Utilizing a Domestic Water Tank to Make the Air Conditioning System in Residential Buildings More Sustainable in Hot Regions

Abstract

Air conditioning (AC) is considered an important requirement for hot environments, but it is facing one of the most demanding obstacles as concerns the use of electrical energy resources. In 2019, electricity consumption in the residential sector in Gulf Cooperation Council states reached approximately 43% of the total national consumption, and about two-thirds of the electrical energy consumed in residential buildings (RBs) was used for AC. Therefore, as these indicators show, there is a need to focus on studying AC. One of the most important reasons for the high consumption of electrical energy in RBs is the big difference between indoor and outdoor temperatures. In this paper, a heat exchanger was designed and tested experimentally to reduce this temperature difference by using a domestic ground water tank (GWT) as a sink/source (water-cooled condensers instead of air-cooling). The results have shown that the water tank made the surrounding temperature around the external coil of the AC more suitable for cooling/heating. The proposed system resulted in a reduction in energy consumption by 28% of the electrical energy needed in the conventional system and an increase in COP by 39%. This means that this system is more efficient and therefore more sustainable.

1. Introduction

Energy is an essential need for all humanity, but energy consumption products are a major problem in the world. The main sources of energy are conventional sources, such as coal, oil, and natural gas, and they are harming the environment. These sources, by most measures, cause more substantial harm than those of renewable energy. The harm caused includes air and water pollution; damage to public health, wildlife, water use, land use; and emissions that lead to more global warming. In 2019, the global energy demand increased by less than half the rate of growth in 2018, well below the average rate for the first time since 2010. This deceleration was due mainly to slower global economic growth, mainly because of COVID-19 [1]. Global energy consumption is the total volume of energy—typically measured on a yearly basis—that is harnessed from all sources and used in all activities, across all industrial and transport sectors—in addition to others—in all countries. For example, in the Kingdom of Saudi Arabia (KSA), the oil and gas consumed in 2018 and 2019 totaled 136.7 and 139.8 Mtoe respectively. The “other purposes” means the use of energy for electric power production, agriculture, and services, as shown in

Table 1. Total oil and gas consumption in the KSA in 2018 and 2019 (Mtoe) [2].

	Industry	Transportation	Non-Energy	Other
2018	37.0	45.5	29.9	24.3
2019	37.9	45.6	33.0	23.3

. In 2019, electricity consumption in the residential sector in Gulf Cooperation Council (GCC) states reached approximately 43% of the total national consumption. In 2018 and 2019, the residential consumption of electricity in Saudi Arabia, which covers the supply for 7.7 million customers in residential areas, accounted, respectively, for about 43.6 and 44.5% of the national consumption [2]. Their electricity is highly used for air conditioning (AC) and cooling due to climate conditions where the demand for electrical energy increases dramatically in the hot summer months (maximum grid power in summer was 62.076 GW, and the minimum in winter was 33.440 GW in the KSA in 2019 [3]). Moreover, studies show that about two-thirds of the electricity consumption in RBs go for AC in GCC [4,5,6] (see

Table 2. Shares of AC in the total electricity consumption.

Country	Location	Share of AC (%)	Reference and Date
Bahrain	Manama	73	Krarti and Dubey [7] 2018
Kuwait	Kuwait	39–49	Azar et al. [8] 2021
	Aqaila	60	Park et al. [9] 2019
Oman	Oman	57 to 71	Al-Saadi and Al-Jabri [10] 2020
KSA	Qassim and Riyadh	70 and 72	Almushaikah and Almasri [11] 2020
	KSA	More than 65	Krarti et al. [12] 2020
	Qassim	66	Esmaeil et al. [13] 2019
	Riyadh and Hail	69 and 64	Alardhi et al. [14] 2020 and Almasri et al. [15] 2021
	Riyadh, Tabuk, Jeddah, and Dhahran	63 to 71	Alaidroos and Krarti [16] 2015 and Krarti et al. [17] 2017
	Riyadh and Tabuk	72 and 70	Krarti [18] 2020
	Dhahran	71	Ahmed and Asif [19] 2020
	Riyadh	69–79	Aldossary et al. [20] 2014
UAE	Dubai	66	AlRemeithi et al. [21] 2020
	Abu Dhabi	60	Al Amoodi and Azar [22] 2018

).

Due to high electricity consumption and the commitment to protecting the environment, for it to be more sustainable in KSA, the government increased prices at the beginning of 2016 and then again at the beginning of 2018. In 2019, Saudi Arabia’s electricity authorities reported that the electricity sold in the Kingdom was 288,598 GWh, which was distributed as 44.5%, 16.0%, 14.1%, 19.6%, and 5.8% for residential, commercial, government, industrial, and other sectors, respectively. In 2018, the number of households that used cool-air conditioning in the KSA was 594,310,179 m², and that of those that used heaters was 232,799,890 m². More than 99% of homes in the country are supplied with electricity from the public network [2].

1.1. Ways to Reduce Electrical Consumption for AC in Residential Buildings

There are three ways to reduce the consumption of electrical energy used for AC in residential buildings:

❖

By improving the thermal insulation of building envelopes, which can be more than 30% in volume [4,6,11,14,15];

❖

By improving the efficiency of ACs, which can be between 28 and 38%; (see

Table 3. AC electricity saving and combination of all energy-efficient measures in GCC (%).

Location	Efficient AC	Combination of All Efficient Measures	Reference and Date
Bahrain	37	71	Krarti and Dubey [7] 2018
Kuwait	31.0	-	Krarti [23] 2015
Oman	38	-	Krarti and Dubey [24] 2017
Qatar	28	60	Krarti et al. [25] 2017
Middle region of KSA	31.7–33.2	61.5–62.8	Almushaikah and Almasri [11] 2020
KSA, Dhahran	34.0	58.0	Ahmed and Asif [19] 2020
Dubai, UAE	37.7	70.0	Krarti and Dubey [26] 2018

); finally,

❖

By reducing the temperature difference between the ambient and the indoor environments, which can be achieved by putting the outside coil of the AC in a more suitable environment than that of the ambient air, which is the aim of this study.

As long as the indoor temperature is determined—as it usually is—based on the thermal comfort conditions indicated in the codes, it is possible to reduce the temperature difference between the internal and external environments by reducing the temperature of the environment surrounding the external coil of the AC in summer (cooling) or raising it in winter (heating). This can be done by placing the external coil in the ground or in an environment in which temperature can be stable throughout the year, as in the case of the domestic ground water tank (GWT).

Thermal comfort can be reached, according to Saudi Building Conservation Code (SBC-602), when the room temperature is in the range of 21.1–23.9 °C. As for ambient temperature, it varies in most of the country, ranging between 40 and 50 °C in summer [27]. It is too high, and that causes high temperature difference (ΔT) as well as high electrical energy consumption for AC owing to the process of creating thermal comfort conditions in the buildings. Refrigerators or AC systems are cyclic devices. Here, Q_L is the heat removed from cold space at temperature T_L, Q_H is the heat projected into the warm environment at temperature T_H, and W is the work input to the cyclic device.

The performance of refrigerators is expressed in terms of the coefficient of performance (COP). The COP in general and of the Carnot refrigerators equation is calculated as follows [28]:

$${\mathrm{COP}}_{\mathrm{R}}=\frac{\mathrm{desired}\mathrm{output}}{\mathrm{required}\mathrm{input}}=\frac{\mathrm{cooling}\mathrm{effect}}{\mathrm{work}\mathrm{input}}=\frac{{\mathrm{Q}}_{\mathrm{L}}}{\mathrm{W}}=\frac{1}{\frac{{\mathrm{T}}_{\mathrm{H}}}{{\mathrm{T}}_{\mathrm{L}}}-1}$$

(1)

Equation (1) shows that COP_R depends on the temperature difference between the cold space and the place of heat rejection. By increasing the temperature difference, the COP_R will be decreased, and thus, the consumption of electricity increases. Hence, temperature difference can be reduced to a low-level ΔT, which, in turn, will lead to lower consumption of electricity and help to make this application economic, environmentally friendly, and more sustainable. Figure 1 shows the difference between the new technique for the influence of the domestic GWT heat exchanger (a) and the typical technique for an air-source AC (b). The figure also shows the method of connecting (cooling) the condenser to the domestic GWT (heat exchanger in a water tank).

1.2. Literature Review

The ground heat exchanger (GHE) has two systems: a closed-loop system and an open-loop one. The open-loop system uses well or surface water to cool or heat the external coil of the ground heat pumps system (GHPs) by circulating the water directly through this system. When the circulation is complete, the water returns to the ground or is used again. Most closed-loop geothermal heat pumps circulate a working fluid (water) through a closed loop that is buried in the ground or submerged in water. As for the heat exchanger, it transfers heat between the refrigerant in the AC cycle and the working fluid in the closed loop. The coil can take a horizontal, vertical, or pond/lake configuration. Vertical ground heat exchangers (VGHEs) are generally used when the ground-surface area is limited. Horizontal ground heat exchangers (HGHEs) are most often used for small systems. Pipes are buried in trenches, typically at a depth of 1.2–3.0 m. Depending on the design, as many as six pipes may be installed in each trench, with trenches typically 3.7–4.6 m apart. The “slinky” is a spiral loop comprised of overlapping coils of the pipe.

Naili et al. [29] evaluated the use of GHE for cooling and the performance of HGSHPs in the north of Tunisia. An HGSHPs with a 25 m underground pipe was used. They confirmed that with a flow mass of 0.12 kg/s and 650 W power, the overall heat transfer coefficient was about 58 W/m² K, which covers a 38.5% load of the tested room (3 × 4 × 3 m). In their turn, Esen et al. [30] evaluated the HGSHPs performance with two GHE depths: 1 m and 2 m—the study was carried out in Elazıg, Turkey. They reported that the COP increased with depth, from 2.5 for 1 m to 2.8 for 2 m. Another study using nano-fluid Al₂O₃ with ethylene glycol water in the HGSHPs was carried out by Kapıcıoğlu and Esen [31] in the cold-climate region of Turkey. This study examined two types of U-type and spiral coils by using volume concentrations of 0.1% and 0.2%. The conclusion was that using a low concentration of nano-fluid in HGSHPs improved COP in general by approximately 3%. The researchers also reported that the average specific capacity for all cases was 22 W/m.

Serageldin et al. [32] studied the effect of groundwater flow rate for different shapes and configurations of the GHE. They reported that when this flow was 1000 m/year, the thermal resistance of the borehole for all cases decreased by 4 to 15.5% compared with the cases without flow. When, however, a single U-tube with an oval cross-section and a single, circular, and also cross-sectioned spacer were used, the resistance decreased by 33% in comparison with the case where a single U-tube with a circular cross-section was used. As for the experiments conducted by Soni et al. [33], these experiments were aimed at comparing the use of a GHE that is directly connected to the AC system with a conventional one for cooling a room space (1.5 TR) in Bhopal, in central India. In this study, the refrigerant flowed directly through the condenser coil, which was composed of copper tubes laid horizontally underground at a depth of 3 m, and vertical configuration of being installed on the side of the pit from a depth of 3 m from the ground surface level. The HGSHP and the vertical ground source heat pump system (VGSHP) consisted of a total of 38 m of copper tubes, which represents more than 1.5 times the normal copper tube length, but the amount of refrigerant was fixed at 2 kg. They reported that in comparison with conventional cases, the electrical energy used for the HGSHP and VGSHP systems was reduced by 11.25–14.68% in the rainy season and 10.5–13.8% in summer. Al-Douri et al. [34] explored the potential of using geothermal energy and suggested applications in this regard, focusing on the need for a sustainable, green, and clean environment. Aprianti et al. [35] compared the use of a ground source heat pump (GSHP) and an air source heat pump (ASHP) for the residential sector in Perth, Western Australia. The results showed that for cooling, the GSHP had an average COP of 3.1, while the COP of the ASHP ranged between 1.3 and 2.8, which was affected by ambient temperatures. The authors concluded that the GSHPs can reduce operating costs and cause a nearly 35% reduction in greenhouse gas emissions. Sedaghat et al. [36] evaluated the effect of various parameters on the performance of the GSHP system with a thermal recovery system for cooling a residential building in a hot desert climate in Bandar-Abbas, Iran. The results showed that this system can increase the average COP up to 20.2%.

A surface water pond/lake can be the lowest-cost option. A supply-line pipe can be laid underground from the building to the water source and then coiled into circles. The coils should be placed only in a water source that meets the required conditions. Domestic GWTs can be treated as a surface water pond/lake and used, in a similar fashion, to reject the heat of the AC in summer easily. Kappler et al. [37] carried out a theoretical as well as an experimental assessment of the use of buried water tanks as a thermal storage facility for AC in a residential building in the state of Rio Grande do Sul, Brazil. Simulations by the EnergyPlus software were used to calculate the load required to maintain a comfortable atmosphere in the building. They calculated the size of the water tank that was to be suitable for the purpose and managed to create a system that was useful for the natural AC in this building. Easow [38] assessed the use of a water-cooled external coil of 1.5 TR AC in a residential building. Water was drawn from the overhead tank (daily-use tank) in the building and used to cool the external coil of the AC by being passed from the upper tank to the lower domestic ground water tank. The results showed an improvement in the performance of the system, and energy saving amounted to 18% even after considering the amount of energy required for pumping water to the upper tank. Kuo and Liao [39] evaluated theoretically and experimentally the use of groundwater to release heat of AC under subtropical conditions in Taipei, Taiwan. The authors compared two types, closed and open water/groundwater circulation, to remove the heat of approximately 15-RT AC. The researchers showed that in the case of a closed type, the system works well for 4 h, after which it must stop for 20 h, while the open system operates continuously without hindrance. They concluded that both types can provide AC that works well and can reduce energy consumption by 22–28% using groundwater. Al-Nimr et al. [40] presented a new mathematical model for examining the performance of ACs in RBs with a GWT as a sink/source in different climate conditions. They reported that this system has higher values of COP than the conventional one, which ranged between 30 and 118%. Moreover, they reported that this system has a payback period that varies from 2 to 10 years, has environmental benefits, and eliminates outdoor fan noise. Cardemil et al. [41] conducted a theoretical study of a water source heat pump using outdoor swimming pools for space heating in three locations in Chile’s central region; the study, which used TRNSYS software, showed that this system yielded favorable results and proved more efficient for such a hot-climate region. The study also showed that this system is more economical in buildings bigger than 310 m². Woolley et al. [42] and Harrington and Modera [43] conducted studies of the use of swimming pools as heat sinks for ACs in various climatic conditions. This study, which was conducted in Davis, California, showed that the use of swimming pools as thermal sinks for ACs could save 35 to 40% of peak cooling load and 25 to 30% of the overall cooling energy in comparison with conventional cases. The study, in which the modeling and measurement values of swimming pool temperature matched well, also showed that this system could help reduce costs by 30–45%.

Finally, it should be noted that there are not enough studies on the use of domestic GWTs or geothermal heat exchangers as sinks for the heat released through the AC condenser in hot regions (water-cooled condensers, instead of air cooling). It is useful, therefore, to study the performance and application of this technique, especially after raising the electricity tariff in most GCC countries. The goal of this paper is, hence, to look for a way to reduce the use of electricity for AC in RBs in hot areas to be more sustainable. It aims to evaluate the utilization of the domestic grounded water tank as a sink for the heat rejected from ACs condensers in summer also the system can be used as an energy source in the winter season (water-cooled condensers, instead of air cooling). Two identical refrigerators will be used for comparison: one was conventional, and the other was connected to the domestic water tank.

2. Water Consumption

The use of domestic GWTs in the KSA was investigated as a case study.

Table 4. The percentage of water resources in the RBs in Qassim and KSA in 2018 (%) [44].

	Public Network	Lorry Water	Well
Qassim	79.39	17.23	3.38
KSA	78.04	20.25	1.71

shows the percentage of water resources in the residential sector, and

Table 5. The percentage of groundwater storage in the RBs in Qassim and KSA in 2018 (%) [44].

	Cement Tank	Tin Tank	Fiberglass Tank
Qassim	94.49	0.09	5.42
KSA	73.67	2.32	23.99

displays the percentage of the total GWT types in RBs in Qassim and the KSA as a whole in 2018. The table shows that the GWT is present in most homes, and the cement tank is used in most buildings, which means that the use of these tanks should be studied to see how they can be utilized as sinks to reduce energy consumption in the AC.

The consumption of water in a house in which seven people lived for over 5 years was determined, as shown in Figure 2, based on a monthly average calculated based on water consumption bills. The average per capita consumption was 280 L/day, which, as published data shows, is close to the normal average. Figure 3 shows daily water consumption per capita in Qassim from 2010 to 2018.

3. Methodology

Domestic GWTs can reduce/raise the temperature of the ambient surroundings of the external coil of the AC by heat transfer processes (convection and conduction), with water in the tank in summer/winter (water-cooled condensers, instead of air cooling). This has been studied experimentally. Usually, domestic GWTs are 25–40 m³ in size. The study was carried out on a 30 m³ tank.

3.1. Experimental Setup and Approach

The location of the experiment is a private home in Burydah, (26°16′59.1″ N, 43°59′08.3″ E), in the KSA. The experimental setup had two identical refrigerators: one was used as a base case, and one was connected to a domestic GWT (project). The external coil of the project was placed inside the box and immersed in water. The water was circulated through a heat exchanger in a domestic GWT to cool the water inside the box in summer. The design was made to show all the parts of the system and the way to connect the loop to the domestic GWT is shown in Figure 4 (the figure shows the parts of the system and the measuring devices installed in the refrigerator that was connected to the domestic GWTs). PVC was used with the coil because it is cost-effective (compared to other materials), easy to use and replace, safe, flexible enough to bend without breaking, and requires minimal maintenance [17].

The tests were carried out in October and November 2021 for sensible thermal loads in the two refrigerators were 200 and 300 W. Refrigerators were used instead of ACs, as the thermal cooling load could be adjusted more accurately and more easily. The refrigerators were placed next to each other in the same conditions. Measurements were taken every three hours, and energy-consumption reading averages were recorded daily. The experiments were conducted as follows:

(1)

The refrigerator condenser was inserted in a water box, insulation boards were applied to that box, and then, the condenser was connected to the refrigerator;

(2)

The box and the circuit were filled with water, the thermocouple sensors were connected, and then, the circulation pump was turned on to get all the air inside the pipes out;

(3)

A specific was set for the supplied electrical power inside each refrigerator;

(4)

The experimental setup was maintained for 10 days, and a measurement was taken every 3 h. Thereafter, measurements were gathered of the following:

❖

Ambient and domestic GWT temperature,

❖

Temperature inside each refrigerator,

❖

Temperature of the water flow at the box inlet and outlet ports of the project refrigerator,

❖

Power and energy consumed in each refrigerator, and

❖

The water flow rate in the coil in the domestic GWT was 240 L/h, with the power of the pump being 6 W.

(5)

This was the end of the first run, after which steps 3 and 4 were repeated for new values of the sensible thermal load;

(6)

A comparison was made between the results of the base case and the suggested system for two sensible thermal loads and the two refrigerators.

Now the system is ready to be tested, as shown in Figure 5.

Table 6. List of the main prototype components.

Instrument	Specifications
Circulation Pump	Small size, high efficiency, low consumption, low noise. Wide temperature resistance ranges from 0 °C to 65 °C. Pump 240 L/H lift 450 cm.
Digital Flow Meter	Maximum working pressure: 20 BAR, unit: L and gal. Accuracy: ±1%, repeatability: ±0.5%.
Digital Thermometer	Dual Channel, Accuracy: ±1.5%. K-type Thermocouple Measuring range: −50~300 °C.
Digital Wattmeter	Power (W), voltage (V), current (A), and electricity usage (kWh) meter. Power factor, cost. Operating max current: 15 A.
Heat Load	200 W and 300 W
Mini Fridge	3.2 cubic feet, 91 L, 220–240 volt.
Thermocouple Sensor	K-type thermocouple cable, 3 m, measure range −50~700 °C.

demonstrates the technical specifications of the equipment and devices that were used in the experiment.

3.2. Equations Governing the Design of the Ground Heat Exchanger

Forced convection occurs when a fluid flow is induced by an external force, such as a pump, fan, or mixer. Natural convection is caused by buoyancy forces due to density differences caused by temperature variations in the fluid. The first step of designing a heat exchanger is to choose the best material and dimensions for the pipes. Equation (2) shows the relationship of the heat transfer ratio to the area (tube length) [46]:

$$\mathrm{Q}=\frac{\mathrm{L}\times {\mathsf{\Delta }\mathrm{T}}_{\mathrm{lm}}}{{\mathrm{R}}_{\mathrm{total}}}$$

(2)

where:

• Q (W): heat transfer ratio;
• ΔT_lm (K): log mean temperature difference;
• R_total (K m/W): total thermal resistance;
• L (m): tube length.

Using Equation (3) was used to calculate the log mean temperature difference:

$$\mathsf{\Delta }{\mathrm{T}}_{\mathrm{lm}}=\frac{({\mathrm{T}}_{\mathrm{i}}-{\mathrm{T}}_{\mathrm{s}})-\left({\mathrm{T}}_{\mathrm{o}}-{\mathrm{T}}_{\mathrm{s}}\right)}{\ln \left(\frac{{\mathrm{T}}_{\mathrm{i}}-{\mathrm{T}}_{\mathrm{s}}}{{\mathrm{T}}_{\mathrm{o}}-{\mathrm{T}}_{\mathrm{s}}}\right)}$$

(3)

where:

• $\mathsf{\Delta }{\mathrm{T}}_{\mathrm{lm}}$ (K): log mean temperature difference;
• T_i (°C): water inlet temperature;
• T_o (°C): water outlet temperature;
• T_s (°C): water tank temperature.

We are dealing here with two kinds of convection: inner (forced) convection and outer (free) convection and conduction through the tube wall. Therefore, to find out the value of these (inner or outer) convection coefficients, we used Equation (4):

$$\mathrm{h}=\frac{\mathrm{Nu}\times \mathrm{k}}{{\mathrm{D}}_{\mathrm{i}/\mathrm{e}}}$$

(4)

where:

• Nu (-): Nusselt number;
• D_i/e (m): pipe internal or external diameter;
• k (W/m K): thermal conductivity of the water.

For finding inner forced convection used Equations (5) and (6), where Pr (Prandtl number = 5.20) at 32 °C and 3.77 at 47 °C

The Nusselt and Reynolds numbers for this case are determined by the following relationship [46]:

$$\mathrm{Nu}=0.027\left({\mathrm{Re}}^{0.8}\right){\mathrm{Pr}}^{0.33}{\left(\frac{\mathsf{\mu }}{{\mathsf{\mu }}_{\mathrm{s}}}\right)}^{0.14}$$

(5)

$$\mathrm{Re}=\frac{\mathrm{D}\times \mathsf{\rho }\times \mathrm{v}}{\mathsf{\mu }}$$

(6)

After finding the Nusselt number, now, the inner forced convection can be found by using Equation (4).

The outer free convection was calculated by using Equation (7) [46].

$$\mathrm{Nu}=0.85\times {\mathrm{Ra}}^{0.188}$$

(7)

Rayleigh number (Ra) and Grashof number (Gr) can be calculated by using Equation (8) and Equation (9), respectively:

$$\mathrm{Ra}=\mathrm{Gr}\times \mathrm{Pr}$$

(8)

$$\mathrm{Gr}=\frac{{\mathrm{D}}^3\times {\mathsf{\rho }}^2\times \mathrm{g}\times \mathsf{\Delta }\mathrm{T}\times \mathsf{\beta }}{{\mathsf{\mu }}^2}$$

(9)

where:

• µ (Pa s): fluid dynamic viscosity;
• ρ (kg/m³): water density;
• v (m/s): fluid velocity in the pipe has been considered 0.4 m/s using volume flow rate and area;
• β (1/K): thermal expansion coefficient.

After determining the internal and external convection coefficients and the thermal conductivity of the pipe and its dimensions, relationship (10) can be used to determine the total resistance of the heat exchanger [46].

$${\mathrm{R}}_{\mathrm{Total}}=\frac{1}{2\mathsf{\pi }\times {\mathrm{r}}_{\mathrm{i}}\times {\mathrm{h}}_{\mathrm{i}}}+\frac{1}{2\mathsf{\pi }\times {\mathrm{r}}_{\mathrm{o}}\times {\mathrm{h}}_{\mathrm{o}}}+\frac{\ln \left(\frac{{\mathrm{r}}_{\mathrm{o}}}{{\mathrm{r}}_{\mathrm{i}}}\right)}{2\mathsf{\pi }\times \mathrm{k}}$$

(10)

where:

• h_i (W/(m² K)): inner forced convection coefficient;
• k (W/(m K)): thermal conductivity for PVC has been considered 0.17;
• h_o (W/(m² K)): outer free convection coefficient;
• r_i (m): inner diameter, 26.2 mm;
• r_o (m): outer diameter, 33.4 mm.

As for the amount of energy transferred to the water tank, it can be calculated by using Equation (11):

$$\mathrm{E}=\mathrm{P}\times \mathrm{t}$$

(11)

where:

• E (Wh): energy;
• P (W): power;
• t (h): running time.

For calculating the temperature difference resulting from the release of the thermal energy of the condenser in the GWT, we used Equation (12) [46]:

$$\mathsf{\Delta }\mathrm{T}=\frac{\mathrm{Q}}{\mathrm{m}\times {\mathrm{C}}_{\mathrm{p}}}$$

(12)

where:

• m (kg): mass of water;
• C_p (kJ/(kg K)): specific heat of water;
• Q (kJ): heat transfer.

3.3. Design of Ground Heat Exchanger

Equation (2) demonstrates the relation of the heat transfer ratio to the area (length of tube), and Equation (4) was used to calculate the inner and outer convection coefficients. The Reynolds and Nusselt numbers for internal forced convection are found by using Equations (5) and (6):

$$\mathrm{Re}=\frac{0.0262\times 990\times 0.4}{577\times {10}^{-6}}=17,981$$

$$\mathrm{Nu}=0.027\left(17,{891}^{0.8}\right) {3.77}^{0.33}{\left(\frac{0.000577 }{0.000695}\right)}^{0.14}=103.4$$

The inner forced convection coefficient was found by using Equation (4):

$${\mathrm{h}}_{\mathrm{i}}=\frac{103.4\times 0.64}{0.0262}=2526\left(\frac{\mathrm{W}}{{\mathrm{m}}^2\mathrm{K}}\right)$$

The outer free convection coefficient was calculated using Equations (7)–(9) as follows:

$$\mathrm{Nu}=0.85\times {8115}^{0.188}=4.61$$

Using Equation (4) the outer free convection coefficient is

$${\mathrm{h}}_{\mathrm{o}}=\frac{4.61\times 0.62}{0.0334}=86\left(\frac{\mathrm{W}}{{\mathrm{m}}^2\mathrm{K}}\right)$$

Calculated by using Equation (10), the total resistance of PVC will be 0.275 (K m²/W). After solving the equations and having reasonable values for free and forced convection coefficients, the surface area of the heat exchanger can be found by using Equation (2). Furthermore, because PVC pipes are generally available and widely used, the diameter was chosen to be 25 mm for the project. If the rate of the heat released is Q = 420 W and this pipe diameter, then the tube length needed for PVC is 6 m and is used for the heat exchangers.

The design of the heat exchanger for the PVC that was tested experimentally is L-shaped. To subtract the heat of the refrigerator condenser, a box was made, and the condenser was placed in it. The box had dimensions of 0.7 m in length, 0.7 m in width, and 0.3 m in height, with a capacity of 147 L of water (used as working fluid). The box was insulated by 5 cm on all sides to reduce the effect of heat exchange with the surroundings. The box was provided with two openings: the first receives cold water from the coil of the heat exchanger located in the water tank (No. 1, at a height of 10 cm). Hot water comes out from the box (No. 2, at a height of 27 cm—to benefit from the activation of the free heat convection) and goes to the heat exchanger in the GWT. The size of the base of the box should be close to that of the area occupied by the refrigerator condenser.

4. Results, Analysis, and Discussions

4.1. Water Tank Temperatures

The average ambient temperature for a year in 2021 in Qassim is shown in

Table 7. Monthly average ambient temperature and domestic GWT temperature in Qassim.

Months	1	2	3	4	5	6	7	8	9	10	11	12
Ambient temperature (°C)	19.0	22.0	27.0	33.0	38.0	40.0	41.0	42.0	39.0	36.0	26.0	23.0
Water tank Temperature (°C)	27.5	27.0	27.8	28.7	29.8	30.4	30.9	31.5	30.6	30.1	29.6	28.3

[47]. The table also contains the measured temperature values of the domestic GWT. Figure 6 also shows these values for Qassim in 2021. The domestic GWT temperature was measured once a week throughout the year, and then, measurements were averaged over the whole period. The temperature sensor was installed at a depth of two meters from the surface of the earth. It was noticed that both the monthly average temperature of the domestic GWT and the air temperature changed from 27 to 31.5 °C and from 19 to 42 °C, respectively. Notably, the temperature of the water tank is lower than the ambient temperature in summer and higher in winter. It should also be noted here that the heat transfer from the working medium to the water is more efficient than through the soil. This makes this system sustainable and more useful, as it helps save energy in both summer and winter.

The heat transfer in the coil in domestic GWTs has three processes:

❖

Forced convection, which occurs in the inner surfaces of the coil (ground heat exchanger);

❖

Conduction from the inner surfaces to the outer surfaces of the coil;

❖

Free, or natural, convection transfers the heat located on the domestic GWT.

Domestic GWTs are renewed daily due to consumption and help to keep the temperature lower than the ambient temperature in summer—assuming the average water tank temperature is 27 °C over a year. The water tanks used in KSA homes range in size between 25 m³ and 40 m³. Using Equation (3), the log means temperature difference calculated as

$$\mathsf{\Delta }{\mathrm{T}}_{\mathrm{lm}}=\frac{\left(65-27\right)-\left(35-27\right)}{\ln \left(\frac{65-27}{35-27}\right)}=19.25°\mathrm{C}$$

4.2. Sensible Thermal Load of 200 W

Measurement with a thermal load of 200 W started on 21 October 2021. The resulting temperature curve for the water tank, the inlet, and the outlet of the box temperature of refrigerators as well as the temperatures inside both refrigerator cases are shown in Figure 7. The average power of the refrigerators used for the project was 82.63 W. As the base case, it was 115.25 W for a sensible thermal load of 200 W. The power of the project and the base case vs. time is shown in Figure 8. The average power and energy consumption and COP for the two cases are shown in

Table 8. Average power, energy, COP, and change percentage for a sensible thermal load of 200 W.

		Project	Base Case	Change Percentage %
Average Power (W)		82.6	115.3	−28.4
COP (–)	Experimental	2.42	1.74	+39.1
	Theoretical (Ideal)	14.93	11.95	+24.9
Energy Consumption (kWh)		1.98	2.77	−28.5

, in which values agree well with previous studies [30]. The obtained COP was approximately 15% of the Carnot (ideal) COP. In addition, experimental results have shown that COP improved by 39.1% compared to the base case. Here, it should be noted that the temperature inside the refrigerators was almost constant and close in both cases. Moreover, there was a rise in the temperature of the outlet of the box compared to the inlet (which is the inlet and outlet of the heat exchanger installed in the GWT, assuming that the pipes are isolated and that there is no heat exchange with the surroundings) by approximately 6 K. The important result was that energy and power consumption was reduced by approximately 28% compared to the base case, which agrees well with the results of previous studies [39,42,43]. The results are also partly in agreement with the results given in [40]—the reason behind the agreement being partial may be that the results were obtained during a year in which the difference between the application temperature (inside the refrigerator) and the ambient air was relatively small, for the percentage is certainly higher in summer when the temperature difference is high.

4.3. Sensible Thermal Load of 300 W

The experiment with a sensible thermal heat load of 300 W started on 1 November 2021 and lasted for 10 days. The resulting curves of the temperatures for the water tank, the inlet, and the outlet of the box in both refrigerator cases are shown in Figure 9. The power of the project and the base case is shown in Figure 10. The average power for the project was 86.38 W, and for the base case, it was, for the given load value, 120.13 W. The final results for the average power, COP, and energy consumption are shown in

Table 9. Average power, energy, COP, and change percentage for a sensible thermal load of 300 W.

		Project	Base Case	Change Percentage %
Average Power (W)		86.38	120.13	−28.1
COP (–)	Experimental	3.47	2.50	+38.8
	Theoretical (Ideal)	37.14	25.40	+46.2
Energy Consumption (kWh)		2.073	2.883	−28.1

. It is noted that the average temperature inside the refrigerators was almost constant between 21 and 22.7 °C. As for the temperature inside the refrigerators, it was within the human comfort temperature conditions. This indicates that the refrigerator compressors were operating continuously and at a sensible thermal load. There was a rise in the temperature of the outlet of the box compared to the inlet. This shows that the average power and energy used by a sensible thermal load of 300 W in this system was approximately 72% compared to the base case. The COPs of project and base case at a sensible thermal load of 300 W, without considering the power of the pump, were 3.47 and 2.50, respectively, which means an improvement of 38.8%. The obtained COP was approximately 10% of the Carnot (ideal) COP. During the testing phase, it was noticed that the project consumed less energy than the base case fridge, which agrees well with the results of previous studies by Woolley et al. [42], Harrington and Modera [43], and Aprianti et al. [35].

Table 10. COP (–) of the project for two loads with and without the power of the pump.

	200 W	300 W
With pump	2.27	3.25
Without pump	2.42	3.47

shows the COP of both cases, the case where pump power is and that where it is not considered. It was also noticed that taking into account the required pump power has a limited effect on the efficiency of the system.

4.4. Using the System for AC

An electrical resistance load of 300 W was placed inside the refrigerator as a thermal sensible load. Then, the power of the refrigerator was measured in the basic case. It was approximately 120 W. Assuming that the tank volume is 30 m³, the operating time of the refrigerator is 12 h. The heat released to the domestic GWT through the external unit per day and the temperature change were determined by using Equations (13) and (14):

E = (120 + 300) × 12 = 5040 Wh = 18,144 kJ

(13)

ΔT = 18,144/(4.18 × 30,000) = 0.145 K

(14)

This result shows that the temperature change in the water tank can be neglected. Figure 11 shows the increase in the temperature of the domestic GWT as a function of the released thermal energy. To ascertain the possibility of using these tanks to reject heat, it was assumed that three ACs with a total electric input capacity of 6 kW and 6 h operating hours per day (or any capacity in this range) were to be used. The thermal energy released to the domestic GWT unit per day and the temperature change was determined by using Equations (13) and (14). The thermal energy released into the tank was approximately 126 kWh when the COP value was 2.5. As a result, the tank temperature rose by about 3.6 K per day, without taking into account the loss of the heat of the tank to the surrounding soil. As long as there is heat loss in the soil and daily water consumption, this system can be relied on to reduce air conditioning energy consumption. The water temperature will increase as the amount of water is less; it will be 54.3 K for a volume of water of 2 m³. This temperature rise is high, but it can be exceeded by increasing the volume of the water tank, reducing the number of ACs connected to the water tank or choosing ACs with better efficiency. A reduction of energy consumption is also can be achieved by connecting the ACs to two condensers: the first releases heat into the water tank, and the second releases heat into the ambient air so that the heat is released into the tank during the day to reduce the peak load, and at night, the heat is released into the air. The heat transfer from the water tank to the surrounding ground is ignored in the analysis because the temperature difference is limited between the water tank and the soil (the soil temperature was measured at a depth of 1.5 m from the surface in the area and is in the range between 27 and 30 °C). However, if the temperature of the water tank is increased, greater heat loss from the tank to the surrounding soil will occur. It is preferable to determine the relationship between water consumed and the capacity of ACs connected to the groundwater tank. Here, there are concerns about the overheating of the groundwater tank. This solution can be used as an alternative to the traditional AC in appropriate cases: if the tank temperature is higher than the average ambient temperature during the summer period, it is returned to the traditional conduction of heat dissipation through the air condenser. It is also suitable in cases where water consumption is high, such as some industrial or commercial processes.

5. Conclusions and Recommendations

The main issue this paper discusses is the reduction of energy consumption by AC in RBs in hot areas, such as Saudi Arabia, and one of the motives behind this study is the problem of the large consumption of electric energy in hot areas—almost 70% of the total electricity consumption is consumed in such areas. Most of the houses in these areas use ground cement tanks for the storage of domestic water. The possibility of using this tank as a heat sink for AC condensers (water-cooled condensers, instead of air cooling) has been examined in this study. The volume of underground tanks often ranges between 25 and 35 m³, and the average house consumes about nearly 1500 L/day, which can help to keep the water in the tank not being hot. The calculation was done by using heat transfer equations, as shown above, to find out the size of the area (tube length) of the heat exchanger that is needed to decrease the temperature of the condenser utilizing this tank.

A comparison was made between two refrigerators of the same size and type: one was normal (base case); the other was a refrigerator connected through a coil located in the ground domestic water tank to cool the condenser. The results of the study can be summarized as follows:

• The important result was that the saving in energy and power was approximately 28% compared to the base case;
• The coefficient of performance of AC was improved at an average of approximately 39% in comparison with the traditional system;
• Considering the pump power required has a limited effect on the efficiency of the new system—approximately 6%;
• The peak load of the power system can be reduced, which is of considerable significance given that the electrical grid loads in summer are about twice as much as those found in winter in winter in the KSA;
• The suggested system has some substantial environmental benefits and can help create a more sustainable residential sector.

An experimental comparison was made in the Qassim region in the KSA, but the results apply only to Gulf Cooperation Council countries and similar regions in terms of climate and habits.

The recommendations for further studies are:

• Changing the mass flow rate of the system in testing can affect the efficiency of the system;
• Conducting studies on larger AC capacities to determining the relationship between the volume of the tank, the heat loss from the tank, the water consumed, and the capacity of ACs that can be connected to the groundwater tank;
• Examining the possibility of using different materials for the pipes of the coils;
• Investigating the possibility of placing the external coil of the AC directly in the domestic GWT.