk C. Modeling of vertical ground loop heat exchangers for
ground source heat pump systems. PhD thesis. USA: Oklahoma State
University; 1999.
[34] Muraya NK, ONeal DL, Hefngton WM. Thermal interference of adjacent legs
in a vertical U-tube heat exchanger for a ground-coupled heat pump. ASHRAE
Trans 1996;102(2):1221.
[35] Rottmayer SP, Beckman WA, Mitchell JW. Simulation of a single vertical U-
tube ground heat exchanger in an innite medium. ASHRAE Trans
1997;103(2):6519.
[36] Li Zhongjian, Zheng Maoyu. Development of a numerical model for the
simulation of vertical U-tube ground heat exchangers. Appl Therm Eng
2009;29(5-6):9204.
[37] Gu Y, ONeal DL. Development of an equivalent diameter expression for
vertical U-Tubes used in ground-coupled heat pumps. ASHRAE Trans
1998;104:34755.
[38] Zeng HY, Diao NR, Fang ZH. Efciency of vertical geothermal heat exchangers
in ground source heat pump systems. J Therm Sci 2003;12(1):7781.
[39] Cui P, Yang H, Fang Z. Numerical analysis and experimental validation of heat
transfer in ground heat exchangers in alternative operation modes. Energy
Build 2008;40(6):10606.
[40] Claesson J, Eskilson P, Hellstrom G. PC design model for heat extraction
boreholes. In: Proc. 3rd WS on SAHPGCS. Gteborg, CIT; 1990: 3. p 99102.
[41] Claesson J. PC design model for thermally interacting deep ground heat
exchangers. In: IEA heat pump centre report HPC-WR-8; 1991. p 95104.
[42] Hellstrom G, Sanner B. Software for dimensioning of deep boreholes for heat
extraction. In: Proc. 6th int. conf. energy storage CALORSTOCK 94; 1994. p.
195202.
[43] Hellstrom G, Sanner B. PC-programs and modeling for borehole heat
exchanger design. In: Proc. intern. geothermal days Germany, Gtv, Geeste,
Supplement; 2001. p. 3544.
[44] Spitler JD. GLHEPRO A design tool for commercial building ground loop heat
exchangers. In: Proceedings of the fourth international heat pumps in cold
climates conference, Aylmer, Qubec; August 178, 2000.
[45] Fisher DE, Murugappan A, Padhmanabhan SK, Rees SJ. Implementation and
validation of ground-source heat pump system models in an integrated
building and system simulation environment. HVAC&R Res 2006;12(3a):
693710.
[46] Liu X. Enhanced design and energy analysis tool for geothermal water loop
heat pump systems. In: Proceedings of 9th international energy agency heat
pump conference. Zrich, Switzerland; May 2022, 2008.
[47] Khan MH, Varanasi A, Spitler JD, Fisher DE, Delahoussaye RD. Hybrid ground
source heat pump system simulation using visual modeling tool for
HVACSIM+. In: Eighth international ibpsa conference. Eindhoven,
Netherlands; August 1114, 2003.
[48] Pahud D, Hellstrom G. The new duct ground heat model for TRNSYS. In: Proc.
eurotherm seminar 49, Eindhoven; 1996. p. 12736.
[49] Kavanaugh SP. A design method for hybrid ground-source heat pumps.
ASHRAE Trans 1998;104(2):6918.
[50] Yavuzturk C, Spitler JD. Comparative study of operating and control strategies
for hybrid ground-source heat pump systems using a short time step
simulation model. ASHRAE Trans 2000;106:192209.
[51] Man Yi, Yang Hongxing, Fang Zhaohong. Study on hybrid ground-coupled heat
pump systems. Energy Build 2008;40(11):202836.
[52] Chiasson AD, Spitler JD, Rees SJ, Smith MD. Model for simulating the
performance of a shallow pond as a supplemental heat rejecter with
closed-loop ground-source heat pump systems. ASHRAE Trans 2000;106:
10721.
[53] Ramamoorthy M, Jin H, Chiasson AD, Spitler JD. Optimal sizing of hybrid
ground-source heat pump systems that use a cooling pond as a supplemental
heat rejecter a system simulation approach. ASHRAE Trans
2001;107(PART1):2638.
[54] Fanney AH, Dougherty BP. Performance of a residential desuperheater.
ASHRAE Trans 1992;98(1):48999.
[55] Kavanaugh SP. Field test of a vertical ground-coupled heat pump in Alabama.
ASHRAE Trans 1992;98(pt 2):60715.
[56] Cui P, Yang HX, Spitler JD, Fang ZH. Simulation of hybrid ground-coupled heat
pump with domestic hot water heating systems using HVACSIM+. Energy
Build 2008;40(9):17316.
[57] Metz PD. The use of ground-coupled tanks in solar-assisted heat-pump
systems. ASME Trans. J Sol Energy Eng 1982;104(4):36672.
[58] Chiasson AD, Yavuzturk C. Assessment of the viability of hybrid geothermal
heat pump systems with solar thermal collectors. ASHRAE Trans
2003;109(PART 2):487500.
[59] Bi, Yuehong, Tingwei Guo, Liang Zhang, Lingen Chen. Solar and ground source
heat-pump system. Appl Energy 2004;78:23145.
[60] Ozgener O, Hepbasli A. Performance analysis of a solar assisted ground-source
heat pump system for greenhouse heating: An experimental study. Build
Environ 2005;40(8):104050.
[61] Trillat-Berdal V, Souyri B, Fraisse G. Experimental study of a ground-coupled
heat pump combined with thermal solar collectors. Energy Build
2006;38(12):147784.
[62] Han Zongwei, Zheng Maoyu, Kong Fanhong, Wang Fang, Li Zhongjian, Bai Tian.
Numerical simulation of solar assisted ground-source heat pump heating
system with latent heat energy storage in severely cold area. Appl Therm Eng
2008;28(11-12):142736.
H. Yang et al. / Applied Energy 87 (2010) 1627 27