Investigation of the Optimal Operation Method of the Heat Recovery Ground Source Heat Pump System Installed in an Actual Building and Evaluation of Energy Saving Effect

Abstract

In this paper, a heat recovery ground source heat pump (HR-GSHP) system, in which the primary pipes of the GSHP for air conditioning and the GSHP for hot water are connected to ground heat exchangers (GHEs) and each GSHP is operated simultaneously or within a short period of time, was installed in a dormitory building on a trial basis. Then, the optimal operation method to minimize the energy consumption of the system was investigated. The operating period of the GSHP for HW was changed and simulations were conducted to determine the operating period with the lowest energy consumption, which was 8 months from April to November. Furthermore, the HR-GSHP system was operated for 8 years from 2012 to 2019, and actual measurements were carried out to verify the system performance and the energy saving effect in optimal operation. In actual operation, it was confirmed that the minimum temperature was about 10 °C or higher even when the GSHP for HW was operated year-round. Therefore, the GSHP for HW was operated year-round after the third year of operation. It was confirmed that the operation of the GSHP for HW in summer, especially in August and September when the cooling load is large, can improve the system’s efficiency by the effect of recovering cooling exhaust heat. In the eighth year of operation, when the GSHP for HW was operated most during the summer season, the system was able to reduce power consumption for air conditioning and hot water supply by approximately 17%.

1. Introduction

Ground source heat pump (GSHP) systems are gradually being introduced in Japan because they were defined as a renewable energy source in 2009 and have high potential for introduction as a renewable energy heat utilization technology. The introduction of GSHPs and GSHP systems is currently underway. However, due to installation cost issues and other factors, the number of installations of GSHP systems in Japan is currently lower than in China, the U.S., and Sweden, which are considered advanced countries, and, in particular, there are only a few cases of large-scale systems with a heat pump capacity of over 500 kW.

A heat recovery ground source heat pump (HR-GSHP) system, as shown in Figure 1, is a system in which the primary piping of two or more GSHPs that supply heating (or hot water) and cooling are connected to ground heat exchangers (GHEs), and each GSHP is operated simultaneously or within a short period of time. This HR-GSHP system can improve system performance and minimize the size of the GHEs for the amount of heat available. Since it is preferable to have the heating (or hot water) and cooling loads occur simultaneously, the HR-GSHP system is expected to be installed in large complexes or between multiple buildings.

A HR-GSHP system, in which the GSHP for air conditioning (AC) and the GSHP for hot water supply (HW) are connected to GHEs and can be operated simultaneously, was installed on a trial basis in an accommodation building with a large domestic hot water load. There are few research cases in which long-term measurements have been carried out and operational methods have been investigated in actual GSHP system installations, not only for HR-GSHP systems but also for normal GSHP systems. Luo et al. conducted 4 years of actual measurements from 2009 to 2012 on a GSHP system with 80 m × 18 borehole GHEs installed in a building in southern Germany and showed changes in ground temperature, COP, SEER, etc. [1]. Naicker and Rees presented the results of three years of measurements on a GSHP system with 100 m × 56 GHEs and discussed the importance of controlling the circulation pumps during part-load operation [2]. Furthermore, based on TRT results and measurements, they show that ground heat transfer is not affected by groundwater advection and that heat extraction and heat dissipation is repeated in UK school buildings, resulting in less significant temperature changes in the source water [3]. Spitler and Gehlin analyzed a GSHP system with 200 m × 20 GHEs [4]. Kindaichi and Nishina analyzed the results of a GSHP system with 100 m × 70 GHEs over three years of operation and found that the amount of heat injection was larger than the amount of heat extraction, resulting in a long-term increase in ground temperature [5]. It was found that the ground temperature could be controlled by reducing the heat injection [5]. Todorov et al. analyzed the results of 18 months of operation of a GSHP system with 74 GHEs with an average effective length of 302.5 m × 74 GHEs, installed as heating and cooling in a shopping center near a university and in a metro station; values of 3.7 for HPCOP and 3.5 for heating SPF were obtained [6]. They further stated that the heat extraction was about seven times larger than the heat injection, which may affect the operation of the GSHP system in the long term [6]. Bockelmann and Norbert Fisch also analyzed the results of long-term operation of six GSHP systems in Germany and stated that the balancing control between the heat injection and the heat extraction and continuous monitoring should not be ignored and that continuous monitoring, control, and maintenance are necessary [7]. Rong et al. conducted actual measurements on a hybrid geothermal heat pump system with 70 m × 544 geothermal heat exchangers and cooling towers, employing radiant air conditioning utilizing TABSs, and showed that the COP for the entire system was 5.49 [8]. Kaneko et al. measured a GSHP system using 100 m × 18 double U-tube GHEs and groundwater wells as hybrid heat sources for four years and showed that the increase in ground temperature could be reduced by reducing the heat injection of the GHEs in summer [9]. There are also several studies that have used simulations to predict the long-term performance or optimize the operation of GSHP systems. The parameter analysis and operational optimization using simulation can be categorized into those that conducted long-term prediction and optimization and those that conducted relatively short-term prediction and optimization. Fan et al. measured the thermal loads and the temperatures of heat carrier fluid in a GSHP system installed in a building in Beijing and predicted the long-term performance of the GSHP system by using the measured data and TRNSYS [10]. Luo et al. predicted a 25-year ground temperature change by using numerical analysis for three different operating conditions for a GSHP system with cooling towers in the cooling dominate area and found that the temperature change was smallest when the heat carrier fluid was circulated in all GHEs at the same time [11]. Weeratunge et al. studied the optimal control method that minimizes energy consumption or running costs for a hybrid GSHP system assisted by solar heat [12]. Yoshinaga et al. measured a hybrid system of GSHPs using 100 m × 36 borehole double U-tube GHEs and an ASHP and used TRNSYS to operate the hybrid system to maximize its efficiency based on the loads obtained from the actual measurements [13]. In recent years, fifth generation district heating and cooling systems (5GDHCSs) have attracted attention as a new way to use renewable energy heat, including the ground thermal energy [14,15]. In paper [14], Yao et al. defined the 5GDHCS concept with three key features: (a) operation at near-ambient network temperatures; (b) deployment of distributed heat pumps at the end-user side; and (c) the ability to provide both heating and cooling services. More than 100 5GDHCSs have already been installed [14], and although there are examples of GSHPs with different applications connected to the same heat source water network [16,17], there are no cases where multiple GSHPs supplying hot water and cold water were operated simultaneously or within a short period of time to reveal the effect of heat recovery. In addition, temperatures of the heat source water network have been presented for several cases [15], but only a few cases analyzed measured data [16,17,18,19,20]. Therefore, none of the papers presented so far have conducted long-term performance verification after optimization and performance prediction of GSHP systems and 5GDHCS using simulation. Therefore, it is important to show the effect of heat recovery when multiple GSHPs supplying hot water and cold water are operated simultaneously or during a short period of time, which is assumed in the large-scale GSHP system or 5GDHCS to be introduced in the future. In addition, the novelty of this study was to show how to optimize the operation of a HR-GSHP system by simulating and verifying the performance over a long period of time.

In this paper, the authors installed a HR-GSHP system on a trial basis in a dormitory building and developed a simulation model to propose an optimal operation method to minimize the energy consumption of the system. Then, the HR-GSHP system was operated for 8 years from 2012 to 2019, and actual measurements were carried out to verify the system’s performance and the energy saving effect of the optimal operation.

2. Outlines of Subject Building and Heat Recovery Ground Source Heat Pump

2.1. Outlines of Subject Building

Figure 2 shows the appearance of the dormitory building where the HR-GSHP system was installed. The building is a seven-story building completed in 2012, with a total floor area of approximately 9400 m². The building is equipped with energy-saving facilities and makes maximum use of solar heat, wind, underground heat, and other energy sources. In particular, the characteristics of the dormitory building mean that it consumes a large amount of energy for domestic hot water, and since individual hot water supply systems using fossil fuel gas are the norm for such buildings, CO₂ emissions are expected to occur. Therefore, this facility aims to reduce CO₂ emissions from hot water by more than 50% by installing a heat pump hot water system with a central heat source that uses solar collectors, a GSHP, and air source heat pumps (ASHPs). In addition, the introduction of LED lighting fixtures and a GSHP air-conditioning system for some of the air-conditioning is expected to reduce overall CO₂ emissions by about 20% compared to the conventional dormitory buildings. In addition to the HR-GSHP system, an energy management system compatible with the regional energy management system and a water heating operation control system compatible with the dynamic pricing have been installed.

2.2. Outlines of Heat Recovery Ground Source Heat Pump

Figure 3 shows the conceptual diagram of the HR-GSHP system and

Table 1. Specifications of the HR-GSHP system and heat source system.

Heat source equipment and subsystem	GSHP for air conditioning (×2)	Water source variable refrigerant flow (VRF) system
		Rated cooling output: 50 kW; Rated electric power for cooling: 10.5 kW
		Rated heating output: 56 kW; Rated electric power for heating: 10.5 kW
	Circulation pump in the primary side of a GSHP for air conditioning	Rated flow rate: 384 L/min
		Rated power consumption: 3.7 kW
	GSHP for hot water (×1)	Water source heat pump
		Rated heating output: 38.5 kW; Rated electric power: 11.5 kW
	Circulation pump in the primary side of a GSHP for hot water	Rated flow rate: 76 L/min
		Rated power consumption: 0.75 kW
	Cooling tower	Rated cooling output: 138.8 kW
		Rated electric power: 1.5 kW
	Thermal storage tank	Capacity: 5 m3
	Heat carrier fluid	Water
	Solar collector (×90)	Area: 1.91 m2
		Rated heating output: 1.3 kW
	ASHP for HW (×4)	Air source heat pump
		Rated heating output: 77 kW; Rated electric power: 20.7 kW
Steel pile GHE	Specification	Double U-tube in steel pile
	Steel pile diameter	Outside diameter: 0.4~0.7 m
	Fluid	Water
	U-tube specification	HDPE25A (Outside diameter: 34 mm; inside diameter: 27 mm)
	Length and number	5.9~12.4 m (Average 7.4 m), 68 piles (GHEs)
Soil condition	Undisturbed temperature	18.8 °C
	Effective thermal conductivity	3.3 W/(m·K)
	Density	1500 kg/m3
	Specific heat	2.0 kJ/(kg·K)

shows the specifications of the HR-GSHP system and heat source system. The water heating system consists of solar collectors, a GSHP for hot water (GSHP for HW), ASHPs for hot water (ASHP for HW), and a backup gas boiler, with priority given to water heating operations in this order. Since the ASHPs for HW and the gas boiler are designed to meet the hot water demand of this building, the GSHP for HW can be operated arbitrarily on an experimental basis. The installed HR-GSHP system consisted of a GSHP air conditioner (GSHP for AC), a GSHP for HW, GHEs used as foundation piles (steel foundation piles), a cooling tower, and a solar thermal storage tank, all of which are connected within one single piping loop. In general, heat recovery heat pumps require simultaneous demand for cooling and heating to be generated in order to recover heat, and if this is not the case, an auxiliary heat source such as a cooling tower is necessary. The addition of GHEs makes it possible to absorb the time difference in the occurrence of demand for cooling and heating. Cooling towers and solar thermal storage tanks serve as auxiliary heat sources to prevent excessive increases and decrease in water temperature in the piping loop. The cooling tower is activated when the fluid temperature at the GHE outlet T_GHEout exceeds the cooling tower operating temperature T_GHEoutctset. The circulation of the heat carrier fluid from the piped loop to the solar thermal storage tank is activated when the fluid temperature at the GHE outlet T_GHEout is below the circulation operating temperature T_GHEoutsset. T_GHEoutctset and T_GHEoutsset are set at 27 °C and 10 °C, respectively.

A total of 68 steel piles were buried, serving as both foundation piles and GHEs. The diameter of the piles was 400 mm or more than 700 mm, and the effective length of the piles was 5.9 to 12.4 m (average 7.4 m), which can be used as the GHEs by inserting double U-tubes. The double U-tube indirect method was used, in which two U-tubes were inserted into the steel pile and filled with water. As for the physical properties of the ground, the values of the disturbed ground temperature and effective thermal conductivity were obtained from thermal response tests. Specific heat and density are general values based on the geology of the ground.

3. Method

3.1. Measurement

Figure 4 shows the system diagram of the HR-GSHP system and the hot water supply system, as well as the main measurement points. The main measurement points were the temperature and flow rate of the fluid in the piping loop, including the GHE and the hot water supply system, and the power consumption of the heat source equipment, as shown in Figure 4. The measurement items, equipment, and units for the main measurement points are shown in Table 2. In addition to these measurement points, meteorological data such as dry-bulb temperature, humidity, horizontal global solar radiation, and wind speed were also monitored. In addition, Table 3 summarizes the calculation formulas for the heating/cooling output of the heat source equipment and the heat extraction/injection of GHEs, which are calculated from the main measurement points.

Figure 4. A system diagram of a HR-GSHP system and hot water supply system and the measurement points.

Figure 4. A system diagram of a HR-GSHP system and hot water supply system and the measurement points.

Table 2. Measurement items, equipment, and units for the main measurement points.

Point (Indicated in Figure 4)	Measuring Items	Equipment	Unit
TG1-1in, TG1-1out	Inlet and outlet fluid temperature in the primary side of GSHP1	Pt-100	°C
TG2-1in, TG2-1out	Inlet and outlet fluid temperature in the primary side of GSHP2
TGHW-1in, TGHW-1out	Inlet and outlet fluid temperature in the primary side of a GSHP for HW
TGHEin, TGHEout	Inlet and outlet fluid temperature of GHE
Tctin, Tctout	Inlet and outlet fluid temperature of cooling tower
Tscin, Tscout	Inlet and outlet fluid temperature of solar collector
Ttank	Water temperature in thermal storage tank
TGHW-2in, TGHW-2out	Inlet and outlet water temperature in the secondary side of a GSHP for HW
TAHWin, TAHWout	Inlet and outlet water temperature in the secondary side of an ASHP for HW
GfG1	Flow rate of fluid in the primary side of GSHP1	Electromagnetic flow meter	L/min
GfG2	Flow rate of fluid in the primary side of GSHP2
GfGHW-1	Flow rate of fluid in the primary side of a GSHP for HW
GfGHE	Flow rate of fluid in GHEs
Gfct	Flow rate of fluid in cooling tower
Gfsc	Flow rate of fluid in solar collector
GwGHW-2	Flow rate of water in the secondary side of a GSHP for HW
GwAHW	Flow rate of water in the secondary side of an ASHP for HW
EGSHP1	Electric energy of GSHP1	Power meter	W
EGSHP2	Electric energy of GSHP2
EGHW	Electric energy of a GSHP for HW
Ect	Electric energy of cooling tower

Table 2. Measurement items, equipment, and units for the main measurement points.

Point (Indicated in Figure 4)	Measuring Items	Equipment	Unit
TG1-1in, TG1-1out	Inlet and outlet fluid temperature in the primary side of GSHP1	Pt-100	°C
TG2-1in, TG2-1out	Inlet and outlet fluid temperature in the primary side of GSHP2
TGHW-1in, TGHW-1out	Inlet and outlet fluid temperature in the primary side of a GSHP for HW
TGHEin, TGHEout	Inlet and outlet fluid temperature of GHE
Tctin, Tctout	Inlet and outlet fluid temperature of cooling tower
Tscin, Tscout	Inlet and outlet fluid temperature of solar collector
Ttank	Water temperature in thermal storage tank
TGHW-2in, TGHW-2out	Inlet and outlet water temperature in the secondary side of a GSHP for HW
TAHWin, TAHWout	Inlet and outlet water temperature in the secondary side of an ASHP for HW
GfG1	Flow rate of fluid in the primary side of GSHP1	Electromagnetic flow meter	L/min
GfG2	Flow rate of fluid in the primary side of GSHP2
GfGHW-1	Flow rate of fluid in the primary side of a GSHP for HW
GfGHE	Flow rate of fluid in GHEs
Gfct	Flow rate of fluid in cooling tower
Gfsc	Flow rate of fluid in solar collector
GwGHW-2	Flow rate of water in the secondary side of a GSHP for HW
GwAHW	Flow rate of water in the secondary side of an ASHP for HW
EGSHP1	Electric energy of GSHP1	Power meter	W
EGSHP2	Electric energy of GSHP2
EGHW	Electric energy of a GSHP for HW
Ect	Electric energy of cooling tower

Table 3. Calculation formulas for the heating/cooling output of the heat source equipment and GHEs.

Parameters	Evaluated Items	Equation	Unit
QGSHP1-1	Heat extraction/injection rate of GSHP1	$c_{\mathrm{p}\mathrm{f}}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{G}1}\left(T_{\mathrm{G}1-1\mathrm{i}\mathrm{n}}-T_{\mathrm{G}1-1\mathrm{o}\mathrm{u}\mathrm{t}}\right)$	W
QGSHP2-1	Heat extraction/injection rate GSHP2	$c_{\mathrm{p}\mathrm{f}}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{G}2}\left(T_{\mathrm{G}2-1\mathrm{i}\mathrm{n}2}-T_{\mathrm{G}2-1\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
QGSHP1-2	Heating/Cooling output from GSHP1	$Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}1-1}+E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}1}$
QGSHP2-2	Heating/Cooling output from GSHP2	$Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}2-1}+E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}2}$
QGSHP2	Heating/Cooling output from GSHP (GSHP1 and GSHP2)	$Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}1-2}+Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}2-2}$
QGHW1	Heat extraction/injection rate of a GSHP for HW	$c_{\mathrm{p}f}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{W}-1}\left(T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{i}\mathrm{n}}-T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
QGHW2	Heating output from a GSHP for HW	$c_{\mathrm{p}w}{\rho }_{\mathrm{w}}{G}_{\mathrm{w}\mathrm{G}\mathrm{H}\mathrm{W}-2}\left(T_{\mathrm{G}\mathrm{H}\mathrm{W}-2\mathrm{i}\mathrm{n}}-T_{\mathrm{G}\mathrm{H}\mathrm{W}-2\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
QAHW	Heating output from an ASHP for HW	$c_{\mathrm{p}w}{\rho }_{\mathrm{w}}{G}_{\mathrm{w}\mathrm{A}\mathrm{H}\mathrm{W}}\left(T_{\mathrm{A}\mathrm{H}\mathrm{W}-2\mathrm{i}\mathrm{n}}-T_{\mathrm{A}\mathrm{H}\mathrm{W}-2\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
QGHE	Heating extraction/injection rate of ground heat exchangers	$c_{\mathrm{p}f}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}\left(T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{i}\mathrm{n}}-T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
Qsc	Heating output from solar collector	$c_{\mathrm{p}f}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{s}\mathrm{c}}\left(T_{\mathrm{s}\mathrm{c}\mathrm{i}\mathrm{n}}-T_{\mathrm{s}\mathrm{c}\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
Qct	Heating (cooling) output from cooling tower	$c_{\mathrm{p}f}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{c}\mathrm{t}}\left(T_{\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{n}}-T_{\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{u}\mathrm{t}}\right)$

Table 3. Calculation formulas for the heating/cooling output of the heat source equipment and GHEs.

Parameters	Evaluated Items	Equation	Unit
QGSHP1-1	Heat extraction/injection rate of GSHP1	$c_{\mathrm{p}\mathrm{f}}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{G}1}\left(T_{\mathrm{G}1-1\mathrm{i}\mathrm{n}}-T_{\mathrm{G}1-1\mathrm{o}\mathrm{u}\mathrm{t}}\right)$	W
QGSHP2-1	Heat extraction/injection rate GSHP2	$c_{\mathrm{p}\mathrm{f}}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{G}2}\left(T_{\mathrm{G}2-1\mathrm{i}\mathrm{n}2}-T_{\mathrm{G}2-1\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
QGSHP1-2	Heating/Cooling output from GSHP1	$Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}1-1}+E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}1}$
QGSHP2-2	Heating/Cooling output from GSHP2	$Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}2-1}+E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}2}$
QGSHP2	Heating/Cooling output from GSHP (GSHP1 and GSHP2)	$Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}1-2}+Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}2-2}$
QGHW1	Heat extraction/injection rate of a GSHP for HW	$c_{\mathrm{p}f}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{W}-1}\left(T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{i}\mathrm{n}}-T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
QGHW2	Heating output from a GSHP for HW	$c_{\mathrm{p}w}{\rho }_{\mathrm{w}}{G}_{\mathrm{w}\mathrm{G}\mathrm{H}\mathrm{W}-2}\left(T_{\mathrm{G}\mathrm{H}\mathrm{W}-2\mathrm{i}\mathrm{n}}-T_{\mathrm{G}\mathrm{H}\mathrm{W}-2\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
QAHW	Heating output from an ASHP for HW	$c_{\mathrm{p}w}{\rho }_{\mathrm{w}}{G}_{\mathrm{w}\mathrm{A}\mathrm{H}\mathrm{W}}\left(T_{\mathrm{A}\mathrm{H}\mathrm{W}-2\mathrm{i}\mathrm{n}}-T_{\mathrm{A}\mathrm{H}\mathrm{W}-2\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
QGHE	Heating extraction/injection rate of ground heat exchangers	$c_{\mathrm{p}f}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}\left(T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{i}\mathrm{n}}-T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
Qsc	Heating output from solar collector	$c_{\mathrm{p}f}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{s}\mathrm{c}}\left(T_{\mathrm{s}\mathrm{c}\mathrm{i}\mathrm{n}}-T_{\mathrm{s}\mathrm{c}\mathrm{o}\mathrm{u}\mathrm{t}}\right)$
Qct	Heating (cooling) output from cooling tower	$c_{\mathrm{p}f}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{c}\mathrm{t}}\left(T_{\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{n}}-T_{\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{u}\mathrm{t}}\right)$

3.2. Establishment of Simulation Model

A simulation model for the HR-GSHP system and the hot water supply system was established to investigate the optimal operation method that minimizes electric energy consumption. The outline of the simulation model is shown in Figure 5. The simulation model consists of (1) the ground temperature calculation model and GHE model, (2) the GSHP model and ASHP model, (3) the cooling tower model, and (4) the solar collector and thermal storage tank model, etc. The inputs were the meteorological data such as the outdoor air temperature and the humidity, the horizontal global solar radiation, the tap water temperature, the amount of hot water usage, the air conditioning thermal load, and the hot water thermal load, and the outputs were the components’ energy consumption. An overview of each model is given below.

(1)

Ground temperature calculation model and GHE model

The GHE was considered as a hollow cylinder in an infinite medium, and temperature changes were calculated as a transient heat transfer problem in a two-dimensional cylindrical coordinate system. The superposition of temperature fields in space was applied to calculate subsurface temperature due to heat extraction/injection from multiple GHEs.

The surface temperature change in one GHE (steel pile) among multiple GHEs (steel piles) from the initial temperature $T_{\mathrm{s}0}$ can be calculated using the following equation:

$$\Delta T_{\mathrm{s},\mathrm{i}}\left(r_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t},\mathrm{i}},t\right)=\Delta T_{\mathrm{s},\mathrm{i},\mathrm{i}}\left(r_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t},\mathrm{i}},t\right)+{\sum }_{j=1}^m\Delta T_{\mathrm{s},\mathrm{i},\mathrm{j}}\left(r_{\mathrm{d},\mathrm{i},\mathrm{j}},t\right)$$

(1)

where i is the temperature change due to heat extraction/injection of the GHE for which the surface temperature is calculated, and j is the surface temperature change due to extraction/injection of the surrounding GHE. $r_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t},\mathrm{i}}$ and $r_{\mathrm{d},\mathrm{i},\mathrm{j}}$ are the diameter of GHE (steel pile) and the interval of GHEs. The surface temperature change is calculated by the following equation, which is based on the time superposition of the temperature response calculated by the infinite cylinder model, infinite source model, etc.

$$\Delta T_{s,i,i}\left(r_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t},i},t\right)=\left(q_{\mathrm{G}\mathrm{H}\mathrm{E}}\ast {\displaystyle \frac{d{G}_{\mathrm{I}\mathrm{C}\mathrm{S}}}{dt}}\right)\left(t\right)$$

(2)

$${\Delta T}_{\mathrm{s},\mathrm{i},\mathrm{j}}\left(r_{\mathrm{d},\mathrm{i},\mathrm{j}},t\right)=\left(q_{\mathrm{G}\mathrm{H}\mathrm{E}}\ast {\displaystyle \frac{d{G}_{\mathrm{I}\mathrm{L}\mathrm{S}}}{dt}}\right)\left(t\right)$$

(3)

where $\left(q_{\mathrm{G}\mathrm{H}\mathrm{E}}\ast {\displaystyle \frac{d{G}_{\mathrm{I}\mathrm{C}\mathrm{S}}}{dt}}\right)\left(t\right)$ represents the convolution of the heat flux $q_{\mathrm{G}\mathrm{H}\mathrm{E}}\left(t\right)$ from the GHE (pile) surface and the time derivative ${\displaystyle \frac{d{G}_{\mathrm{I}\mathrm{C}\mathrm{S}}}{dt}}$ of the temperature response calculated by the infinite cylinder model. Also, $\left(q_{\mathrm{G}\mathrm{H}\mathrm{E}}\ast {\displaystyle \frac{d{G}_{\mathrm{I}\mathrm{L}\mathrm{S}}}{dt}}\right)\left(t\right)$ represents the convolution of the heat flow $q_{\mathrm{G}\mathrm{H}\mathrm{E}}\left(t\right)$ from the GHE (pile) surface and the time derivative ${\displaystyle \frac{d{G}_{\mathrm{I}\mathrm{L}\mathrm{S}}}{dt}}$ of the temperature response calculated by the infinite source theory. For details on these calculations, see previous reports [21].

In the GHE, there is heat carrier fluid circulating inside the U-tube and water filling the space between the U-tube and the steel foundation pile. The heat transfer fluid and water are each considered as one mass system, and the heat balance at a small time dt is considered. The change in the heat capacity of the heat transfer fluid can be considered as the sum of the heat transfer from the piping loop and the heat transfer from the filled water outside the U-tube. The change in the heat capacity of the water filled between the U-tube and the steel foundation pile can be considered as the sum of the heat transfer from the U-tube and the heat transfer from the surrounding ground. Therefore, the following two equations can be obtained.

Fluid circulating through U-tube:

$$c_{\mathrm{p}\mathrm{f}}{\rho }_{\mathrm{f}}{V}_{\mathrm{f}\mathrm{U}}{\displaystyle \frac{{dT}_{\mathrm{f}\mathrm{U}}}{dt}}=c_{\mathrm{p}\mathrm{f}}{\rho }_{\mathrm{f}}{G}_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}\left(T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{o}\mathrm{u}\mathrm{t}}{-T}_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{i}\mathrm{n}}\right)+K_{\mathrm{U}-\mathrm{o}\mathrm{u}\mathrm{t}}{A}_{\mathrm{U}-\mathrm{o}\mathrm{u}\mathrm{t}}\left(T_{\mathrm{W}\mathrm{G}\mathrm{H}\mathrm{E}}{-T}_{\mathrm{f}\mathrm{U}}\right)$$

(4)

Water filled in steel pile:

$$\begin{array}{c}{c}_{p\mathrm{w}f}{\rho }_w{V}_{w\mathrm{G}\mathrm{H}\mathrm{E}}{\displaystyle \frac{{dT}_{w\mathrm{G}\mathrm{H}\mathrm{E}}}{dt}}=K_{\mathrm{U}-\mathrm{o}\mathrm{u}\mathrm{t}}{A}_{\mathrm{U}-\mathrm{o}\mathrm{u}\mathrm{t}}\left(T_{\mathrm{f}\mathrm{U}}-T_{W\mathrm{G}\mathrm{H}\mathrm{E}}\right)+\\ K_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t}}{A}_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t}}\left(T_{\mathrm{s}}{\left(r_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t},}t\right)-T}_{\mathrm{w}\mathrm{G}\mathrm{H}\mathrm{E}}\right)\end{array}$$

(5)

where $K_{\mathrm{U}-\mathrm{o}\mathrm{u}\mathrm{t}}$ in Equations (4) and (5) is the overall heat transfer coefficient based on the U-tube outer surface area from the water filling inside the steel pile to the heat transfer fluid Appendix A, and $A_{\mathrm{U}-\mathrm{o}\mathrm{u}\mathrm{t}}$ is the U-tube outer surface area. $K_{\mathrm{U}-\mathrm{o}\mathrm{u}\mathrm{t}}$ in Equation (4) is the overall heat transfer coefficient based on the outer surface area of the steel pile from the surface of the GHE (steel pile) to the water filled inside the steel pipe Appendix A, and $A_{\mathrm{U}-\mathrm{o}\mathrm{u}\mathrm{t}}$ is the pile outer surface area. By giving $q_{\mathrm{G}\mathrm{H}\mathrm{E}}\left(t\right)$ at each time step from the equation $q_{\mathrm{G}\mathrm{H}\mathrm{E}}\left(t\right)=K_{\mathrm{G}\mathrm{H}\mathrm{E}-out}\left(T_s{\left(r_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t},}t\right)-T}_{\mathrm{w}\mathrm{G}\mathrm{H}\mathrm{E}}\right)$, the surface temperature of the GHE can be calculated as $T_s\left(r_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t},}t\right)=T_{s0}\left(r_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t},}t\right)+\Delta T_{\mathrm{s},\mathrm{i}}\left(r_{\mathrm{G}\mathrm{H}\mathrm{E}-\mathrm{o}\mathrm{u}\mathrm{t},i},t\right)$ and the temperature of the heat carrier fluid at the outlet of GHE can be calculated as $T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{o}\mathrm{u}\mathrm{t}}=T_{\mathrm{f}\mathrm{U}}$, respectively. As a result, the temperatures in the GHE and the surrounding ground can be evaluated; the temperatures for GHEs placed in series, in parallel, etc., were calculated using the method described in a previous report [22].

Using the developed GHE model, the outlet temperature $T_{\mathrm{G}\mathrm{H}\mathrm{E}out}$ was calculated by giving the inlet temperature $T_{\mathrm{G}\mathrm{H}\mathrm{E}in}$ and circulation flow rate of the GHE G_fGHE and compared with the outlet temperature $T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{o}\mathrm{u}\mathrm{t}}$ obtained from actual measurements for verification. Figure 6 shows the comparison results. As shown in Figure 6, it was confirmed that the calculated values reproduced the measured values well by giving 3.3 W/(m-K) as the effective thermal conductivity of the ground.

(2)

GSHP model and ASHP model

The GSHP for AC, which is the subject of performance prediction in this paper, was the water cooled variable refrigerant flow (VRF) air conditioning system. The calculation model was produced by referring to a paper [23]. In this model, the power consumption of heat pump $E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}}=E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}1}+E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}2}$ during cooling operation (Q_GSHP-2 < 0) was calculated by the following equation [23].

$$E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}}=\left(Q_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{c}}\right)\left(EI{R}_{\mathrm{c}}\right)\left(P{F}_{\mathrm{c}}\right)$$

(6)

Here, Q_totalc is the total cooing capacity of all indoor units connected to one outdoor unit. EIR_c is the energy input ratio, which is the reverse of COP. PF_c is the modifier for operation performance under the part-load condition based on the total cooling capacity. (EIR_c)(PF_c) can be approximated as a function of the wet bulb temperature of the air entering the indoor unit T_aiWB, the inlet temperature at the primary side of heat pump unit T_G1-1in, the flow rate in the primary side G_fG = G_fG1 + G_fG2, and the hourly cooling load Q_GSHP-2.

$$\left(EI{R}_{\mathrm{c}}\right)\left(P{F}_{\mathrm{c}}\right)\cong f\left(T_{\mathrm{a}\mathrm{i}\mathrm{W}\mathrm{B}},T_{\mathrm{G}1-1\mathrm{i}\mathrm{n}},G_{\mathrm{f}\mathrm{G}},Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}-2}\right)$$

(7)

On the other hand, the power consumption of heat pump $E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}}=E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}1}+E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}2}$ during heating operation (Q_GSHP-2 > 0) was calculated by the following equation.

$$E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}}=\left(Q_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{h}}\right)\left(EI{R}_{\mathrm{h}}\right)\left(P{F}_{\mathrm{h}}\right)$$

(8)

Here, Q_totalh is the total heating capacity of all indoor units connected to one outdoor unit. EIR_h is the energy input ratio, which is the reverse of COP. PF_h is the modifier for operation performance under the part-load condition based on the total heat capacity. (EIR_h)(PF_h) can also be approximated as a function of the dry-bulb temperature of the air entering the indoor unit (T_aiDB), T_G1-1in, G_fG, and Q_GSHP-2.

$$\left(EI{R}_{\mathrm{h}}\right)\left(P{F}_{\mathrm{h}}\right)\cong f\left(T_{\mathrm{a}\mathrm{i}\mathrm{D}\mathrm{B}},T_{\mathrm{G}1-1\mathrm{i}\mathrm{n}},G_{\mathrm{f}\mathrm{G}},Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}-2}\right)$$

(9)

Comparing the COP of the heat pump calculated by the above equation with the COP of the heat pump obtained by actual measurement, the model was modified, resulting in the COP according to the heat pump inlet temperature and heat output, as shown in Figure 7.

Next, the heat extraction rate (heat exchange rate) $Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}-1}$ at the primary side of the GSHP unit can be calculated by the following equation, using $Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}-2}$ and $E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}}$.

$$Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}-1}=Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}-2}-E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}}$$

(10)

Furthermore, $Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}-1}$ can also be expressed in the following manner.

$$Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}-1}=c_{pf}{\rho }_f{G}_{f\mathrm{G}-1}\left(T_{\mathrm{G}-1\mathrm{i}\mathrm{n}}-T_{\mathrm{G}-1\mathrm{o}\mathrm{u}\mathrm{t}}\right)$$

(11)

The outlet temperature $T_{\mathrm{G}-1\mathrm{o}\mathrm{u}\mathrm{t}}$ of the primary side of the GSHP unit can be calculated using the following equation.

$$T_{\mathrm{G}-1out}=T_{\mathrm{G}-1\mathrm{i}\mathrm{n}}-{\displaystyle \frac{Q_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}-1}}{c_{pf}{\rho }_f{G}_{f\mathrm{G}-1}}}$$

(12)

Next, for the GSHP for HW, the water supply temperature T_GHW-2out = 65 °C was set, and the COP according to the heat pump inlet temperature, as shown in Figure 8, was obtained based on the technical data and actual measurements of heat pumps. The power consumption $E_{\mathrm{G}\mathrm{H}\mathrm{W}}$ can be obtained from the following equation.

$$E_{\mathrm{G}\mathrm{H}\mathrm{W}}=Q_{\mathrm{G}\mathrm{H}\mathrm{W}2}/\mathrm{C}\mathrm{O}\mathrm{P}$$

(13)

Then, the heat extraction rate $Q_{\mathrm{G}\mathrm{H}\mathrm{W}1}$ in the primary side can be calculated by the following equation, using $Q_{\mathrm{G}\mathrm{H}\mathrm{W}2}$ and $E_{\mathrm{G}\mathrm{H}\mathrm{W}}$.

$$Q_{\mathrm{G}\mathrm{H}\mathrm{W}1}=Q_{\mathrm{G}\mathrm{H}\mathrm{W}2}-E_{\mathrm{G}\mathrm{H}\mathrm{W}}$$

(14)

Furthermore, the outlet temperature $T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{o}\mathrm{u}\mathrm{t}}$ of the primary side can be calculated using the following equation.

$$T_{\mathrm{G}\mathrm{H}\mathrm{W}-1out}=T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{i}\mathrm{n}}-{\displaystyle \frac{Q_{\mathrm{G}\mathrm{H}\mathrm{W}1}}{c_{pf}{\rho }_f{G}_{f\mathrm{G}\mathrm{H}\mathrm{W}-1}}}$$

(15)

In the absence of cooling tower operation, the GHE inlet temperature $T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{i}\mathrm{n}}$ can be obtained by the following equation, in which the primary side outlet temperature of the GSHP for air conditioning $T_{\mathrm{G}-1\mathrm{o}\mathrm{u}\mathrm{t}}$ and the primary side outlet temperature of the GSHP for hot water supply $T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{o}\mathrm{u}\mathrm{t}}$ are weighted by the circulation flow rate.

$$T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{i}\mathrm{n}}={\displaystyle \frac{T_{\mathrm{G}-1\mathrm{o}\mathrm{u}\mathrm{t}}{G}_{f\mathrm{G}-1}+T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{o}\mathrm{u}\mathrm{t}}{G}_{f\mathrm{G}\mathrm{H}\mathrm{W}-1}}{G_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}}}$$

(16)

Here, $G_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}=G_{f\mathrm{G}-1}+G_{f\mathrm{G}\mathrm{H}\mathrm{W}-1}$.

For the COP of the ASHP for HW, COP_AHW = 2.4 was set as constant. The power consumption $E_{\mathrm{A}\mathrm{H}\mathrm{W}}$ can be obtained from the following equation.

$$E_{\mathrm{A}\mathrm{H}\mathrm{W}}=Q_{\mathrm{A}\mathrm{H}\mathrm{W}}/{COP}_{\mathrm{A}\mathrm{H}\mathrm{W}}$$

(17)

(3)

Cooling tower model

The cooling capacity of the cooling tower Q_ct was approximated as a function of the cooling tower inlet temperature T_ctin, outside wet bulb temperature T_aWB, and flow rate G_fct.

$$Q_{\mathrm{c}\mathrm{t}}\cong f(n_{\mathrm{c}\mathrm{t}},T_{\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{n}},T_{\mathrm{a}\mathrm{W}\mathrm{B}},G_{\mathrm{f}\mathrm{c}\mathrm{t}})$$

(18)

where $n_{\mathrm{c}\mathrm{t}}$ is the model number (scale) of the cooling tower and this value is a constant.

T_aWB can be calculated iteratively using the outdoor dry-bulb temperature T_aDB and the absolute humidity, which are obtained from measured data. The cooling tower outlet temperature T_ctout is calculated by the following equation.

$$T_{\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{u}\mathrm{t}}=T_{\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{n}}-{\displaystyle \frac{Q_{\mathrm{c}\mathrm{t}}}{G_{\mathrm{f}\mathrm{c}\mathrm{t}}{c}_{p\mathrm{f}\mathrm{c}\mathrm{t}}{\rho }_{\mathrm{f}\mathrm{c}\mathrm{t}}}}$$

(19)

Figure 9 shows the Q_ct calculated for the cooling tower used in this building, given T_aWB and T_ctin, with G_fct = 390 L/min. In addition, when the cooling tower is in operation, it consumes the rated amount of power, thus,

$$E_{\mathrm{c}\mathrm{t}}=1.5 \mathrm{k}\mathrm{W}$$

(20)

The GHE inlet temperature T_GHEin in the presence of cooling tower operation can be obtained by the following equation. T_GHEin is determined by an equation that weights the temperature of the water after it passes through the cooling tower and the temperature of the water that bypasses the cooling tower but does not pass through the cooling tower by the flow rate.

$$\begin{array}{c}{T}_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{i}\mathrm{n}}=\left({\displaystyle \frac{T_{\mathrm{G}-1\mathrm{o}\mathrm{u}\mathrm{t}}{G}_{f\mathrm{G}-1}+T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{o}\mathrm{u}\mathrm{t}}{G}_{f\mathrm{G}\mathrm{H}\mathrm{W}-1}}{G_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}}}-{\displaystyle \frac{Q_{\mathrm{c}\mathrm{t}}}{G_{\mathrm{f}\mathrm{c}\mathrm{t}}{c}_{p\mathrm{f}\mathrm{c}\mathrm{t}}{\rho }_{\mathrm{f}\mathrm{c}\mathrm{t}}}}\right){\displaystyle \frac{G_{\mathrm{f}\mathrm{c}\mathrm{t}}}{G_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}}}\\ +{\displaystyle \frac{T_{\mathrm{G}-1\mathrm{o}\mathrm{u}\mathrm{t}}{G}_{f\mathrm{G}-1}+T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{o}\mathrm{u}\mathrm{t}}{G}_{f\mathrm{G}\mathrm{H}\mathrm{W}-1}}{G_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}}}{\displaystyle \frac{G_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}{-G}_{\mathrm{f}\mathrm{c}\mathrm{t}}}{G_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}}}\end{array}$$

(21)

(4)

Solar collector and thermal storage tank model

The heat collection rate of the solar collector was calculated after calculating the slope solar radiation from the horizontal global solar radiation, using the following equation.

$$Q_{\mathrm{s}\mathrm{c}}=J_{\mathrm{s}\mathrm{l}}{\eta }_{\mathrm{s}\mathrm{c}}{A}_{\mathrm{s}\mathrm{c}}$$

(22)

where J_sl is the slope solar radiation, η_sc is the solar collector efficiency, and A_sc is the solar collector area. The water in thermal storage tank was regarded as one mass system as the heat carrier fluid in the GHE, and the water temperature was calculated based on the heat balance equation.

$$c_{\mathrm{p}\mathrm{w}}{\rho }_{\mathrm{w}}{V}_{\mathrm{w}\mathrm{t}}{\displaystyle \frac{{dT}_{\mathrm{w}\mathrm{t}}}{dt}}=Q_{\mathrm{s}\mathrm{c}}+c_{\mathrm{p}\mathrm{w}}{\rho }_{\mathrm{w}}{G}_{\mathrm{h}\mathrm{w}}\left(T_{\mathrm{t}\mathrm{w}}{-T}_{\mathrm{w}\mathrm{t}}\right)+K_{\mathrm{e}\mathrm{x}}{A}_{\mathrm{e}\mathrm{x}}\left(T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{o}\mathrm{u}\mathrm{t}}{-T}_{\mathrm{w}\mathrm{t}}\right)$$

(23)

The first term on the right side is the heat collection rate of the solar collector, the second term is the heat flow from the thermal storage tank to the hot water tank, and the third term indicates the heat exchange between the heat carrier fluid in the piping loop and the water in the heat storage tank when the heat carrier fluid is circulated to the thermal storage tank. The third term occurs only when the circulation of the heat carrier fluid to the thermal storage tank is activated. G_hw is the hot water consumption (the amount of water supplied from the thermal storage tank to the hot water tank), T_tw is the tap water temperature, and K_ex and A_ex are the heat transfer coefficient and area of the heat exchanger that exists between the heat carrier fluid from the piped loop and the water in the thermal storage tank. If the heat carrier fluid is heated by the water in the tank, the GSHP primary inlet temperatures $T_{\mathrm{G}-1\mathrm{i}\mathrm{n}} \mathrm{a}\mathrm{n}\mathrm{d} T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{i}\mathrm{n}}$ are determined by the following equation, which weights the fluid temperature after it passes through the storage tank and the fluid temperature that bypasses the tank and does not pass through the tank by the flow rate.

$$T_{\mathrm{G}-1\mathrm{i}\mathrm{n}}=T_{\mathrm{G}\mathrm{H}\mathrm{W}-1\mathrm{i}\mathrm{n}}=\left(T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{o}\mathrm{u}\mathrm{t}}-{\displaystyle \frac{K_{\mathrm{e}\mathrm{x}}{A}_{\mathrm{e}\mathrm{x}}\left(T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{o}\mathrm{u}\mathrm{t}}{-T}_{\mathrm{w}\mathrm{t}}\right)}{c_{pf}{\rho }_f{G}_{f\mathrm{w}\mathrm{t}}}}\right){\displaystyle \frac{G_{f\mathrm{w}\mathrm{t}}}{G_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}}}+T_{\mathrm{G}\mathrm{H}\mathrm{E}\mathrm{o}\mathrm{u}\mathrm{t}}{\displaystyle \frac{G_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}-G_{f\mathrm{w}\mathrm{t}}}{G_{\mathrm{f}\mathrm{G}\mathrm{H}\mathrm{E}}}}$$

(24)

In addition, when T_tw is below 15 °C, the water in the storage tank is heated by ASHP. The electric power consumption of the heating of water in the tank, E_tank, is calculated by the following equation.

$$E_{\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{k}}=c_{\mathrm{p}\mathrm{w}}{\rho }_{\mathrm{w}}{V}_{\mathrm{w}\mathrm{t}}\left(15{-T}_{\mathrm{w}\mathrm{t}}\right)/{COP}_{\mathrm{A}\mathrm{S}\mathrm{H}\mathrm{P}}$$

(25)

Here, the COP of the ASHP, COP_ASHP = 2.4, was set as constant.

Figure 10 shows the calculation flow for the HR-GSHP system. Given the specifications of the equipment and the physical properties of the ground as the calculation conditions, and given the heating and cooling load and hot water supply load at each time of day, the temperature and energy consumption of each part of the system can be calculated.

To validate the simulation model, actual measured values of air conditioning load and hot water heating load in 2013 were given and simulated to compare with the measured values. The simulation model shown in Figure 5 was used and the specifications of each device shown in Table 1 were given as the calculation conditions. The effective thermal conductivity of the ground was given as 3.3 W/(m-K).

Table 4. Comparison of measured and simulated electric energy consumption.

	Electric Energy Consumption [kWh]
	Storage Tank Heating	GSHP for AC		Cooling Tower	ASHP for HW	GSHP for HW	Total
		Cooling	Heating
Measurement	0	29,722	24,319	535	147,123	12,337	214,036
Calculation	259	29,244	25,235	797	144,689	12,166	212,390
Relative error [%]	-	1.6	3.8	49.0	1.7	1.4	0.8

shows a comparison of measured and simulated electric energy consumption. The table shows that the error between the measured and simulated values is within about 5%, confirming the reproducibility of the simulation. Although the error in the cooling tower is large (approximately 50%), it does not have a significant impact on the simulation results because the error is small and accounts for only a small percentage of the total error.

3.3. Operation and Performance Evaluation of the HR-GSHP System

The HR-GSHP system was actually operated for eight years from 2012 to 2019, and the optimal operation method that minimizes the energy consumption of the whole system was investigated. Since equipment malfunctions occur during actual operation, load conditions varied each year depending on weather conditions and facility usage, and as simulation results do not completely reproduce actual measurements, adjustments were made to the operation method to minimize energy consumption while confirming the actual measurement results.

Table 5. Adjustments and changings during operation.

Date	Details of the Changing the Operation
April 2013	Operation of the GSHP for HW based on simulation results (April–November)
April 2014	Extension of operation period of the GSHP for HW (all year round)
2014, 2015	GSHP for HW operation failure (August–September)
April 2016	Suspension of cooling tower operation due to malfunction
April 2018	Resumption of cooling tower operation

shows the specific adjustments made to the operation. Table 5 also includes changes in operation due to equipment malfunctions. Starting from April 2013, the second year of operation, the optimal operating period of the GSHP for HW obtained by simulation was modified. Based on the results of the first and second year of operation, the temperature TGHEout in winter only dropped to 12–14 °C. Therefore, the authors extended the operation period of the GSHP for HW from 8 to 12 months, considering that the temperature TGHEout would be below 10 °C for a small period of time even if the GSHP for HW was operated during the winter season. In August and September of 2014 and 2015, the GSHP for HW experienced operational failures. In addition, the cooling tower failed in April 2016, and the cooling tower operations were suspended in 2016 and 2017. The cooling tower was restored in 2018.

4. Results

4.1. Investigation of Optimal Operation Method by Simulation Model

The established simulation model was used to investigate the optimal operation method. Specifically, the optimal operation method was to minimize the total energy consumption of the system E_system, as shown in the equation below, by suppressing excessive temperature increases and decreases.

$$E_{\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}}=E_{\mathrm{G}\mathrm{S}\mathrm{H}\mathrm{P}}+E_{\mathrm{G}\mathrm{H}\mathrm{W}}+E_{\mathrm{A}\mathrm{W}\mathrm{H}}+E_{\mathrm{C}\mathrm{T}}+E_{\mathrm{T}\mathrm{a}\mathrm{n}\mathrm{k}}$$

(26)

In addition, the following constraints were given based on the operable range of inlet temperature T_G1-1in in GSHP for AC.

$$T_{\mathrm{G}1-1\mathrm{i}\mathrm{n}}\le 30 °\mathrm{C}, T_{\mathrm{G}1-1\mathrm{i}\mathrm{n}}\ge 10 °\mathrm{C}$$

(27)

Of the equipment comprising the HR-GSHP system, the GSHP for AC must independently provide the air conditioning load for the part to be air-conditioned, while the GSHP for HW has an ASHP for HW, which can provide the whole of heating load for HW, installed as an auxiliary heat source, and the operation time can be set arbitrarily. Therefore, in this system, simulations were conducted by changing the operation period of the GSHP for HW at monthly intervals, and the operation period with the lowest energy consumption was set as the optimal operation condition, and the actual system was also operated under this condition.

Table 6. Conditions of operation period of the GSHP for HW.

	Operating Period of the GSHP for HW		Operating Period of the GSHP for HW
CASE0	Nothing (0 month)	CASE6	May~October (6 months)
CASE1	August (1 month)	CASE7	May~November (7 months)
CASE2	July~August (2 months)	CASE8	April~November (8 months)
CASE3	July~September (3 months)	CASE9	April~December (9 months)
CASE4	June~September (4 months)	CASE10	March~December (10 months)
CASE5	June~October (5 months)	CASE11	March~January (11 months)

shows the conditions of the operation period of the GSHP for HW. In conducting the simulation for setting the optimum operating conditions, the actual measured values for 2013 shown in Figure 11 and Figure 12 were used for the air conditioning load and hot water supply load. For the hot water supply load, the operation period of the GSHP for HW was assumed to be the period shown in Table 6. Since the rated output of the GSHP for HW was 38.5 kW, the hot water load was given as the total hot water load for each time period shown in Figure 12 if it was less than 38.5 kW and 38.5 kW if it was greater than 38.5 kW. Therefore, when the GSHP for HW was operated all year round, the GSHP for HW load was as shown in Figure 12.

As a result of the simulation, the energy consumption of the whole HR-GSHP system for the operation period of the GSHP for HW is shown in Figure 13. The energy consumption required for water heating decreased when the operation period of the GSHP for HW is increased because the COP of the GSHP for HW was higher than that of ASHP for HW. However, when the GSHP for HW was operated even in winter, the temperature of heat carrier fluid in the piping loop decreased, resulting in heat exchange between the water in the thermal storage tank. This requires energy to heat the thermal storage tank, and this increases energy consumption. Therefore, the condition with the lowest E_system was the case where the GSHP for HW was operated for 8 months from April to November.

4.2. Performance Evaluation of the HR-GSHP System for Each Year

The performance of the HR-GSHP system was evaluated for each year, and the effects of optimal operation on system performance and energy consumption reduction were analyzed.

Since energy consumption increases or decreases depending on load conditions in actual operation, the following items were used to evaluate whether optimal operation was being performed. Since the HR-GSHP system was expected to perform well during periods when both heating (or hot water) and cooling demand occurred, the evaluation was conducted mainly during the cooling period (April to November), when both hot water demand (hot water heating load) and cooling demand (cooling load) occurred at the same time.

(1)

Piping loop heat carrier fluid temperature.

(2)

GSHP for AC performance during the cooling period.

(3)

Hot water system performance.

(4)

Overall system energy consumption.

The results of the evaluation are presented below.

(1)

Piping loop heat carrier fluid temperature.

T_GHEout was evaluated as a representative value of the piping loop heat carrier fluid temperature. The hourly variation in T_GHEout is shown in Figure 14. It can be seen that T_GHEout rises to about 30 °C until the fourth year, except for an instantaneous rise in summer. It was confirmed that the temperature rise was suppressed by the operation of the cooling tower as an auxiliary heat source when T_GHEout exceeded 27 °C until the fourth year. On the other hand, in the fifth and sixth years, the cooling tower was not operated, resulting in a temperature rise of approximately 35 °C in the fifth year and 37 °C in the sixth year. The temperature T_GHEout was around 12~14 °C until the second year, and dropped to around 10 °C after the third year. This is because the GSHP for HW was operated during the winter season after the third year, since the results of the first and second years of operation indicated that the measured T_GHEout was higher than the simulated T_GHEout. After the third year of operation of the GSHP for HW during the winter season, the minimum and maximum temperatures were about 10 °C and 30 °C, respectively, except in the fifth and sixth years when the cooling tower was not operated, confirming that the system was operated properly.

(2)

GSHP for AC performance during the cooling period

Figure 15 shows the total cooling output and electric energy consumption of the GSHP for AC (GSHP1 and GSHP2), COP (=(Q_GSHP1-2 + Q_GSHP2-2)/(E_GSHP1 + E_GSHP2)), and SCOP (=(Q_GSHP1-2 + Q_GSHP2-2)/(E_GSHP1 + E_GSHP2 + E_ct)) from the first year (2012) to the eighth year (2019). The COP in the fifth and sixth years was 2.8 and 3.0, respectively, due to the increase in the heat source water temperature caused by the cooling tower shutdown. In the eighth year, the operation of the cooling tower was minimized and the SCOP decreased due to the operation of the cooling tower, resulting in COP and SCOP both improving to 3.8. The reason why the cooling tower operation was minimized in the eighth year is discussed in the discussion in Section 4.1.

(3)

Hot water system performance

Figure 16 shows the integrated values of heat output and power consumption of the GSHP for HW, the ASHPs for HW, the solar collectors that make up the water heating system, and the COP (=(Q_GHW2 + Q_AHW2 + Q_s)/(E_GHW + E_AHW)) from the first year (2012) to the eighth year (2019). For hot water heating, the COP values had been low for a period of time since 2014, when the GSHP for HW began in earnest, but returned to their original levels in the eighth year.

The COP values were calculated by summing up the heating output from the three pieces of equipment (GSHP for HW, ASHPs for HW, and solar collectors) and the total energy consumption of the GSHP for HW and ASHPs for HW. Since the ratio of the heating output from the solar collectors was small, the COP was considered to have improved as the operation of the GSHP for HW increased.

(4)

Overall system energy consumption

Figure 17 shows the energy consumption of the whole system during the cooling season (form April to November). Here, the “All ASHP” case is the energy consumption calculated by using the measured thermal load and COP (cooling COP = 2.6, heating COP = 2.0, and hot water COP = 2.4) for the case where only the ASHP operates for air conditioning and hot water supply without heat recovery. Figure 16 shows that the reduction in the eighth year was 17%, which was the largest. As explained above, the reason for this is that the GSHP for HW operated more in the summer of the eighth year, and the operation of the cooling tower was suppressed due to heat extraction by the GSHP for HW, and the GSHP for HW had a higher COP than the ASHP for HW, which reduced power consumption for hot water.

Compared to the conventional GSHP system for AC and the GSHP system for HW installed independently, the GSHP for HW improved the COP by recovering the waste heat occurred by the GSHP for AC during the cooling period. In addition, the operation of the cooling tower was reduced by the heat extraction of the GSHP for HW. To explain the specific effects in this system, the improvement in the COP of the GSHP for HW during the cooling period was about 6%, since TGHW-1in rose by approximately 10 °C, resulting in a reduction in electricity consumption of approximately 2100 kWh. The reduction in electricity consumption due to the reduction in cooling tower operation can be estimated to be at least approximately 3000 kWh based on a comparison of the first and fourth years. The total reduction in electricity consumption was, thus, approximately 5100 kWh, which was 2.5% of the total electricity consumption.

The CO₂ emission reduction rate was examined based on the electricity consumption over the 8-year period; the overall reduction in electricity consumption over the 8-year period was 232 MWh, or a reduction rate of about 12%. Since this system did not use a gas boiler and all energy was provided by electricity, the electricity consumption and CO₂ emissions were considered to be in a proportional relationship and, thus, the CO₂ emission reduction rate was also about 12%. The cost reduction effect was estimated to be approximately JPY 5.56 million, given that the electricity rate in this area is approximately 24 JPY/kWh.

5. Discussion

5.1. Heat Recovery Effect in the HR-GSHP System

In this section, the authors analyzed the effect of the GSHP for HW operation (heat extraction) results on the HR-GSHP system and discussed how to operate the system to achieve the best heat recovery. Analysis was conducted on the following items.

(1)

Heat recovery-related equipment (GSHP, GHE, and cooling tower) during the cooling period (April to November).

Figure 18 shows the comparison results of the heat injection rate of the GSHP for AC, the heat extraction rate of the GSHP for HW, the heat injection rate of the GHE, and the heat release rate of the cooling tower from April to November in the first year (2012) to the eighth year (2019). Under normal circumstances, the heat injection from the GSHP for AC was handled by the heat injection to the ground via the GHE and the heat release to the atmosphere from the cooling tower, while the heat extraction from the GSHP for HW was handled by the heat extraction from the ground via the GHE. In the first and second years, the GSHP for HW operation time was still small, so the heat release from the cooling tower was large. In the eight year, the heat extraction from the GSHP for HW increased compared to the other years, and it was confirmed that the heat injection to the ground via the GHE and the heat release from the cooling tower was smaller than the heat injection from the GSHP for AC. The increase in heat extraction from the GSHP for HW increased the effectiveness of heat recovery. This minimizes the increase in heat carrier fluid temperature due to heat injection via the GHE and the cooling tower operation. Then, it yielded the highly efficient operation of the GSHP for AC, as shown in Figure 15.

(2)

Monthly variation of heat extraction/injection rate

Comparing year 4 and year 8 in Figure 18, it can be seen that the heat injection from the GSHP for AC and the heat injection via the GHE are comparable and that the heat release from the cooling tower is larger in year 4, while the heat extraction from the GSHP for HW is larger in year 8. The reason for this is discussed by showing the monthly variation in the heat extraction/injection rate. Figure 19 shows the monthly extraction/injection rate in the fourth year, and Figure 20 shows the monthly heat extraction/injection rate in the eighth year. From the figure, it can be seen that in the eighth year, the amount of heat extracted from the GSHP for HW was larger, especially from June to September. This minimized the heat release from the cooling tower, which led to an improvement in the SCOP during the cooling period. The above results confirm that, when operating the HR-GSHP system, the ideal operation is to generate heat extraction and heat injection simultaneously, as much as possible, to suppress changes in the heat carrier fluid temperature and improve the system operation efficiency.

5.2. Shortcoming of the Study

In this building, the heat recovery effect could not be achieved in winter because there was no cooling demand in winter. When this system is installed in the future, it will be important to combine it with facilities that generate cooling demand even in winter, such as cold storage warehouses, computer rooms, and data centers. Future tasks include planning regional facilities in consideration of the heating and cooling demand of each facility, constructing an HR-GSHP system that can effectively recover heat, and developing technology to predict the heat recovery and energy conservation effects of the HR-GSHP system when it is introduced.

6. Conclusions

In this paper, a HR-GSHP system, in which the primary piping of a GSHP for air conditioning and a GSHP for hot water were connected to a GHE and each heat pump was operated simultaneously or within a short period of time, was installed in a dormitory building on a trial basis, and the optimal operation method to minimize the energy consumption of the system was investigated using simulation. The study was conducted to determine the optimal operation method to minimize the energy consumption of the system using simulation. Furthermore, the HR-GSHP system was operated for 8 years from 2012 to 2019, and actual measurements were conducted to verify the system’s performance and the energy saving effect in optimal operation. The results are shown below.

• The operating period of the GSHP for hot water supply was changed, and the optimal operation method of the HR-GSHP system that minimizes the energy consumption of the system was studied using simulation. As a result, the operation period of the GSHP for HW that minimizes the energy consumption of the system is 8 months from April to November.
• Based on the operation results of the HR-GSHP system up to the second year, the temperature drop in the piping loop was 12~14 °C, which was smaller than the simulation. Therefore, the GSHP for HW was operated year-round after the third year. The minimum temperature of heat carrier fluid was about 10 °C and the maximum temperature was about 30 °C, except for the fifth and sixth years when the cooling tower failed, confirming that the system was being operated properly.
• Verification of the system performance based on the actual measurement results of the HR-GSHP system for 8 years showed that the SCOP for cooling in the eighth year of operation was the highest, and the COP for hot water supply was also the second highest. The reasons for this are that the GSHP for HW operated more in the summer in the eighth year, which reduced the operation of the cooling tower due to the heat extraction of the GSHP for HW, and that the GSHP for HW had a higher COP than the ASHP for HW. The reduction in power consumption compared to the case where air conditioning and hot water supply were operated only with the ASHP was also the largest in the eighth year, with a reduction of approximately 17%.
• Analysis of the amount of heat taken from the GSHP, the GHE, and the cooling tower during the cooling period in each year showed that the amount of heat taken from the GSHP for HW increased in the eighth year compared to the other years, thereby reducing the amount of heat released from the GHE and cooling tower relative to the amount of heat released by the GSHP for AC. In addition, it was also confirmed that the amount of heat extracted from the GSHP for HW increased from June to September, especially in the eighth year, and that when operating the HR-GSHP system, the ideal operation is to generate heat extraction and heat dissipation simultaneously, as much as possible, to suppress changes in the heat source water temperature and increase the system operation efficiency. The actual operation results confirm this.