SSRN Id4355816

1 Development and application of a new calculation method for
ed
2 double spiral ground heat exchangers
iew
3 Kunning Yanga*, Takao Katsurab, Shigeyuki Nagasakac*, Katsunori Naganob*
4 aGraduate School of Engineering, Hokkaido University, Sapporo, 060-8628, Japan
5 bFaculty of Engineering, Hokkaido University, Sapporo, 060-8628, Japan
6 c Shin Nippon Air Technologies Co., Ltd., Chino, 391-0013, Japan
v
7
re
8 *Kunning Yang. Tel.: +81-011-706-6287; fax: +81-011-706-6287.
9 E-mail address: kunning.yang.h4@elms.hokudai.ac.jp
10 Abstract er
11 This study provides a new calculation method for double spiral pipe ground heat exchangers (GHEs) and
12 involves the combination of the analytical, infinite line source (ILS), and infinite cylindrical source (ICS)
pe
13 models with the capacity resistance model (CaRM). Method adopts the concept of fin efficiency, enabling it
14 to calculate underground temperature changes easily and accurately, followed by integration into a simulation
15 tool for ground source heat pump (GSHP) systems. Verification was done by comparing simulation results
ot
16 with actual operating data, measured from a zero-energy building (ZEB) in Sapporo, Japan. Analyses
17 conducted using the new calculation method prove that simulation results can be obtained with increased
tn
18 precision and validity. Under conditions of instantaneous large heat injection, significant differences were
19 observed between the simulation and the actual measurements during the initial 50 h. The number of
20 temperature nodes in the network was increased to effectively reduce the error between simulations and
rin
21 measurements. Moreover, it was seen that the efficiency of double spiral pipe GHEs can be potentially
22 improved by shortening the spiral pitch. This study thus confirms that the proposed new calculation method
23 can be applied to future designs of GSHP systems by using thermal piles with double spiral pipe GHEs.
ep
24
25 Keywords: ground source heat pump system; double spiral pipe ground heat exchangers; capacity resistance
26 model; thermal pile; simulation

Pr
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4355816
27 1. Introduction
ed
28 Building energy consumption accounts for approximately 40% of the total global energy consumption [1]
29 of which space cooling and heating alone account for 50%–60% of the building operations [2]. To substantially
30 reduce greenhouse gas emissions, energy supplies for space cooling and heating must be shifted from fossil
iew
31 fuels to renewable energy sources. Ground source heat pump (GSHP) systems have gained popularity
32 worldwide in recent years [Fig. 1] [3]. The ground heat exchangers (GHEs) of the GSHP system use circulating
33 fluids such as ethylene glycol, which can inject heat into the ground in summers and absorb heat from the
34 ground in winters, thus providing space heating and cooling for buildings through heat pump units [Fig. 2 (a)].
v
35 Compared to the other commonly used renewable energy technologies, GSHP systems have better thermal
re
36 efficiency and reliability owing to the continuity of operations [4] [5].
37 However, GSHPs involve high installation costs. The previously used borehole U-shaped pipe GHEs had a
38 drilling depth of more than 100 m, and the drilling cost is high. Whereas the recently introduced large-scale
er
39 GSHP systems required larger areas for installation of multiple GHEs. Thermal piles have thus been adopted
40 in recent years to tackle these problems. Thermal pile is a type of energy structure that incorporates the GHEs
pe
41 of GSHP systems through the foundation elements of the building. Thermal piles can not only support the
42 structure of the building, but also serve as GHEs [Fig. 2(b)] [6]. Moreover, they are extremely cost effective
43 given that their installation does not require additional land and is at much smaller depths than that of the U-
44 tube GHEs. Furthermore, as compared to the other commonly used boreholes, they have larger radius due to
ot
45 their load-bearing function. In recent years, many countries, such as the United Kingdom and Japan, have been
46 increasingly utilizing the thermal piles; there are more than 300,000 thermal piles currently under construction
tn
47 in Japan [Fig. 3, Fig. 4] [7] [8].

rin
ep
Pr
48
49 Fig. 1. The installed capacity and annual utilization of GSHP systems from 1995-2020
2
50
ed
iew
51
52 Fig. 2. (a) Operation of GSHP system with borehole U-tube GHEs; (b) Buildings with GSHP system using thermal piles and double
v
53 spiral pipe GHEs
54
re
er
pe
55
56 Fig. 3. The resultant annual CO2 savings from thermal piles installed in the UK
57
ot
tn
rin
58
59 Fig. 4. Categories of thermal piles being used in Japan, 2019
60 The heat capacitance of the grouting material in the thermal piles greatly affects their thermal performance.
ep
61 Thermal piles can be classified into three categories based on the pile material and the heat exchanger pipes
62 used [9]: cast in-situ concrete, pre-stressed high-strength concrete (PHC), and steel piles. The GHEs used can
Pr
63 be further classified into three common types: single U-shaped, double U-shaped, and spiral pipes.
64 The focus of this study was to analyze the double spiral pipe GHEs [Fig. 5 (c)]. The advantage of using
65 double spiral pipe GHEs is that the depth of thermal piles can be further shortened, thereby saving on drilling
3
66 costs. Moreover, for same depths, the length of the double spiral pipe GHE is much longer than that of the U-
67 shaped pipe GHE, meaning there is a larger area available for heat transfer between underground soil and
ed
68 circulating fluid through the pipe surface, making it more energy efficient [Fig. 5 (d)].
iew
69
v
70 Fig. 5. GHE configurations: (a) single U-shaped; (b) double U-shaped; (c) spiral (d) Comparison between U-shaped and double spiral
re
71 pipe GHEs
72 The “geothermal tornado method” developed by Japan Pile Corporation eliminated the installation problems
73 posed by double spiral pipe GHEs by making them safe and convenient to use. They not only reduced the
74
75
er
installation time and labor, but also the risk of damage, drilling work, and initial costs. Figure 6 illustrates how
the construction of double spiral pipe GHEs is more cost effective as compared to the conventional GHEs [10].
pe
ot
tn
76
77 Fig. 6. Cost reduction by using thermal pile and double spiral pipe GHE
78 Determination of the required size of the GHEs is of great importance prior to the construction of the GSHP
rin
79 system. Smaller GHE systems are not able to provide high thermal load for the building and may fail shortly
80 or eventually as the operations begin. whereas larger GHEs involve high costs. Therefore, the models for sizing
81 of the GHEs have been under discussion for a long time.

ep
82 This study focuses mainly on double spiral pipe GHEs, which in previous studies have been extensively
83 studied, and the main calculation methods can be classified into three categories: numerical, analytical, and
Pr
84 CaRM model.
85  Numerical model: Albetro et al. analyzed the heat transfer performance of the thermal piles using
86 GHEs of different configurations (U-shaped, double U-shaped, Triple U-shaped, and spiral pipes)
4
87 [11]. The obtained results show that the spiral pipe has the best heat transfer performance. Zhao et al.
88 conducted comparative analyses of different types of GHEs in thermal piles and estimated that spiral-
ed
89 shaped GHEs have better thermal performance than the other two, both for long-term and short-term
90 operations [12]. Simulation results obtained from a 3D model of the double spiral pipe GHEs built by
iew
91 Serageldin, A. A. et al. using ANSYS workbench revealed that the double spiral pipe GHEs had better
92 thermal efficiency, higher heat transfer rate, and lower thermal resistance under turbulent flow
93 conditions than those with single U-shaped pipes [13].
94  Analytical model: A transient ring-coil heat source model proposed by Cui et al. to study the transient
v
95 heat conduction around the buried spiral pipe failed to consider the size of the pipe, causing the
re
96 simulated temperatures to often deviate from the actual conditions [14]. Improvements were made by
97 Zhang et al. to simplify the calculations by taking into account the arrangement and pitch of the spiral
98 pipe [15]. This new model, however, cannot be applied if the grouting material differs from that of
99
er
the surrounding soil. Later, an efficient spiral coil source model along with its analytical solution was
100 proposed by Park et al., but its long-term operational efficiency is yet to be verified [16]. A solid
pe
101 cylindrical source model of the pile foundation that considered both the radial dimension and heat
102 capacity of the borehole or pile was then suggested by Man et al. [17] [18]. The analytical solution
103 expressions for 1D and 2D new models were constructed by Li et al. for the model using the Green’s
104 function method, which resolved the problem of heat conduction in infinite or semi-infinite
ot
105 anisotropic media with a line, spiral-line, or cylindrical-surface heat sources; the solution derived new
106 temperature response functions for pile GHEs with spiral coils and borehole GHEs with single or
tn
107 double U-shaped pipes [19][20].
108  CaRM model: The capacity resistance model (CaRM) was first applied by Zarrella et al. to analyze
109 the heat transfer in thermal piles [21]. A comparative analysis was performed between spiral pipe and
rin
110 double U-shaped pipe GHEs, considering long- and short-term operations. For long-term operations,
111 although the spiral pipe GHEs were shorter than the double U-shaped pipe GHEs by 33%, the amount
ep
112 of heat exchanged with the ground remained the same, indicating that the spiral pipe GHE can be
113 utilized to reduce installation costs. During short-term operations, spiral pipe GHE allowed for higher
114 peak loads than double U-shaped pipe GHE owing to the greater heat capacity of thermal pile.
Pr
115 Moreover, field experiments and numerical analyses were used to compare the spiral pipe and triple
116 U-shaped pipe configurations inside a foundation pile [22]; the spiral pipe exhibited higher thermal
117 performance, with an increase in peak value of approximately 23%.
ed
118 Despite their functionality, these models pose some problems. Few hours of simulation can be
119 computationally intensive for numerical models, making them unsuitable for long-term simulations. Most of
iew
120 the analytical models available assume the materials of grouting, pile body, and ground soil to be the same,
121 making them inappropriate in cases where the materials differ. Interestingly, though most analytical models
122 focus on calculating the temperatures at one point, they take a long time to obtain the results if average
123 temperatures at multiple points are calculated.
v
124 Previous research [23] proposed a simulation tool combining the CaRM with analytical model, which
re
125 improved the precision but was only applicable to double spiral pipe GHE with pitch (spiral distance) lower
126 than 150 mm. The current study addresses this problem by introducing a new calculation method for the heat
127 transfer of thermal piles using double spiral pipe GHEs for varied pitch sizes. This method is based on the
128
er
CaRM and adopts the concept of fin efficiency. A new simulation tool developed by combining this calculation
129 method (for temperature change inside the thermal pile) with a ground temperature simulation program for
pe
130 GSHP systems (for temperature change outside the thermal pile) was found to be suitable for temperature
131 simulations of thermal piles with double spiral pipe GHEs; the simulation accuracy was reported to remain
132 unaffected by the value of the spiral pitch, thus allowing the material differences between grouting and soil.
133 We believe that this simulation tool will be helpful in designing the GSHP systems using thermal piles with
ot
134 double spiral pipe GHEs.

tn
135 2. Methodology
136 2.1. CaRM for double spiral pipe GHEs

137 The larger diameters of the thermal pile than the common value of the borehole make it essential to consider
rin
138 the heat capacity of the grouting material when studying the thermal performance of double spiral pipe GHE.
139 Previous works [24] presented a numerical model, CaRM-He, to solve this problem and analyzed the thermal
140 behavior of a short-length spiral pipe GHE. CaRMs are derived from the lumped thermal capacity model [25],
ep
141 a simple transient heat conduction approach, in which the temperature of the solid is assumed to be spatially
142 independent and time-dependent, assuming no temperature variations within the solid. In the CaRM-He model,
143 temperature nodes are arranged both vertically and radially across the spiral pipe, grouting, and ground. Higher
Pr
144 precisions demand multiple temperature nodes, which in turn hinder simulation performance due to large scale
145 calculations.
146 This study also employed the CaRM to account for the thermal capacitance of grouting material;
147 temperature nodes were arranged in the radial direction to reduce the amount of calculation while obtaining
ed
148 acceptable calculation accuracy. Figure 7 illustrates the top view of a thermal pile in a large-diameter borehole.
149 The volume of the thermal pile is divided into three parts:
iew
150  “core” part: grouting material
151  double spiral pipe
152  “shell” part: thermal pile material between spiral pipe and borehole boundary
153 The temperature nodes were set at the following locations: 1. between the center and inner surface of the
v
154 pipe 𝑇𝑐𝑜𝑟𝑒; 2. the inner and outer surface of a double spiral pipe 𝑇𝑝,𝑠𝑝; 3. the circulating fluid inside the pipe
re
155 𝑇𝑓,𝑠𝑝; 4. between outer surface and the boundary of the borehole 𝑇𝑠ℎ𝑒𝑙𝑙; 5. boundary of the borehole 𝑇𝑔 [26].
156 The heat transfer process inside a thermal pile can be divided into three parts: heat conduction in the “core”
157 part, heat convection between flowing circulating fluid and pipe surface, and heat conduction in the “shell”
er
158 part [23] [24].
pe
ot
159
160 Fig. 7. (a) Sectional view of thermal pile, (b) Top view of thermal pile, and (c) Partial detail of thermal pile
tn
161 As the outer diameter of the pipe 𝑑𝑝,𝑜 is much smaller than the borehole diameter, the double spiral pipe
162 was considered as a line that injects/absorbs heat into/from the ground in summers/winters. Assuming this, the
163 heat transfer can be simplified to one-dimensional steady-state heat transfer in a multi-layer cylindrical wall
rin
164 [24]. The thermal resistances between “core” part and double spiral pipe, double spiral pipe and “shell” part,
165 and “shell” part and surrounding soil can be respectively expressed in form of equations as under [Fig. 8]:
166 𝑅𝑠𝑝 ‒ 𝑐 = 2𝜋 ∙ 𝑙
1
( )
𝑟𝑠𝑝,𝑖
∙ 𝑙𝑛 𝑟 (1)
ep
𝑏 ∙ 𝜆𝑐𝑜𝑟𝑒 𝑐𝑜𝑟𝑒
167 𝑅𝑠𝑝 ‒ 𝑠 = 2𝜋 ∙ 𝑙
1
𝑏 ∙ 𝜆𝑠ℎ𝑒𝑙𝑙
( )
∙ 𝑙𝑛
𝑟𝑠ℎ𝑒𝑙𝑙
𝑟𝑠𝑝,𝑜 (2)
1
( )
𝑟𝑏
Pr
168 𝑅𝑠 ‒ 𝑏 = 2𝜋 ∙ 𝑙
𝑏 ∙ 𝜆𝑠ℎ𝑒𝑙𝑙
∙ 𝑙𝑛 𝑟𝑠ℎ𝑒𝑙𝑙 (3)
169 where 𝑙𝑏 is the depth of the thermal pile, 𝜆𝑐𝑜𝑟𝑒 and 𝜆𝑠ℎ𝑒𝑙𝑙 are the thermal conductivity of grouting and pile
170 materials, respectively; 𝑟𝑐𝑜𝑟𝑒 and 𝑟𝑠ℎ𝑒𝑙𝑙 are the distances from center of the circle to the temperature nodes
ed
171 of “core” and “shell” parts, respectively; 𝑟𝑏 is the radius of thermal pile; 𝑟𝑠𝑝,𝑖 and 𝑟𝑠𝑝,𝑜 are the distances
172 from the center of the circle to the inner surface and the outer surface of the pipe, respectively.
v iew
173
re
174 Fig. 8. Sectional drawing of the thermal pile
175 As this study only considers the heat transfer in the horizontal direction, it is necessary to consider spiral
176 pitch 𝑙𝑝𝑖𝑡𝑐ℎ in the vertical direction and the outer diameter of the pipe 𝑑𝑝,𝑜. The vertical section of the double
er
177 spiral pipe GHE is similar to a “vertical” floor heating panel [Fig. 8], which distributes heat to the “core” and
178 the “shell” parts. The value of the heat transfer rate can be calculated using the Kollmar-Liese method, which
pe
179 adopts the concept of fin efficiency and is widely used for designing the hot water floor heating systems.
180 2.2. Method for calculating the heat transfer using fin efficiency
181 2.2.1. Calculation for heat transfer rate using Kollmar-Liese method
ot
182 Previous studies proposed a calculation method for the thermal emission of flat floor heating panels with a
tn
183 cylindrical heat source [27]. They assumed that a virtual fin with a width (𝑑) equal to the outer diameter (𝑑𝑝,𝑜)
184 of the pipe is attached to the pipe [Fig. 9 (b)]. 𝛼 is the convective heat transfer coefficient between the floor
185 and surrounding air, while 𝜆 is the thermal conductivity of the floor material.
rin
ep
186
Pr
187 Fig. 9. (a) A typical radiant floor heating system (b) Principle of Kollmar-Liese method (c) Similarity between floor heating panel and
188 double spiral pipe GHE
189 Figure 9 (b) depicts the combination of a virtual fin and a pipe, where the fin efficiency 𝜂 of the virtual fin
190 can be calculated using following equation:
ed
2𝛼 𝑙𝑝𝑖𝑡𝑐ℎ ‒ 𝑑𝑝,𝑜 𝑡𝑎𝑛ℎ (𝑚 ∙ ℎ)
191 𝑚= 𝜆∙𝑑
;ℎ = 2
;𝜂 = 𝑚 ∙ ℎ (4)
192 Assuming that the surface temperature of the pipe is 𝑇𝐻 and the ambient temperature is 𝑇𝐿, the mean
iew
193 temperature 𝑇𝑚 of the virtual fin can be calculated using the Kollmar-Liese method [28]:
194 𝑇𝑚 = 𝑇𝐿 + 𝜂 ∙ (𝑇𝐻 ‒ 𝑇𝐿) (5)
195 The heat transfer rate from lower floor to upper air can be given by the equation:
𝑄 = 𝛼 ∙ 𝜂 ∙ (𝑇𝐻 ‒ 𝑇𝐿) = 𝛼 ∙ (𝑇𝑚 ‒ 𝑇𝐿)
v
196 (6)
re
197 2.2.2. Calculations for temperature of approximate line using Kollmar-Liese method
198 The vertical section of the double spiral pipe GHE resembles a “vertical” floor heating panel [Fig. 9 (c)].
199 This part can also be seen as an approximate line source (ignoring the width) that injects heat into the ground
200
er
in summers and extracts heat from it in winters. By analogy, the temperature 𝑇𝑝𝑛𝑙 of the approximate line
201 (corresponding to the mean temperature 𝑇𝑚 of the virtual fin in Fig. 9 (b)) can be expressed by the surface
pe
202 temperature 𝑇𝑝,𝑠𝑝 of the spiral pipe (corresponding to the surface temperature 𝑇𝐻 of the pipe) and the
203 temperature in “core” and “shell” parts (corresponding to the ambient temperature 𝑇𝐿). The relations between
204 these temperatures can be expressed as:
205 𝑇𝑝𝑛𝑙 = 𝑇𝑐𝑜𝑟𝑒 + 𝜂𝑐𝑜𝑟𝑒 ∙ (𝑇𝑝,𝑠𝑝 ‒ 𝑇𝑐𝑜𝑟𝑒) (7)

ot
206 𝑇𝑝𝑛𝑙 = 𝑇𝑠ℎ𝑒𝑙𝑙 + 𝜂𝑠ℎ𝑒𝑙𝑙 ∙ (𝑇𝑝,𝑠𝑝 ‒ 𝑇𝑠ℎ𝑒𝑙𝑙) (8)
207 Combining Equation (7) and (8), the formula for temperature of the approximate line 𝑇𝑝𝑛𝑙 can be expressed
tn
208 as:
1 1
209 𝑇𝑝𝑛𝑙 = 2[(𝑇𝑝,𝑠𝑝 ‒ 𝑇𝑐𝑜𝑟𝑒) ∙ 𝜂𝑐𝑜𝑟𝑒 + 𝑇𝑐𝑜𝑟𝑒] + 2[(𝑇𝑝,𝑠𝑝 ‒ 𝑇𝑠ℎ𝑒𝑙𝑙) ∙ 𝜂𝑠ℎ𝑒𝑙𝑙 + 𝑇𝑠ℎ𝑒𝑙𝑙] (9)
rin
210 2.2.3. Calculation of fin efficiency
211 Using parameters of the double spiral pipe GHE, the fin efficiency to calculate the heat transfer rate from
ep
212 the approximate line to “core” part 𝜂𝑐𝑜𝑟𝑒 and “shell” part 𝜂𝑠ℎ𝑒𝑙𝑙 can be expressed by Equation (4). The
213 convective heat transfer coefficient, 𝛼𝑐𝑜𝑟𝑒, between “vertical” floor heating panel and “core” part, 𝛼𝑠ℎ𝑒𝑙𝑙
214 between “vertical” floor heating panel and “shell” part, are calculated using the thermal resistance 𝑅𝑠𝑝,𝑐, 𝑅𝑠𝑝,𝑠,
Pr
215 the heat transfer area to “core” part 𝐴𝑐𝑜𝑟𝑒 and to “shell” part 𝐴𝑠ℎ𝑒𝑙𝑙:
𝑅𝑠𝑝 ‒ 𝑐 𝑅𝑠𝑝 ‒ 𝑐
216 𝛼𝑐𝑜𝑟𝑒 = 𝐴𝑐𝑜𝑟𝑒
= 2𝜋 ∙ 𝑙
𝑏 ∙ 𝑟𝑠𝑝,𝑖
(10)
9
𝑅𝑠𝑝 ‒ 𝑠 𝑅𝑠𝑝 ‒ 𝑠
217 𝛼𝑠ℎ𝑒𝑙𝑙 = 𝐴𝑠ℎ𝑒𝑙𝑙
= 2𝜋 ∙ 𝑙
𝑏 ∙ 𝑟𝑠𝑝,𝑜
(11)
ed
218 Subsequently, the fin efficiency in two directions can be expressed as:
( )
2𝛼𝑐𝑜𝑟𝑒 𝑙𝑝𝑖𝑡𝑐ℎ ‒ 𝑑𝑝,𝑜
𝑡𝑎𝑛ℎ 𝜆𝑐𝑜𝑟𝑒 ∙ 𝑑𝑝,𝑜
∙ 2
219 𝜂𝑐𝑜𝑟𝑒 = 2𝛼𝑐𝑜𝑟𝑒 𝑙𝑝𝑖𝑡𝑐ℎ ‒ 𝑑𝑝,𝑜 (12)
iew
𝜆𝑐𝑜𝑟𝑒 ∙ 𝑑𝑝,𝑜
∙ 2
( )
2𝛼𝑠ℎ𝑒𝑙𝑙 𝑙𝑝𝑖𝑡𝑐ℎ ‒ 𝑑𝑝,𝑜
𝑡𝑎𝑛ℎ 𝜆𝑐𝑜𝑟𝑒 ∙ 𝑑𝑝,𝑜
∙ 2
220 𝜂𝑠ℎ𝑒𝑙𝑙 = 2𝛼𝑠ℎ𝑒𝑙𝑙 𝑙𝑝𝑖𝑡𝑐ℎ ‒ 𝑑𝑝,𝑜 (13)

𝜆𝑐𝑜𝑟𝑒 ∙ 𝑑𝑝,𝑜
∙ 2
221 According to Equation (6), the heat transfer rate (W) of the circulating fluid through the pipe surface to the
v
222 “core” and “shell” parts can be respectively expressed by following equations:
re
223 𝑄𝑐𝑜𝑟𝑒 = 𝛼𝑐𝑜𝑟𝑒 ∙ 𝐴𝑐𝑜𝑟𝑒 ∙ (𝑇𝑝𝑛𝑙 ‒ 𝑇𝑐𝑜𝑟𝑒) (14)
224 𝑄𝑠ℎ𝑒𝑙𝑙 = 𝛼𝑠ℎ𝑒𝑙𝑙 ∙ 𝐴𝑠ℎ𝑒𝑙𝑙 ∙ (𝑇𝑝𝑛𝑙 ‒ 𝑇𝑠ℎ𝑒𝑙𝑙) (15)
225
er
2.2.4. Calculations for convective thermal resistance
226 On obtaining the temperature of the approximate line 𝑇𝑝𝑛𝑙, it is necessary to discern the relationship
pe
227 between the pipe surface temperature 𝑇𝑝, 𝑠𝑝 and the circulating fluid temperature 𝑇𝑓, 𝑠𝑝. The influence of
228 centrifugal force on fluid flow in the spiral pipe must be considered so as to calculate the convective heat
229 transfer coefficient 𝛼𝑓 between the circulating fluid and the pipe surface. The Reynolds number 𝑅𝑒 and
ot
230 Prandtl number 𝑃𝑟 of the circulating fluid flowing in double spiral pipe can be calculated using its
231 thermophysical properties, density 𝜌, kinematic viscosity ν, and thermal diffusivity 𝛼.

tn
232 Based on the 𝑅𝑒 and 𝑃𝑟 obtained, the previous studies expressed the Nusselt number 𝑁𝑢 under two
233 conditions:
234  Previous works always calculated the Nusselt number for laminar flow using equations for straight
rin
235 pipes. During this study however, it was seen that, the centrifugal force generated by the circulating
236 fluid flow in the spiral pipe had a significant influence on the Reynolds number. As such, Dravid’s
237 formula [29] was adopted for single-phase laminar flow in spiral pipes:
ep
𝑑𝑝,𝑜
238 𝐷𝑒 = 𝑅𝑒 ∙ 𝑑𝑠𝑝,𝑜
, 𝑁𝑢 = (0.76 + 0.65𝐷𝑒0.5) ∙ 𝑃𝑟0.175 (16)
239  Gnielinski’s formula was used for turbulent flow [30][31]:

Pr
(𝑓/8) ∙ (𝑅𝑒 ‒ 1000) ∙ 𝑃𝑟

240 𝑓 = 4𝐶𝑓 = 4(0.0791𝑅𝑒 ‒ 1/4), 𝑁𝑢 = (17)
1 + 12.7(𝑓/8)1/2 ∙ (𝑃𝑟2/3 ‒ 1)
10
241 The results show that as the Nusselt number calculated by the Dravid’s formula is greater than that by the
242 Gnielinski’s formula, it can neither be applied for laminar flow nor the turbulent flow region.
ed
v iew
re
243
244 Fig. 10. Nusselt number calculated for laminar and turbulent flow
245 Then the convective heat transfer coefficient 𝛼𝑓 can be expressed as:
𝑁𝑢𝜆𝑓
246
er 𝛼𝑓 = 𝑑𝑝,𝑖 (18)
247 where 𝜆𝑓 is the thermal conductivity of the fluid, 𝑑𝑝,𝑜 and 𝑑𝑝,𝑖 are outer and inner diameters of the pipe.
pe
248 The length of double spiral pipe GHE 𝑙𝑠𝑝 can be simply calculated as:
𝑙𝑏 𝑑𝑠𝑝,𝑖 + 𝑑𝑠𝑝,𝑜 2
249 𝑙𝑠𝑝 = 𝑙
𝑝𝑖𝑡𝑐ℎ
2
∙ 𝑙𝑝𝑖𝑡𝑐ℎ + 𝜋∙ ( 2 ) (19)
250 Convective thermal resistance 𝑅𝑠𝑝 ‒ 𝑓 between circulating fluid and pipe surface can be expressed as:
ot
251 𝑅𝑠𝑝 ‒ 𝑓 = 2𝜋 ∙ 𝑟
1
𝑝,𝑖 ∙ 𝑙𝑠𝑝 ∙ 𝛼𝑓
1
+ 2𝜋𝜆
𝑝𝑙𝑠𝑝
∙ 𝑙𝑛( )𝑟𝑠𝑝,𝑜
𝑟𝑠𝑝,𝑖 (20)
tn
252 2.2.5. Heat balance equations for each part
253 The volume of each part can be expressed by following equations:
 Circulating fluid in double spiral pipe:

rin
254
2
255 𝑉𝑓 = 𝜋 ∙ 𝑟𝑝,𝑖 ∙ 𝑙𝑠𝑝 (21)
256  “core” part:

ep
2
257 𝑉𝑐𝑜𝑟𝑒 = 𝜋 ∙ 𝑟𝑠𝑝,𝑖 ∙ 𝑙𝑏 (22)
258  “shell” part
259 𝑉𝑠ℎ𝑒𝑙𝑙 = 𝜋 ∙ (𝑟2𝑏 ‒ 𝑟𝑠𝑝,𝑜

2
) ∙ 𝑙𝑏 (23)
Pr
260 Using above parameters, the heat balance of circulating fluid in double spiral pipe, “core”, and “shell” parts
261 in every time step 𝑑𝑡 = 60𝑠 can be expressed using Equations (24), (25), and (26), respectively.
11
262  Circulating fluid in double spiral pipe:
𝑑𝑇𝑓 𝑇𝑓,𝑠𝑝 ‒ 𝑇𝑝,𝑠𝑝
ed
263 𝑐𝑓 ∙ 𝜌𝑓 ∙ 𝑉𝑓 ∙ 𝑑𝑡
= 𝑐𝑓 ∙ 𝜌𝑓 ∙ 𝑣𝑓 ∙ (𝑇𝑓,𝑖𝑛 ‒ 𝑇𝑓,𝑜𝑢𝑡) ‒ 𝑅𝑠𝑝 ‒ 𝑓 (24)
264  The “core” part:

𝑑𝑇𝑐𝑜𝑟𝑒 𝑇𝑝𝑛𝑙 ‒ 𝑇𝑐𝑜𝑟𝑒
265 𝑐𝑐𝑜𝑟𝑒 ∙ 𝜌𝑐𝑜𝑟𝑒 ∙ 𝑉𝑐𝑜𝑟𝑒 ∙ = (25)
iew
𝑑𝑡 𝑅𝑠𝑝 ‒ 𝑐
266  The “shell” part:

𝑑𝑇𝑠ℎ𝑒𝑙𝑙 𝑇𝑝𝑛𝑙 ‒ 𝑇𝑠ℎ𝑒𝑙𝑙 𝑇𝑠ℎ𝑒𝑙𝑙 ‒ 𝑇𝑠
267 𝑐𝑠ℎ𝑒𝑙𝑙 ∙ 𝜌𝑠ℎ𝑒𝑙𝑙 ∙ 𝑉𝑠ℎ𝑒𝑙𝑙 ∙ 𝑑𝑡
= 𝑅𝑠𝑝 ‒ 𝑠
‒ 𝑅𝑠 ‒ 𝑏 (26)
268 where 𝑐𝑓, 𝑐𝑐𝑜𝑟𝑒, 𝑐𝑠ℎ𝑒𝑙𝑙, 𝜌𝑓, 𝜌𝑐𝑜𝑟𝑒, 𝜌𝑠ℎ𝑒𝑙𝑙 are heat capacities and densities of circulating fluid, grouting
v
269 material and pile material, 𝑇𝑐𝑜𝑟𝑒, 𝑇𝑠ℎ𝑒𝑙𝑙, 𝑇𝑏 are temperature of temperature nodes in “core” part, “shell” part
re
270 and boundary of the thermal pile.
271 Here, inlet temperature of the spiral pipe equals to the inflow temperature of the circulating fluid 𝑇𝑓,𝑖𝑛 =
272 𝑇𝑝,𝑖𝑛, outlet temperature of the spiral pipe equals to the outflow temperature of the circulating fluid 𝑇𝑓,𝑜𝑢𝑡 =
er
273 𝑇𝑝,𝑜𝑢𝑡.
pe
274 2.3. Combination of new calculation method and simulation tool for GSHP systems
275 GroundClub, a GSHP system simulation program developed by Environmental System Research Laboratory
276 of Hokkaido University [32], is based on the analytical model of Infinite Line Source (ILS) and Infinite
277 Cylindrical Source (ICS) models., A new simulation tool was developed by integrating the above innovative
ot
278 calculation method into GroundClub in MATLAB R2020b (released in September 17, 2020, developed by
279 MathWorks, Portola Valley, California, U.S.), the flow calculations for which have been shown in Figure 11.
tn
rin
ep
Pr
280
12
281 Fig. 11. Calculation flowchart for the GSHP system with double spiral pipe GHEs
282 For one time step 𝑑𝑡 = 60𝑠, the calculation procedures are under:
ed
283  Electricity consumption 𝐸ℎ𝑝 of the heat pump unit and fluid temperature 𝑇1,𝑜𝑢𝑡 flowing out of the
284 heat pump unit are calculated using the building load 𝑄2 (secondary side), heat extraction/injection
iew
285 from/into the ground 𝑄1 (primary side). 𝑇1,𝑜𝑢𝑡 is equal to the inflow temperature of the circulating
286 fluid 𝑇𝑝,𝑖𝑛.
287  The heat transfer processes inside the thermal pile and inflow temperature of the circulating fluid
288 𝑇𝑝,𝑜𝑢𝑡 have been calculated using the new calculation method proposed. 𝑇𝑝,𝑜𝑢𝑡 is equal to the fluid
v
289 temperature 𝑇1,𝑖𝑛 flowing into the heat pump unit, which will be used for the next time step.
re
290  The surrounding soil temperature 𝑇𝑠 (outside the thermal pile) is calculated in GroundClub, 𝑇𝑠 is
291 equal to the thermal pile boundary temperature 𝑇𝑏, which will also be used for the next time step.
292 The simulation results of the GHSP system for long-term operations can be obtained by performing
er
293 continuous step-by-step simulations of temperature changes in pipe inlet 𝑇𝑝,𝑖𝑛, outlet 𝑇𝑝,𝑜𝑢𝑡, and ground 𝑇𝑠.
294 The calculation speed of this simulation tool has been noted to be much faster than that of the numerical tool;
pe
295 for verifications performed in ensuing section 3.1.1, 3019 h of simulations for space heating took
296 approximately 7 s (specifications of the workstation used: OS: Windows 10 Pro; processor: 1.80 𝐺𝐻𝑧 × 2,
297 RAM: 16 GB).

ot
298 3. Results and analyses
299 The verification work was carried out in a 3-story house in Sapporo, Japan [Fig. 12]. Table 1 lists the basic
tn
300 information about the house. Energy efficient technologies used in this house include a photovoltaic panel
301 system for electricity supply, heat recovery ventilation system, and GHSP system, FCU (fan coil unit), and
302 radiant air conditioning system. The house has been in operation since 2021/7 and was verified as a nearly-
rin
303 ZEB (zero energy building).
304
ep
Pr
305
13
306 Fig. 12. The house in Sapporo, Japan, used for verification work
307
308 Table 1. Basic information of the house
ed
Building Energy saving technologies
Location: Sapporo, Japan Heat recovery ventilation system
Floor area: 650.85 m2 High thermal insulation
Number of floors: 3 PV system
Operation time: 2021/7 ~ now Radiant air conditioning system
Structure: Wooden GSHP system
iew
309
310 There are 3 GSHP units (rated power 10 kW) and 24 thermal piles installed on each floor of the house to
311 provide space heating and cooling and snowmelt in winter. Figure 14 illustrates the system diagram of each
312 floor. The thermal pile configuration for each floor is as follows: 4 (2 series × 2 parallel) for 1st floor (stores),
v
313 6 (2 series × 3 parallel) for 2nd floor (office rooms), 4 (1 + 1 + 2) for 3rd floor (office rooms and meeting
re
314 rooms) and 10 for snowmelt (Fig. 15).
er
pe
315
316 Fig. 13. Heat pump units on the first floor
ot
317
tn
rin
ep
318
Pr
319 Fig. 14. Schematic diagram of the 3-story house

320
14
ed
v iew
re
321
322 Fig. 15. Layout diagram of thermal piles installed in the house
323 er
324 Each GSHP unit has temperature sensors and electromagnetic flow meters that collect real time
325 measurements of pipe inlet temperature 𝑇𝑝,𝑖𝑛, outlet temperature 𝑇𝑝,𝑜𝑢𝑡, circulating fluid flow 𝑣𝑓, and power
326 consumption of heat pump unit 𝐸ℎ𝑝. Data was collected over a period of one year and was used for the purpose
pe
327 of verification. A schematic diagram of one GHSP system and its attached measurement points has been
328 illustrated in Figure 16.

ot
tn
rin
ep
329
330 Fig. 16. Schematic diagram of GSHP system and measuring points
331
Pr
332 All the thermal piles use double spiral pipe GHEs and are installed according to the “Geothermal Tornado
333 Method” developed by Japan Pile Corporation [10]. The thermal properties (thermal conductivity, specific
15
334 heat capacity, and density) of grout and pile material were measured by the “HC-10 Quick λ Thermal
335 Conductivity Tester” by EKO Instruments [33], the parameters for which have been listed in Table 2.
ed
336 Table 2. GSHP system specifications
Description Unit Value
Thermal pile
Depth 𝑙𝑏 m 20
Outer radius 𝑟𝑏
iew
m 0.3
Inner radius 𝑟𝑠𝑝,𝑜 m 0.2
“Core” part (cement & soil)
Thermal conductivity 𝜆𝑐𝑜𝑟𝑒 W/m∙K 0.6
Specific heat capacity 𝑐𝑐𝑜𝑟𝑒 kJ/kg∙K 0.9
Density 𝜌𝑐𝑜𝑟𝑒 kg/m3 2100
“Shell” part (concrete)
Thermal conductivity 𝜆𝑠ℎ𝑒𝑙𝑙 W/m∙K 2.0
Specific heat capacity 𝑐𝑠ℎ𝑒𝑙𝑙 kJ/kg∙K 0.95
v
Density 𝜌𝑠ℎ𝑒𝑙𝑙 kg/m3 2500
Double spiral pipe GHE
Pipe outer diameter 𝑑𝑝,𝑜 m 0.032
re
Pipe inner diameter 𝑑𝑝,i m 0.026
Spiral distance 𝑙𝑝𝑖𝑡𝑐ℎ m 0.25
Thermal conductivity of pipe material 𝜆𝑝 W/m∙K 0.38
Length of spiral pipe 𝑙𝑠𝑝 m 94.63
Soil
Specific heat capacity 𝜆𝑠 kJ/kg∙K 2
Density 𝜌𝑠 kg/m3 1500
Initial temperature 𝑇𝑠
er
Circulating fluid: 40% ethylene glycol solution
℃ 12
337
pe
338 3.1. Verification
339 To determine the effective thermal conductivity of the ground, an in-situ Thermal Response Test (TRT) was
340 carried out on a thermal pile from the 3rd floor system. The schematic diagram of the TRT test is shown in
341 Figure 17 (a).

ot
tn
rin
ep
342
343 Fig. 17. a) Schematic diagram of TRT test and b) 338-h measurement
344 Figure 17 (b) shows the pipe inlet/outlet temperature and flowrate of the circulating fluid recorded for 338
Pr
345 h (2021/10/06 13:00–2021/10/20 15:00). The data were analyzed using the Kelvin Line Theorem [34]. The
346 effective thermal conductivity of the ground where the house is located was calculated according to the
347 following procedure:

16
348 A transient 1D heat conduction analysis [35] gives the fluid temperature as a function of time:
( 4𝛼𝑠
)
ed
𝑞 𝑞
349 𝑇𝑓(𝑡) = 4𝜋 ∙ 𝜆 ∙ 𝑙𝑛 (𝑡) + 4𝜋 ∙ 𝜆 ∙ 𝑙𝑛
𝑠 𝑠 𝑟2𝑏
‒ 𝛾 + 𝑞 ∙ 𝑅𝑏 + 𝑇𝑠 (27)
350 where 𝑇𝑓 is fluid temperature, 𝑇𝑠 is the initial ground temperature, 𝑞 is the heat load per unit length, 𝜆𝑠
351 is the effective thermal conductivity of the ground, 𝑟𝑏 is radius borehole, 𝛼𝑠 is thermal diffusivity of the
iew
352 ground, 𝛾 is Euler’s constant, 𝑅𝑏 is borehole thermal resistance.
353 Substituting the 𝑚' and 𝑏 for the constants in Equation (27), the equation becomes:
354
𝑞
𝑇𝑓(𝑡) = 𝑚' ∙ 𝑙𝑛 (𝑡) + 𝑏; 𝑚' = 4𝜋 ∙ 𝜆 ; 𝑏 = 4𝜋𝜆 𝑙𝑛
𝑠
𝑞
(
𝑠
4𝛼𝑠
𝑟2𝑏 )
‒ 𝛾 + 𝑞𝑅𝑏 + 𝑇𝑠 (28)
v
355 The equation is only valid if the time is not too short, which means that the time criterion 𝑡 > 5 ∙ 𝑟2𝑏/𝛼𝑠
re
356 must be satisfied. Assuming the thermal diffusivity of the ground 𝛼𝑠 = 0.002 𝑚2/𝑠, and the borehole radius
357 𝑟𝑏 = 0.3 𝑚, the effective start time of the TRT analysis becomes:
𝛼𝑠 ∙ 𝑡 er0.002𝑡
358 𝑟2𝑏
>5→
0.32
> 20 →𝑡 > 225 ℎ (29)
pe
ot
359
360 Fig. 18. a) Mean fluid temperature and b) temperature in a logarithmic scale with the fitting linear function
tn
361 The mean fluid temperature over time during the TRT test and the measured temperature with linear and
362 logarithmic time scales have been plotted in Figures 18 a and b, respectively. Thermal conductivity is
rin
363 determined from the slope 𝑘 of the linear curve from the effective start time of 225 h to the end time of 338
364 h.
𝑞
365 𝜆𝑠 =
4𝜋 ∙ 𝑚'
(30)
ep
366 After calculation, the effective thermal conductivity of the ground is 𝜆𝑠 = 1.846(𝑊/𝑚 ∙ 𝐾). By applying
367 this value to the simulation, the verification work was performed by comparing the measured and simulated
368 data and simulation under two cases discussed below.

Pr
17
369 3.1.1. Space cooling in summer
370 During the space cooling period, the indoor room temperatures were set to 26 °C. As majority of the rooms
ed
371 on the first floor are occupied, GSHP unit 1 ran the longest. As such we chose this unit for data analyses. The
372 heat injection from the house into ground can be calculated using the hourly recorded pipe inlet/outlet
iew
373 temperature and the flowrate data of the circulating fluid during the space cooling period [Fig. 19].
374 𝑄𝑖𝑛𝑗𝑒𝑐𝑡𝑖𝑜𝑛 = 𝜌𝑓 ∙ 𝑐𝑓 ∙ v𝑓 ∙ (𝑇𝑝,𝑖𝑛 ‒ 𝑇𝑝,𝑜𝑢𝑡) (31)
375 Then heat injection during this period was used as the primary side load 𝑄1 for subsequent simulations.
376 Except for the preparation time, the verification period was set as 2021/8/1–2021/8/31. Figure 20 shows the
v
377 comparative hourly changes of the pipe inlet temperature 𝑇𝑝,𝑖𝑛 for both, measured and simulated data.
re
378 The RMSE (Root Mean Squared Error) value between measurement and simulation data from 8/1 to 8/31
379 was reported to be 0.73, indicating the accuracy of the simulation tool. The simulated and measured average
380 temperatures were reported to be 19.71 °C and 19.77 °C, respectively. Changes in both temperatures display
381
er
similar trends, with minor differences when the temperatures change rapidly.
pe
ot
tn
382
383 Fig. 19. Heat injection into the ground during space cooling period
384
rin
ep
385
386 Fig. 20. Pipe inlet temperature during space cooling period
Pr
18
387 3.1.2. Space heating in summer
388 All GSHP units were set to provide 45 °C hot water during the space heating period, while the indoor room
ed
389 temperatures were set to 22 °C. The recorded data was used to calculate the heat extracted from the ground,
390 the results for which are shown in Figure 21:
iew
391 𝑄𝑒𝑥𝑡𝑟𝑎𝑐𝑡𝑖𝑜𝑛 = 𝜌𝑓 ∙ 𝑐𝑓 ∙ v𝑓 ∙ (𝑇𝑝,𝑜𝑢𝑡 ‒ 𝑇𝑝,𝑖𝑛) (32)
392 Figure 22 plots the comparison between measured and simulated data for pipe inlet temperature. Excluding
393 maintenance and holiday time, the verification period was selected from 1/7 to 2/7, which is the coldest period
394 in Sapporo, when space heating is most needed. The results show that the temperatures between measurement
v
395 and simulation are well matched with an RMSE value of 1.06. This value is acceptable and demonstrates the
re
396 effectiveness of this simulation tool.
er
pe
397
398 Fig. 21. Heat extraction from the ground during space heating period
399
ot
tn
rin
400
401 Fig. 22. Pipe inlet temperature during space heating period
ep
402 This simulation tool has been designed to generate high accuracy performance assessment results to make
403 sound estimates supporting the early-stage decision-making. The summer and winter verifications show that
404 this simulation tool for performance prediction can be applied in the preliminary design phase of future GSHP
Pr
405 systems using thermal piles and double spiral pipe GHEs.
19
406 4. Discussion
ed
407 The comparison between simulated and measured data validates the functionality of this simulation tool.
408 The tool, however, has a few shortcomings that need to be discussed further.
409 4.1. Superiority to another CaRM model
iew
410 Our analyses shows that this new simulation tool can yield better performance predictions than the well-
411 accepted CaRM-He model [24], which also analyzed the thermal behavior of helical GHEs. The new
412 calculation method using fin efficiency assumes the spiral pipe to be a vertical heat release/absorption panel,
v
413 where the temperature nodes can be set at any point in the grout/pile, irrespective of their position with respect
re
414 to the pipe [Fig. 23].
415 As the CaRM-He model considers the heat transfer along the depth direction, it is assumed that the heat
416 flow in/out of the temperature points closest to the spiral pipe are along the four directions: the pipe surface,
er
417 the point above, the point below, and the point next on the parallel line [24]. However, if the pitch (spiral
418 distance) is larger than 100 mm, at temperature points other than those parallel to the pipe, the heat flow from
pe
419 pipe surface will be reduced, thus making the heat balance irrelevant. This makes the CaRM-He model suitable
420 only for small-pitch spiral pipes.

ot
tn
421
422 Fig. 23. Comparison of the new calculation method (a) and model CaRM-He (b)
rin
423 4.2. Applicability to short-term operations
424 The measurement results of the TRT test were compared with the simulation results during the verification
425 works. Although the two sets of data agree well after a period of 100 h, there are significant errors in both pipe
ep
426 inlet and outlet temperatures during initial 50 h [Fig. 24]. This means that, when applying the CaRM model,
427 there will be a significant difference in the simulated and calculated temperature changes over a short period
428 of time in case of large instantaneous heat injections. This is mainly because the CaRM or lumped thermal
Pr
429 capacity model assumes that the temperature of each solid is spatially independent (uniform) and can be
430 represented by the same temperature. If the volume of the solids corresponding to the temperature nodes is too
20
431 large, the represented temperatures may deviate from the actual temperatures.
ed
v iew
432
re
433 Fig. 24. Inlet and outlet temperature of double spiral pipe GHE during TRT
434
435 To solve this problem, the size of solid can be reduced so that the network of interconnected thermal
436 resistances and capacities has a finer mesh. The “core” and “shell” parts are divided into 𝑛 regions [Fig. 25],
437
er
corresponding to 𝑛 temperature nodes. With this, the heat load and circulating fluid flow in the initial 50 h of
438 TRT [Section 3.1] can be used as an input for the simulation. Figure 26 shows that the duration between the
pe
439 simulation results reduces with increase in temperature nodes during the initial 30 h. This thus proves that the
440 simulation tool can also be used for short-term operation predictions.
441
ot
tn
442
443
rin
Fig. 25. Concept of dividing “core” and “shell” part into 𝑛 annual regions
444
ep
Pr
21
ed
iew
445
446
v
Fig. 26. Comparison of pipe inlet temperature in initial 30 h
447 4.3. Coefficient of heat extraction/injection
re
448 The effect of pitch on the efficiency of the double spiral pipe GHE was investigated, given that this
449 simulation tool can be used for double spiral pipe GHE with any pitch size. A single U-shaped pipe GHE with
450 the same underground depth, pipe material, and smaller borehole diameter has also been included to compare
er
451 efficiency of GHEs with different configurations. To evaluate the performance of GHE, this study introduces
452 the heat extraction/injection rate per unit length of GHE, which is defined as the coefficient of heat
pe
453 extraction/injection [36]. The associated equation can be expressed as follows:
𝑇𝑝,𝑖𝑛 + 𝑇𝑝,𝑜𝑢𝑡 𝑄
454 ∆𝑇 = 2
‒ 𝑇𝑠; 𝑞' = 𝑙 (33)
𝑏 ∙ ∆𝑇
455 where 𝑇𝑝,𝑖𝑛 and 𝑇𝑝,𝑜𝑢𝑡 are the pipe inlet and outlet temperature, 𝑇𝑠 is the initial ground temperature, 𝑄
ot
456 is heat load, 𝑙𝑏 is depth of borehole, 𝑞' is the coefficient of heat extraction/injection.
457 The parameters of the thermal pile, spiral pipe, soil, and circulating fluid were kept constant [Tables 2 and
tn
458 3]. The simulation results depict the change in the coefficient of heat extraction or injection per 100 h [Fig.
459 28]. At 100 h, the efficiency of spiral GHE increased by 6.7%, 14.3%, 22.9%, 32.7%, and 39.6% for
460 corresponding changes in pitch from 0.07 m to 0.1, 0.15, 0.2, 0.25, and 0.3 m. Moreover, at 800 h, the efficiency
rin
461 of double spiral pipe GHE was seen to be increased by 4.6%, 9.5%, 14.8%, 20.7%, and 24.7% for
462 corresponding pitch changes from 0.07 m to 0.1, 0.15, 0.2, 0.25, and 0.3 m.
463 Table 3. Simulation condition settings

ep
Description Unit Value

Heat load 𝑄 W 2000
Flow rate of circulating fluid v𝑓 L/min 6.5
Length of borehole for U-shaped pipe GHE and m 20
double spiral pipe 𝑙𝑏
Pr
Radius of borehole for U-shaped pipe GHE m 0.06

Radius of thermal pile m 0.3
Duration time 𝑡 h 800
Spiral distance of spiral pipe 𝑙𝑝𝑖𝑡𝑐ℎ m 0.07, 0.1, 0.15, 0.2,
0.25, 0.3
22
464
465 The lowest efficiency of the double spiral pipe GHE (𝑝𝑖𝑡𝑐ℎ = 0.3𝑚) is still far superior to the U-shaped
ed
466 pipe GHE. At 800 h, the efficiency of double spiral pipe GHE (𝑝𝑖𝑡𝑐ℎ = 0.3𝑚) was noted to be 67.5%, more
467 than of the U-shaped pipe GHE, further proving the added advantage of double spiral pipe GHE.
468 To sum up, for constant underground depths, the efficiency of double spiral pipe GHE can be greatly
iew
469 improved by shortening the pitch size, irrespective of the duration of operations.
470 However, there are some limitations in this study: first, the GSHP units were not operating continuously
471 during the measurement process, but were turned on when in use and turned off when not in use, so some of
v
472 the measured data may be affected by frequent starting and stopping of the system; second, although the new
473 method simplified the heat transfer calculations between the ground and double spiral pipe and greatly reduced
re
474 the required calculation time, the results of this study are not as accurate but acceptable compared to some
475 numerical model that may require several weeks of calculation time. More verification work will be performed
476 in the future to seek modifications of this method that can yield better results.
er
pe
ot
tn
477
478 Fig. 27. The average temperature of inlet and outlet of spiral pipe with different pitches and U-shaped pipe for 800 h
479
rin
ep
480
Pr
481 Fig. 28. Coefficient of heat extraction/injection for spiral pipe for varied pitches every 100 h
23
482 5. Conclusion
483 The proposed new calculation method based on the CaRM model is an improvement over the previous ones
ed
484 as it adopts the concept of fin efficiency, and integrates a simulation program for GSHP system in MATLAB:
485  Different from the traditional U-shaped pipe GHE, the double spiral pipe GHE adopts shorter
iew
486 underground borehole, but has a larger area for heat exchange between the ground and circulating
487 fluid without additional drilling work, which saves the initial costs of the GSHP system and promotes
488 its use in the future.
489  By replacing the spiral pipe with a “vertical” heat generating/absorbing panel, the temperature
v
490 changes in each part of the thermal pile can be easily obtained without a huge amount of calculation.
re
491  Verification work carried out in an energy-efficient house proved that the simulation tool produces
492 reliable and high-precision simulation results for the long-term operation of GSHP systems, both in
493 summers and winters. This would assist in the preliminary designing phase of future GSHP systems.
494
er
Compared to the widely accepted CaRM-He model [24], this simulation tool can do better in the following
495 two points:
 As it is derived from a calculation method for floor heating systems, its applicability is not limited by
pe
496
497 the pitch size of spiral pipes.
498  For short-term simulations of less than 50 h, the simulated results can still agree well with the
499 measured results by adding more temperature nodes to the thermal network.
ot
500 Finally, the simulation results show that the thermal efficiency can be greatly improved by using the double
501 spiral pipe GHE with smaller pitch.

tn
502 It is thus believed that this simulation tool will help in designing of other GSHP systems using double spiral
503 pipe GHEs, thereby facilitating the transition to energy-efficient buildings in the future.
rin
504 Nomenclature
Symbols Subscripts
l length/depth, m b borehole of thermal pile
η fin efficiency sp spiral pipe
T temperature, ℃ s ground soil
ep
R heat resistance, 𝐾/𝑊 f circulating fluid

r radius, m in pipe inlet
d diameter, m out pipe outlet
α heat transfer coefficient, 𝑊/𝑚2∙𝐾 o outer
λ thermal conductivity, 𝑊/𝑚∙𝐾 i inner
v flow rate, m3/s core “core” part
Q thermal load, k𝑊 shell “shell” part
Pr
ρ density, kg/m3 pnl “vertical” panel

c specific heat capacity, J/kg∙𝐾 p pipe
E electricity consumption, kW hp heat pump
V volume, m3
q thermal load per unit length, 𝑊/m
24
q’ coefficient of heat extraction/injection
505
ed
506 CRediT authorship contribution statement
507 Kunning Yang: Methodology, Conceptualization, Writing - original draft, Validation, Software. Takao
508 Katsura: Conceptualization, Funding acquisition, Writing - review & editing, Supervision. Shigeyuki
iew
509 Nagasaka: Supervision. Katsunori Nagano: Supervision.
510
511 Acknowledgements
512 This work was supported by JST SPRING, Grant Number JPMJSP2119. And this study is based on results
v
513 obtained from the project “Renewable energy heat utilization technology development for cost reduction”,
re
514 commissioned by the Japan national agency NEDO. The authors also thank M's Industry Co., Ltd and Japan
515 Pile Corporation for their assistance in this research project.
516 References
er
517 [1] P. Nejat, F. Jomehzadeh, M.M. Taheri, M. Gohari, M.Z. Abd. Majid, A global review of energy
518 consumption, CO2 emissions and policy in the residential sector (with an overview of the top ten CO2
pe
519 emitting countries), Renew. Sustain. Energy Rev. 43 (2015) 843–862.
520 https://doi.org/10.1016/j.rser.2014.11.066.
521 [2] D. Ürge-Vorsatz, L.F. Cabeza, S. Serrano, C. Barreneche, K. Petrichenko, Heating and cooling energy
ot
522 trends and drivers in buildings, Renew. Sustain. Energy Rev. 41 (2015) 85–98.
523 https://doi.org/10.1016/j.rser.2014.08.039.
tn
524 [3] J.W. Lund, A.N. Toth, Direct utilization of geothermal energy 2020 worldwide review, Geothermics. 90
525 (2021). https://doi.org/10.1016/j.geothermics.2020.101915.
526 [4] S. Naicker, Performance Analysis of a Large-Scale Ground Source Heat Pump System (Ph.D), De
rin
527 Montfort University, Leicester, UK, 2016.
528 [5] J. Luo, J. Rohn, W. Xiang, D. Bertermann, P. Blum, A review of ground investigations for ground source
529 heat pump (GSHP) systems, Energy Build. 117 (2016) 160–175.
ep
530 https://doi.org/10.1016/j.enbuild.2016.02.038.
531 [6] D. Wang, L. Lu, W. Zhang, P. Cui, Numerical and analytical analysis of groundwater influence on the
Pr
532 pile geothermal heat exchanger with cast-in spiral coils, Appl. Energy. 160 (2015) 705–714.
533 https://doi.org/10.1016/j.apenergy.2015.04.037.
25
534 [7] T. Amis, F.A. Loveridge, Energy piles and other thermal foundations, developments in UK practice and
535 research, REHVA J. (2014) 32–35.
ed
536 [8] http://www.geohpaj.org/introduction/qa/1-10. (Accessed December 16, 2022).
537 [9] J. Fadejev, R. Simson, J. Kurnitski, F. Haghighat, A review on energy piles design, sizing and modelling,
iew
538 Energy. 122 (2017) 390–407. https://doi.org/10.1016/j.energy.2017.01.097.
539 [10] Zero-emission by low cost construction method of geothermal heat collection | Challenge zero.
540 https://www.challenge-zero.jp/en/casestudy/673. (Accessed November 22, 2022).
541 [11] A. Carotenuto, P. Marotta, N. Massarotti, A. Mauro, G. Normino, Energy piles for ground source heat
v
542 pump applications: Comparison of heat transfer performance for different design and operating
re
543 parameters, Appl. Therm. Eng. 124 (2017) 1492–1504.
544 https://doi.org/10.1016/j.applthermaleng.2017.06.038.
545 [12] Q. Zhao, B. Chen, F. Liu, Study on the thermal performance of several types of energy pile ground heat
546
er
exchangers: U-shaped, W-shaped and spiral-shaped, Energy Build. 133 (2016) 335–344.
547 https://doi.org/10.1016/j.enbuild.2016.09.055.
pe
548 [13] A.A. Serageldin, A. Radwan, T. Katsura, Y. Sakata, S. Nagasaka, K. Nagano, Parametric analysis,
549 response surface, sensitivity analysis, and optimization of a novel spiral-double ground heat exchanger,
550 Energy Convers. Manag. 240 (2021). https://doi.org/10.1016/j.enconman.2021.114251.
551 [14] P. Cui, X. Li, Y. Man, Z. Fang, Heat transfer analysis of pile geothermal heat exchangers with spiral
ot
552 coils, Appl. Energy. 88 (2011) 4113–4119. https://doi.org/10.1016/j.apenergy.2011.03.045.
553 [15] W. Zhang, H. Yang, L. Lu, P. Cui, Z. Fang, The research on ring-coil heat transfer models of pile
tn
554 foundation ground heat exchangers in the case of groundwater seepage, Energy Build. 71 (2014) 115–
555 128. https://doi.org/10.1016/j.enbuild.2013.12.016.
556 [16] Sk. Park, S.R. Lee, H. Park, S. Yoon, J. Chung, Characteristics of an analytical solution for a spiral coil
rin
557 type ground heat exchanger, Comput. Geotech. 49 (2013) 18–24.
558 https://doi.org/10.1016/j.compgeo.2012.11.006.
ep
559 [17] Y. Man, H. Yang, N. Diao, P. Cui, L. Lu, Z. Fang, Development of spiral heat source model for novel
560 pile ground heat exchangers, HVAC R Res. 17 (2011).
561 [18] Y. Man, H. Yang, N. Diao, J. Liu, Z. Fang, A new model and analytical solutions for borehole and pile
Pr
562 ground heat exchangers, Int. J. Heat Mass Transf. 53 (2010) (2593–2601).
563 https://doi.org/10.1016/j.ijheatmasstransfer.2010.03.001.
26
564 [19] M. Li, A.C.K. Lai, Heat-source solutions to heat conduction in anisotropic media with application to
565 pile and borehole ground heat exchangers, Appl. Energy. 96 (2012) 451–458.
ed
567 [20] M. Li, A.C.K. Lai, New temperature response functions (G functions) for pile and borehole ground heat
iew
568 exchangers based on composite-medium line-source theory, Energy. 38 (2012) 255–263.
569 https://doi.org/10.1016/j.energy.2011.12.004.
570 [21] A. Zarrella, A. Capozza, M. de Carli, Analysis of short helical and double U-tube borehole heat
571 exchangers: A simulation-based comparison, Appl. Energy. 112 (2013) 358–370.
v
re
573 [22] A. Zarrella, M. de Carli, A. Galgaro, Thermal performance of two types of energy foundation pile:
574 Helical pipe and triple U-tube, Appl. Therm. Eng. 61 (2013) 301–310.
575 https://doi.org/10.1016/j.applthermaleng.2013.08.011.
576
er
[23] T. Katsura, T. Higashitani, Y. Fang, et al, A new simulation model for vertical spiral ground heat
577 exchangers combining cylindrical source model and capacity resistance model, Energies. 13 (2020).
pe
578 https://doi.org/10.3390/en13061339.
579 [24] A. Zarrella, M. de Carli, Heat transfer analysis of short helical borehole heat exchangers, Appl. Energy.
580 102 (2013) 1477–1491. https://doi.org/10.1016/j.apenergy.2012.09.012.
581 [25] J. Wojtkowiak, Lumped thermal capacity model, in: Encyclopedia of Thermal Stresses, 2014.
ot
582 [26] M. de Carli, M. Tonon, A. Zarrella, R. Zecchin, A computational capacity resistance model (CaRM) for
583 vertical ground-coupled heat exchangers, Renew. Energy. 35 (2010) 1537–1550.

tn
584 https://doi.org/10.1016/j.renene.2009.11.034.
585 [27] A. Kollmar, Thermal Emission of Flat Heating Surfaces and Ceiling Heating Panels with Cylinderical
586 Heat Sources (Wärmeabgabe von Plattenheizflächen und Heizdecken mit Zylindrischen Wärmequellen)
rin
587 D.A. Sinclair (Tr. by), 37p, 1961.
588 [28] H. Ishino, Research on calculation method of thermal design load in radiant heating and cooling
ep
589 systems. Sixth International IBPSA Conference (BS′ 99), 1999, pp. 885–892.
590 [29] A.N. Dravid, K.A. Smith, E.W. Merrill, P.L.T. Brian, Effect of secondary fluid motion on laminar flow
591 heat transfer in helically coiled tubes, AIChE J. 17 (1971) 1114–1122.

Pr
592 https://doi.org/10.1002/aic.690170517.
27
593 [30] V. Gnielinski, New equations for heat and mass transfer in turbulent pipe and channel flow, Int. Chem.
594 Eng. 16 (1976).
ed
595 [31] P.K. Namburu, D.K. Das, K.M. Tanguturi, R.S. Vajjha, Numerical study of turbulent flow and heat
596 transfer characteristics of nanofluids considering variable properties, Int. J. Therm. Sci. 48 (2009) 290–
iew
597 302. https://doi.org/10.1016/j.ijthermalsci.2008.01.001.
598 [32] T. Katsura, K. Nagano, Y. Sakata, H. Wakayama, A design and simulation tool for ground source heat
599 pump system using energy piles with large diameter, Int. J. Energy Res. Published online. 43 (2019)
600 1505–1520. https://doi.org/10.1002/er.4372.
v
601 [33] HC-10. https://www.eko-instruments.com/us/categories/products/thermal-analysis-instruments/hc-10.
re
602 (Accessed November 17, 2022).
603 [34] B. Nordell, Thermal Response Test (TRT) State-of-the Art 2011: IEA ECES, 2011 ANNEX 21.
604 [35] L.R. Ingersoll, H.J. Plass, Theory of the ground pipe heat source for heat pump, ASHVE journal section,
605 Heat. Pip. Air Cond. 119–122 (1948).

er
606 [36] T. Katsura, K. Nagano, Y. Nakamura, A study on design method for the ground heat exchanger’s
pe
607 specification of ground source heat pump system, Nihon Kenchiku Gakkai Kankyokei Ronbunshu. 76
608 (2011) 59–66. https://doi.org/10.3130/aije.76.59.
609
610
ot
611
tn
rin
ep
Pr
28

SSRN Id4355816

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SSRN Id4355816

Uploaded by

Copyright:

Available Formats

1 Development and application of a new calculation method for

4 aGraduate School of Engineering, Hokkaido University, Sapporo, 060-8628, Japan

5 bFaculty of Engineering, Hokkaido University, Sapporo, 060-8628, Japan

6 c Shin Nippon Air Technologies Co., Ltd., Chino, 391-0013, Japan

9 E-mail address: kunning.yang.h4@elms.hokudai.ac.jp

26 model; thermal pile; simulation

47 in Japan [Fig. 3, Fig. 4] [7] [8].

81 of the GHEs have been under discussion for a long time.

93 conditions than those with single U-shaped pipes [13].

107 double U-shaped pipes [19][20].

117 performance, with an increase in peak value of approximately 23%.

123 temperatures at multiple points are calculated.

134 double spiral pipe GHEs.

136 2.1. CaRM for double spiral pipe GHEs

151  double spiral pipe

188 double spiral pipe GHE

190 can be calculated using following equation:

194 𝑇𝑚 = 𝑇𝐿 + 𝜂 ∙ (𝑇𝐻 ‒ 𝑇𝐿) (5)

𝑄 = 𝛼 ∙ 𝜂 ∙ (𝑇𝐻 ‒ 𝑇𝐿) = 𝛼 ∙ (𝑇𝑚 ‒ 𝑇𝐿)

204 these temperatures can be expressed as:

205 𝑇𝑝𝑛𝑙 = 𝑇𝑐𝑜𝑟𝑒 + 𝜂𝑐𝑜𝑟𝑒 ∙ (𝑇𝑝,𝑠𝑝 ‒ 𝑇𝑐𝑜𝑟𝑒) (7)

206 𝑇𝑝𝑛𝑙 = 𝑇𝑠ℎ𝑒𝑙𝑙 + 𝜂𝑠ℎ𝑒𝑙𝑙 ∙ (𝑇𝑝,𝑠𝑝 ‒ 𝑇𝑠ℎ𝑒𝑙𝑙) (8)

210 2.2.3. Calculation of fin efficiency

219 𝜂𝑐𝑜𝑟𝑒 = 2𝛼𝑐𝑜𝑟𝑒 𝑙𝑝𝑖𝑡𝑐ℎ ‒ 𝑑𝑝,𝑜 (12)

220 𝜂𝑠ℎ𝑒𝑙𝑙 = 2𝛼𝑠ℎ𝑒𝑙𝑙 𝑙𝑝𝑖𝑡𝑐ℎ ‒ 𝑑𝑝,𝑜 (13)

224 𝑄𝑠ℎ𝑒𝑙𝑙 = 𝛼𝑠ℎ𝑒𝑙𝑙 ∙ 𝐴𝑠ℎ𝑒𝑙𝑙 ∙ (𝑇𝑝𝑛𝑙 ‒ 𝑇𝑠ℎ𝑒𝑙𝑙) (15)

231 thermophysical properties, density 𝜌, kinematic viscosity ν, and thermal diffusivity 𝛼.

239  Gnielinski’s formula was used for turbulent flow [30][31]:

(𝑓/8) ∙ (𝑅𝑒 ‒ 1000) ∙ 𝑃𝑟

252 2.2.5. Heat balance equations for each part

253 The volume of each part can be expressed by following equations:

 Circulating fluid in double spiral pipe:

256  “core” part:

258  “shell” part

259 𝑉𝑠ℎ𝑒𝑙𝑙 = 𝜋 ∙ (𝑟2𝑏 ‒ 𝑟𝑠𝑝,𝑜

264  The “core” part:

266  The “shell” part:

286 fluid 𝑇𝑝,𝑖𝑛.

297 RAM: 16 GB).

298 3. Results and analyses

303 ZEB (zero energy building).

319 Fig. 14. Schematic diagram of the 3-story house

328 illustrated in Figure 16.

341 Figure 17 (a).

347 following procedure:

368 data and simulation under two cases discussed below.

374 𝑄𝑖𝑛𝑗𝑒𝑐𝑡𝑖𝑜𝑛 = 𝜌𝑓 ∙ 𝑐𝑓 ∙ v𝑓 ∙ (𝑇𝑝,𝑖𝑛 ‒ 𝑇𝑝,𝑜𝑢𝑡) (31)

390 the results for which are shown in Figure 21:

409 4.1. Superiority to another CaRM model

420 only for small-pitch spiral pipes.

423 4.2. Applicability to short-term operations

447 4.3. Coefficient of heat extraction/injection

463 Table 3. Simulation condition settings

Description Unit Value

Radius of borehole for U-shaped pipe GHE m 0.06

488 its use in the future.

495 two points:

497 the pitch size of spiral pipes.

501 spiral pipe GHE with smaller pitch.

R heat resistance, 𝐾/𝑊 f circulating fluid

ρ density, kg/m3 pnl “vertical” panel

515 Pile Corporation for their assistance in this research project.

525 (2021). https://doi.org/10.1016/j.geothermics.2020.101915.