Issue 
Renew. Energy Environ. Sustain.
Volume 6, 2021



Article Number  1  
Number of page(s)  19  
DOI  https://doi.org/10.1051/rees/2020008  
Published online  19 January 2021 
Research Article
Performance analysis of horizontal ground source heat pump for building cooling in arid Saharan climate: thermaleconomic modeling and optimization on TRNSYS
^{1}
Laboratory of New and Renewable Energies in Arid and Saharan Areas–LENREZA, University of Kasdi MerbahOuargla Po Box 511 30000, Ouargla. Algeria
^{2}
Laboratory of Thermal Processes, Research and Technology Center of Energy, Hammam Lif, B.P. 95, 2050 Tunis, Tunisia (CRTEn)
^{3}
Department of Electrical Engineering, Faculty of Science and Technology, University Ziane Achour of Djelfa, Djelfa 17000, Algeria
^{*} email: sabrinskemr@gmail.com
Received:
8
November
2020
Received in final form:
8
December
2020
Accepted:
8
December
2020
This paper presents a feasibility and performance study of ground source heat pump (GSHP) coupled with horizontal ground heat exchanger (HGHX) used for cooling residential unit equipped with radiant floor system (RFs) under the meteorological conditions of Saharan environment in Ouargla, city located in Southeast Algeria. A dynamic simulation system is developed using TRNSYS software for modeling the performance of the GSHP system. To verify the reliability of GSHP including HGHX system programs, the modeling procedure was validated against experimental data from a horizontal ground source heat pump system (HGSHPs) installed at the Research and Technology Center of Energy (CRTEn), Tunisia, and a good agreement was obtained. Then, to obtain an acceptable balance between system efficiency and total cost of HGSHPs an economic analysis was carried out to determine the optimum design parameters of the HGHX. The simulation results obtained from this study indicated that the HGSHPs could effectively solve cooling problem and reduce traditional energy consumption in the Saharan areas; it is possible to lower the mean indoor air temperature below 27 °C and raise the average relative humidity to reach 73.97%. By concentrating principally on the thermaleconomic optimization, the optimized COP of the GSHP that combines the reliability and economy of cooling in long term was found to be 3.89.
© S. Korichi et al., published by EDP Sciences, 2021
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
1 Introduction
The energy consumption in urban buildings has gained an increasing demand recently due to the high comfort standards [1]. This energy reaches about 40% of the world's energy use, for example, in Algeria, the residential sector is the largest consumer of energy, accounting about 43% of the total national energy consumption [2]. This value will increase to reach 179.78 TWh in 2040 with the increasing of population and because of urbanization and improvement of the living standard [2]. Indeed, these values indicate that the residential sector represents the largest source of greenhouse gas emissions (GHG) and ecological problems in the world.
In recent years, global warming has become a major threat to the environment, especially in arid and semiarid areas where the desert regions in southern Algeria recorded a significant increase in temperatures that exceeded the normal rate, which constituted a burden on conventional air conditioners to provide the cooling needs for the buildings during the summer. Therefore, it is necessary to highlight the use of new technologies based on renewable energies, which still represents only 0.1% of Algerian energy consumption [3]. The renewable energy that appears to be very well adapted in the building sector is the geothermal energy; it represents the solution for reducing the large energy consumption and environmental problems of the building sector.
The shallow geothermal energy is a clean, sustainable and low carbon solution for the heating and cooling of buildings by pumping heat from or to the shallow subsurface to provide the thermal loads. Compared with other renewable sources, the shallow geothermal energy has important advantage, it can be used everywhere in the world especially in residential sector areas where the quantity of exploitable geothermal energy stored in the shallow subsurface is greater than its annual requirements for cooling and heating.
In the last decades, shallow geothermal systems such as ground source heat pump systems have been increasingly applied to heating and cooling residential buildings and infrastructure. Among the different types of GSHPs, the closed ground source heat pump system (GCHP) is one of the best configurations in terms of practicability and independent installation. It consists of heat exchanger loops that utilize the heat conduction mechanism of the ground to reject or extract heat.
The ground loop pipes or the ground heat exchangers, which are placed horizontally or standing vertically in ground, have the largest portion of investment cost in ground source heat pumps installation [4]. Compared with vertical GHXs, the horizontal GHXs require longer pipe length and ground surface area but they are typically less expensive due to the lower cost of drilling and refilling. However, the investment price remains relatively high compared to the air conditioning system, Therefore, the process of improvement of the system must be carried out during the sizing of GHX in order to obtain the most costeffective design parameters and thus avoiding the significant financial losses [4–6].
Over the last two decades, several studies have shown the importance of conducting a predesign analysis to determine the influence of well properties on ground source heat pump performance and installation costs, modeling a software program based on a model of ground heat exchanger and evaluating the optimum technical characteristics of the GSHP. Sanaye et al. [7] presented a modeling and optimizing processes for a GSHP coupled with HGHX by defining the total annual cost (total of investment and operation costs) as an objective function using Nelder–Mead and Genetic algorithm techniques to estimate the optimum design parameters of the system as well as the configuration of HGHX. Their study indicated that the system parameters (saturated temperature condenser and evaporator, inlet and outlet temperatures of the water source and GHX nominal pipe diameter) have a significant effect on ground coupled heat pump performance and the total annual cost of the system. Therefore, it is necessary to preinvestigate the exploitation of closed geothermal systems before any field application in given climatic conditions in order to obtain optimum system parameters that ensure high efficiency and lower cost. Kayaci et al. [8] studied the energy and economic analysis of HGHX used for cooling and heating an office with a total conditioned space of about 200 m^{2} in Turkey. The effect of all geometric properties of the designed horizontal GHE as well as the effects of the rate of increase in electricity prices was investigated by defining a reference function (total cost) as an optimization parameter. In their simulation study, found that the number of parallel pipe, burial depth, pipe spacing, tube diameter and pipe length have a major impact on the performance of the GSHP and the reference function because it affects the excavation and labor costs and the amounts of heat extracted from soil or transferred to soil, simultaneously. Thus, they concluded that it is feasible to examine all parts of GSHP system in the predesign analysis to obtain the optimal design parameters that ensure a balance between efficiency and costs. Suggested values for all of these engineering properties are recommended.
Inalli et al. [9] the effects of various system parameters such as the buried depth of earth coupled heat exchanger, the mass flow rate of the water antifreeze solution and the sewer water on the performance of ground source heat pumps were examined based on experimental analysis conducted in Turkey. It was found that the poor design of the GSHP system caused lower performance factors when compared their results to other heat pumps operating under the same design values. They concluded that it is necessary to conduct a predesign analysis to determine optimal system parameters that will ensure minimum energy consumption and favorable costs with the help of GSHP simulation models. Reliable numerical models and methods were shown in various studies [10–12].
The properties of the soil, actual loads of buildings and meteorological factors play a significant role in the design of the geothermal heat exchangers. In recent years, many studies dealing with the influence of soil thermophysical characteristics on ground source heat pump performance have been published [13,14]. Most researchers have found that the undisturbed soil temperature and ground resistance, which vary with time and space strongly affects the amount of geothermal energy that can be transferred to or extracted from the ground. This leads to significant differences in GSHP thermal performance from one region to another which changes the design criteria for the ground heat exchangers, especially for the horizontal ones, since they are the closest to the ground surface and the most affected by external factors. Geological and climate conditions can also strongly influence the GSHP operation and the total cost of the GHX system.
In this context, the main aim of this study was to preinvestigate the exploitation of horizontal ground source heat pump systems for cooling in arid Saharan environment. A TRNSYS model of HGSHPs equipped with RFs is proposed to provide the cooling for residential building of about 95 m^{2} area located in Ouargla, SouthEast of Algeria. The model was validated using experimental data collected from a HGSHPs conducted in the Research and Technology Center of Energy at Borj Cédria, northern Tunisia. For optimal system design, a thermaleconomic optimization was carried out to estimate the optimum HGHX design parameters for the case study region. The effect of the pipe length, burial depth, pipe diameter and mass flow rate on the thermal performance and the system total cost were investigated. Then to ensure the reliability of the optimization results, the simulation inlet and outlet temperatures of the HGHX were calibrated using ASHRAE guideline constraints.
2 The test region
As a future working project, the optimized HGSHPs is mainly designed to be installed and tested in hotdry and desert environments. It has been proposed to meet the thermal comfort requirements of a house model under climatic conditions of Ouargla city, Algeria. The city of Ouargla is situated in northeastern of the Algerian Sahara, it is located at 164 m altitude, 31°57′ N latitude and 5°21′ E longitude and covers a total area of about 163.233 km^{2}. It has a dry desert climate, which is characterised by mild winters and very hotdry summers. The average, minimum and maximum monthly variation of temperatures, including the average monthly variation of the wind speed and relative humidity measured with the meteorological weather station of Ouargla during the year 2019 were displayed in Table 1.
The study region of this work is one of the main geothermal areas in Algeria, as illustrated in Figure 1. It is characterized by a mean gradient of 21 °C/km, which corresponds very high geothermal heat flow values (ranging between 80 and 100 mW/m^{2}) [15].
The hydraulic drilling that crosses the Saharan territory indicates that this area is mainly covered with Tertiary and Quaternary materials [16], which are characterized by a sandy limestone majority in the surface layers of lithological classifications.
The min max average monthly variation of outdoor temperatures and the average monthly variation of relative humidity and wind speed in the year 2019.
3 Basics of the design
3.1 Thermal modeling analysis
Using the first law of thermodynamics, the heat absorbed from the building Q _{ b } and the heat transferred to the ground Q _{ g } can be computed from the following equations:$${Q}_{b}={{\displaystyle \stackrel{}{m}}}_{wRF}{C}_{p{H}_{2}O}\left({T}_{inRF}{T}_{outRF}\right)$$(1) $${Q}_{g}={{\displaystyle \stackrel{}{m}}}_{wGHX}{C}_{p{H}_{2}O}\left({T}_{inGHX}{T}_{outGHX}\right)$$(2)
The heat transfer rate from tank storage to refrigerant via the evaporator is:$${Q}_{Evp}={{\displaystyle \stackrel{}{m}}}_{wEvp}{C}_{p{H}_{2}O}\left({T}_{inEvp}{T}_{outEvp}\right)$$(3)
The heat transfer rate from condenser to ground loop is:$${Q}_{Con}={{\displaystyle \stackrel{}{m}}}_{wCon}{C}_{p{H}_{2}O}\left({T}_{outCon}{T}_{inCon}\right)$$(4)
The compressor power is expressible as:$${{\displaystyle \dot{W}}}_{Com}={Q}_{Con}{Q}_{Evp}$$(5)
The heat pump coefficient of performance (COP) estimated by the ratio between the heat absorbed from building and the compressor power input:$$COP=\frac{{Q}_{Evp}}{{{\displaystyle \dot{W}}}_{\mathrm{C}\mathrm{o}\mathrm{m}\mathrm{p}}}$$(6)
The seasonal energy efficiency ratio (SEER) can be calculated from:$$SEER=\frac{\left(1.12\sqrt{\left(1.25440.08EER\right)}\right)}{0.04}$$(7)
Where EER is the energy efficiency ratio.
3.2 Economic performance analysis
Because of the high investment cost of the ground heat exchanger installation and the large cost of longterm electricity consumption, the objective of this part was to minimize the initial and operation costs of HGSHPs for the building cooling capacity in which the thermal performance of the GSHPs was considered (i.e. COP). Thus, an economic analysis was performed by defining an objective function to obtain the optimized values of GHX parameters design.
3.2.1 Objective function
An objective function, OF ($/kW), was defined by dividing the total cost into the total annual cooling loads:$$OF=\frac{TAC}{{Q}_{c}\times {T}_{c}}$$(8)where Q _{ c } and T _{ c } are the cooling load and annual cooling operation hours per year respectively, the TAC is the total annual cost which contains the investment and the electricity costs, was defined as:$$TAC=CRF({C}_{inv}+{C}_{elec})$$(9)Then
where CRF, is the capital recovery factor, expressed as:$$CRF=\frac{i}{1{\left(i+1\right)}^{n}}$$(10) i and n are interest rate and number of operation years, respectively.
Thus, the objective function can be reformulated by following equation:$$OF=O{F}_{inv}+O{F}_{elec}$$(11)Where:$$O{F}_{inv}=\frac{CRF\times {C}_{inv}}{{Q}_{c}\times {T}_{c}}$$(12) $$O{F}_{elec}=\frac{CRF\times {C}_{elec}}{{Q}_{c}\times {T}_{c}}$$(13)
3.2.2 The annual cost of the power consumption
The annual electricity consumption of GSHP systems includes the annual energy consumption of compressor and circulating pump:$$E={E}_{\mathrm{p}\mathrm{u}\mathrm{m}\mathrm{p}}+{E}_{\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}}$$(14)
The annual electricity consumption of the compressor, E _{comp } (kWh), can be determined as:$${E}_{\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}}={T}_{c}{W}_{\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}}$$(15)
The annual electricity consumption of the circulating pump, E _{pump } (kWh), can be determined as:$${E}_{\mathrm{p}\mathrm{u}\mathrm{m}\mathrm{p}}={\tau}_{c}\frac{{{\displaystyle \stackrel{}{m}}}_{w}{H}_{\mathrm{p}\mathrm{u}\mathrm{m}\mathrm{p}}}{{\rho}_{w}{\eta}_{\mathrm{p}\mathrm{u}\mathrm{m}\mathrm{p}}}$$(16)where τ _{ c } (h) is the total running hours in the cooling period. H _{pump } is the total pumping head in the HGHX (Δp _{ GHX }) and heat pump (Δp _{ GHP }).
Thus, the total annual electricity consumption, C _{ elec } ($), is obtained as follows:$${C}_{\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{c}}=\left({E}_{\mathrm{p}\mathrm{u}\mathrm{m}\mathrm{p}}+{E}_{\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}}\right){E}_{tariff}$$(17)where E _{tariff } ($/kWh), is the price of electricity for each kWh, which can be obtained from Sonelgaz [17].
And finally, the total cost of electricity consumption for 10year period is obtained as follows:$${C}_{elec}={C}_{elec,1}+{\displaystyle \sum _{n=2}^{n=10}}{C}_{elect,n}e$$(18)where e is the increases in electricity prices.
3.2.3 The annual capital cost
For the total investment cost of GSHP system including the GHX, C _{ inv } ($) a reference function was defined as:$${C}_{inv}={C}_{\text{Pipe}}+{C}_{\text{earthwork}}+{C}_{\text{heatpump}}+{C}_{\text{pump}}$$(19)
For the cost of pipes,earthwork, pump, heat pump unit, the following cost functions were applied [18]:$${C}_{\text{Pipe}}={c}_{\text{Pipe}}\left(NPT\right){L}_{GHX}$$(20) $${C}_{\text{earthwork}}={a}_{1}\left(NPT\right){L}_{GHX}Z$$(21) $${C}_{\text{heatpump}}={a}_{2}{Q}_{\text{heatpump}}^{a3}$$(22) $${C}_{\text{pump}}={a}_{4}{W}_{\text{pump}}^{a5}$$(23)where:
c _{Pipe } is the regional cost of polyethylene pipe per meter which depending on the diameter of pipes, (Tab. 2) NPT is the number of parallel pipe and L _{ GHX } is the total required length of the HGHX. Z is the depth of the ground. a _{1} is the regional price of the drilling and refilling ($/m^{2}). a _{2}– a _{5} are constant coefficients which were obtained based on the manufacturer's data according to the given required cooling capacity and the regional price of the equipments. (Tab. 2)
4 Model development and validation
As stated in Section 1, the model in this work focuses on preinvestigating the exploitation of HGSHPs for cooling a residential unit in a dry desert climate. A simulation model of the geothermal system has been proposed consisting of watertowater heat pump, a water tank with immersed heat exchangers, horizontal ground heat exchanger, water circulation pumps and the RF system.
A novel method based on multiyear energy and economic analysis has been developed in order to achieve the two main goals of the model: reducing the total investment and operation costs as much as possible while ensuring a good system efficiency. The energy performance of GSHP was evaluated with a developed TRNSYS model. Then, the numerical outputs were used in the economic performance analysis to obtain the moste costeffective design parameters of the HGHX. Because of the massive computations in the optimization process, a MATLAB code was used to evaluate the objective function based on equation (8).
The description of the HGSHP dynamic simulation model and its validation, which was performed by comparing the simulation results with measured data obtained by experimental device will be presented in the following sections, while the schematic diagram containing the algorithm procedure is shown in Figure 2.
Fig. 2 Block diagram of the design procedure. 
4.1 TRNSYS model description
The entire model of the HGSHPs was simulated in the TRNSYS 16 simulation environment. Figure 3 shows the simulation system developed for the cooling of the building exemple. The overall model is composed of four important submodels: the building, the ground heat pump, the data acquisition (DAQ) system and the horizontal ground heat exchanger. This model includes the main components for the installation which can be provided from the TRNSYS library (standard models). It is summarized in Table 3.
The simulation time step was set at 5 min in order to capture any thermal behavior of the system during the simulation periods. The set point temperature for cooling is 25.5 °C when the compressor switch on and 26.5 °C when the compressor switch off and the set point temperature inside the water storage tank was set as 6 °C when the compressor switch on and 13 °C when the compressor switch off.
Fig. 3 The model of HGSHP system developed in TRNSYS. 
The technical information of the main system components and TRNSYS types.
4.1.1 Building thermal model description
A onestory residential house with a floor area of 95 m^{2} and an internal height of 3 m was created in Room Arranger for the simulation performance analysis in the TRNSYS environment. The model of the building with northern orientation was divided into six thermal zones corresponding to the living requirements (three bedrooms, a kitchen, a living room and one toilet).
The following suggestions are also included in the house model and load characteristics topics:

The flat roof made with 20 cm heavy concrete block, 8 cm screed and 1 cm asphalt.

The walls made with 15 cm brick, 5 cm insul and 6 cm plaster.

The ground made with 0.5 cm floor, 6 cm stone, 4 cm silence, 24 cm concrete and 6cm concrete.

The house have a eight double glazing windows with 3 m^{2}.

The windows are primarily southfacing to block direct sunlight during the summer [20].
The house model was imported into TRNSYS simulation studio, Figure 4 clearly shows the dimensions of the building thermal zones.
Fig. 4 3D schematic view of the typical building (1): Zoon1 (kitchen), (2), (3), (6): Zoon2, 3, 6 (bedrooms), 4: Zoon4 (bathroom) and (5): Zonn5 (living room). 
4.1.2 GHP description
In the proposed GSHPs under study, the parameters of the watertowater heat pump model were determined based on the product specification available from the manufacturers and based on the design load so that its nominal capacity is adequate for the thermal loads of the building. The reversible heat pump is designed to satisfy the cooling demands of a large area, so to adapt it to our case study, we used a water storage tank (insulating tank, variable inlets) of 1 m^{3} to increase the thermal inertia of the system. The storage tank is directly connected to the two inputs of the GHP unit and on the other hand linked through a small circulation pump to ensure the circulation of water in the RF system.
4.1.3 Radiant floor system
In the TRNSYS program, the TRNBuild model is equipped with a radiant floor system that provides cooling to the various zones, the circuit of the RF system consists of a multilayer (Crosslinked polyethylene) heat exchanger with 0.017 m of outside diameter and with a 0.02 m distance between the pipes. The heat exchangers are distributed throughout the ground floor to ensure a uniform internal temperature distribution.
4.1.4 Ground source heat exchanger
The proposed GSHPs uses a geothermal field as a heat sink designed according to the thermal loads of the house, this field is composed of straight ground heat exchanger installed horizontally at different depths in the ground. It consists of HDPE (highdensity polyethylene, PE 100, type SDR11) pipe with different inner diameters where the circulating fluid (water with antifreezer) flows as a heat transfer fluid.
4.1.5 Weather data
As already mentioned in the previous sections, the meteorological factors are a very important database to represent the most important physical phenomena related to the calculation of thermal loads and soil temperatures. For TRNSYS thermal modeling, the weather data was collected from weather stations and converted to TMY2 data format by an actual meteorological data processor.
4.1.6 Annual energy consumption
In this work, the dailay load profile was categorized into three time periods as illustrated in Figure 5, this categorization was mainly designed for ground recharge purposes by assuming that there is no cooling demand during the transition periods. In order to calculate the cooling thermal load, the cooling thermostat was set as 26 °C from 10 to18 and 20 to 06 for all the spaces according to the thermal comfort standards [21].
Figure 6 shows the variation of the annual building cooling loads which were obtained using the TRNSYS simulation software based on the weather data of the south climate. According to the load simulation results, the cooling period started from 15 May to 31 October, the remaining periods were considered as transition and heating periods.
Fig. 5 Daily cooling and transition periods. 
Fig. 6 Hourly coolin loads of the proposed building. 
4.1.7 Variation of the soil temperature
The design of geothermal heat exchangers depends strongly on the undisturbed soil temperature (i.e. the soil temperature before any heat injection or extraction) as they affects soil resistance [13,14], which limits the heat exchange between the soil and the GHX wall [7], and thus affects the GSHP performance coefficient as well as the design parameters of the ground heat exchanger.
The initial subsurface temperature distribution T _{ s } can be computed as a function of the time of year and the depth below the ground surface as the following correlation [14]:$${T}_{s}={\displaystyle \stackrel{}{{T}_{s}}}{A}_{s}\times \mathrm{exp}\left[Z\left(\sqrt{\frac{\pi}{365\times {\alpha}_{sol}}}\right)\right]\times cos\left[\frac{2\pi}{365}\left({t}_{now}{t}_{shift}\frac{Z}{2}\times \sqrt{\frac{365}{\pi \times {\alpha}_{sol}}}\right)\right]$$(24)
$\stackrel{}{{T}_{s}}$ is the mean value of the ground surface temperature for a location of Ouargla city over an entire year, A _{ s } is the amplitude that the surface temperature experiences throughout a year at the location, the parameter t _{ shift } is the difference in time between the beginning of the year and the occurrence of the minimum surface temperature, α _{ sol } is the thermal diffusivity of the ground $\left(\frac{{m}^{2}}{day}\right)$. The thermal properties of the soil were obtained referring to Table 4.
Figure 7 shows the undisturbed soil temperature profiles over a year for the given climate. It can be observed that the amplitude of the soil temperature signal decreases as the depth increases. In the upper 3 m, where soil temperature is affected by the external factors of climatic conditions, the soil has a big thermal slope. In deeper distances that correspond to the thermally stable zone, the soil mean temperature is almost constant.
The soil properties.
Fig. 7 Variations of soil temperature with depth (m), Ouargla city. 
4.2 Model validation
This section presents the validation of TRNSYS type against experimental measurements of one of the three heat exchangers of the CRTEn ground source heat pump projet. For this purpose, all the technical characteristics of the experimental device as well as the climate teste building have been integrated into the developed TRNSYS model. The following assumptions have been made:

The model has been validated with the consideration of the experimental measurements of mass flow, pipe length, pipe diameter, burial depth as inputs for the HGHX type.

The number of horizontal nodes has been increased in order to increase the accuracy at the expense of simulation time.

Average value of function (time interval) was imposed by the library (TESS) controller to correctly reproduces the evolution of the on/off cycle of the heat pump unit over the simulation periods.

In order to ensure good accuracy in predicting the response of the drill, the thermophysical properties of the soil and the backfill material around the horizontal buried pipes were precisely determined before being integrated in the HGHX type.
4.2.1 CRTEn ground source heat pump projet
The experimental system (Fig. 8) installed in a test office of the Laboratory of Thermal Processes (LPT) at the Research Center and Technologies of Energy (CRTEn). The system consists of a GSHP unit manufactured by CIAT that provides the air conditioning for 12 m^{2} area. The GHP with a cooling capacity of 12.7 kW is connected to an internal and external circuits. The internal circuit includes a radiant floor system, a water storage tank and a circulation pump. On the other hand, the external circuit comprises a circulation pump and three buried heat exchangers (GHX):

Heat exchanger of length 25 m buried at 1 m of depth.

Heat exchanger of length 50 m buried at 1 m of depth.

Heat exchanger of lenght 50 m buried at 0.6 m of depth.

Heat exchanger of length 100 m buried at 1 m of depth.
A detailed description of the installation, the measurement equipments and the particular conditions of its operation is provided in Refs. [22–24].
4.2.2 Experimental verification results
In order to test the reliability of the simulation model that we have developed, the validation results presented in this section will mainly aim to evaluating the ability of the model to qualitatively represent the observed physical phenomena in pilot plan experimentation.
The experimental measurements has been compared against the simulated inlet and outlet fluid temperature of the GHX, the heat exchange rate to ground and the coefficient of performance. Figure 9 reports the relativeerror analysis between experimental and simulation results, we can observe from this illustration that the maximum error between the simulation and experimental performance coefficient does not exceed 4.9%, the same for the outlet and inlet water temperature and the heat exchanger rate, the maximum relativeerror reaches about 1.4C%, 2.3C% and 0.5%, respectively. This results showing that the TRNSYS model is able to reproduce the thermal behavior of the HGSHP with a high accuracy. Therefore, it can be concluded that by means of the developed model of GSHP in TRNSYS environment, the GHP system and the drill outputs can be preinvestigated individually based on meteorological and geological databases.
Fig. 9 Experimental vs numerical coefficient of performance of the GSHP. 
5 ASHRAE guidelines constraints
In the ASHRAE design methods, the water mass flow rate and the outlet ground heat exchanger temperatures constraints applied in optimization procedures according to practical conditions for the cooling mode are [25]:

The maximum value of the T _{ out−GHX } is 11–17 °C greater than the undisturbed soil temperature (T _{ s }) to guarantee the proper heat exchange between the soil and hot water and refrigerant in condenser.

In warm climates (T _{ s } > 15 °C) the optimal level tends to the minimum for this range (11 °C).
6 Results and discussions
The TRNSYS model presented and validated above is used in this section to envisage the effectiveness of the HGSHPs for airconditioning a typical house in Saharan climate. The first part was conducted to determine the optimal GHXs design using a thermaleconomic optimization. In the second part, the optimized values of all the design parameters were performed in the HGSHP model that used for cooling a residential building with 95 m^{2} of surface in the city of Ouargla, Southeast of Algeria.
6.1 Optimization results
The optimal design of the GHX is obtained by considering the long time operating costs and the first investment costs simultaneously under technical and economic conditions. Thus, the effects of the design parameters such as the pipe length, pipe diameter, burial depth and mass flow rate on both the objective function and the performance of GSHP were investigated in ten years of cooling period in different cases as summarized in the Table 5.
The system operating conditions and the constant economical factors were collected in Table 6. The obtained optimum values of all effective parameters in objective function are listed in Table 7.
Design parameters of all cases.
The list of input parameters used in economic analysis.
The list of optimum design parameters including the results values of elements of the objective function.
6.1.1 Effect of the pipe length
In this section, the effects of pipe length on the thermal and economic performance of the HGSHPs, were investigated by assuming constant mass flow rate, pipe diameter and burial pipe depth (Cases 1–9).
The coefficient of performance of the GSHP and the objective function values are shown in Figure 10 (left) and (right), respectively. As expected, the results demonstrates that an increase in pipe length leads to an increase in the thermal performance of the heat pump and the first investment cost. In the initial stage, the electric power consumption of the compressor decreases with the increase of pipe length, which reduces the objective function until getting a minimum point at 500 m of pipe length (Case 4, optimal point). After this stage, the system investment caused by increased heat exchanger area as well as the excavation works exceeds the decrease of operating cost which explain the increase in the objective function in this range of pipe lengths.
Fig. 10 Effect of HGHX length on the objective function and the COP of the GSHP. 
6.1.2 Effect of the pipe diameter
The variations of the objective function and the performance coefficient of the GSHP system as functions of pipe diameter in constant mass flow rate, burial depth and pipe length (Cases 10–13) are shown in Figure 11. Results illustrate that the objective function and the COP of the GSHP were directly affected by pipe diameter values. The electric power consumption of the compressor and the well pump decreases with the increase of pipe diameter, indicating a degradation in the objective function of about 63.16% from 0.02 to 0.04 m of outer pipe diameter.
Fig. 11 Effect of HGHX diameter on the objective function and the COP of the GSHP. 
6.1.3 Effect of the burial depth
To determine the optimal burial depth in which we must put the GHX pipes a series of simulations were carried out. The results which plotted the objective function and the COP of the GSHP as functions of the burial pipe depth in constant mass flow rate, pipe diameter and pipe length (Cases 10–19) are shown in Figure 12. With increasing the burial depth, the electric power consumption of the compressor decreases and the COP increases as a result of the improvement in the amounts of heat transferred to soil in the lower depths. But it is clear that from 1 m of depth (Case 11, optimal point), the investment cost caused by increased the cost of earthworks is far exceeds the decrease of operating cost, which explain the increase in the objective function by more than 7.22% from 1 to 3 m of burial depth.
Fig. 12 Effect of burial depth on the objective function and the COP of the GSHP. 
6.1.4 Effect of the mass flow rate
Figure 13 displays the objective function and the coefficient of performance of the GSHP as function of mass flow rate in constant burial depth, pipe diameter and pipe length (Cases 20–27). In the initial stage (Cases 20–23), the COP increases and the electric power consumption of the well pump decreases with the increase of mass flow rate of water circulating in the GHX, which explain the decline in the objective function until getting a minimum value at 1 kg/s (optimal point). In higher mass flows rate (Cases 24–27), the pressure drop in the GHX was increased gradually with the increase of fluid velocity that increased the electric power consumption of the well pump and simultaneously reduced the COP values.
Fig. 13 Effect of mass flow rate on the objective function and the COP of the GSHP. 
6.2 The optimized HGSHPs operation analysis
In this section, investigating the effects of the all optimum design parameters on the operation of the HGSHPs were performed. The effect of the long time HGSHPs operation on the drill outputs and soil temperature were assessed as follows:
6.2.1 Drill outputs and soil temperature estimates
The inlet/outlet GHX temperatures and the average soil temperature of the optimized HGSHPs for the first year and for ten years in Ouargla were shown in Figures 14–16. The inlet and outlet GHX temperatures for ten years presented relatively parallel trend and the average soil temperature around the GHX was stably equal in all this period. Therefore, the employment of the HGSHPs was very suitable for buildings of Ouargla in the long term. As shown in Figure 14b, the maximum difference between the outlet and inlet GHX fluid temperature ranging between 22 and 22.2 °C and the maximum difference between the outlet GHX temperature and soil temperature is 14.5 °C. Due to the very hot climate, these values showing that the simulation results present an acceptable agreement with ASHRAE constraints.
The heats injected into ground during the cooling season is shown in Figure 17. Results illustrate that the heats rejected from the GHX into ground vary between 0.37 and 83.6 kW, which reflect the high potential of saharan territories for cooling buildings.
Fig. 14 Evolution of water temperatures at the external and the internal circuits in (a) the first year period, (b) typical day for L_{GHX} = 500 W/m, D_{iGHX} = 0.0363 m and Z = 1 m. 
Fig. 15 Variation of fluid inlet, outlet ground heat exchanger water temperatures in nine years for L_{GHX} = 500 W/m, D_{iGHX} = 0.0363 m and Z = 1 m. 
Fig. 16 Variation of average soil temperature surrounding the ground heat exchanger at the first, fifth and tenth year. 
Fig. 17 Evolution of the heat injected into ground and the heat rejected from building in (a) the first year period, (b) typical day for L_{GHX} = 500 W/m, D_{iGHX} = 0.0363 m and Z = 1 m. 
6.2.2 Building thermal conditions
In order to show the rentability of optimized HGSHPs for refreshing the different typical building zones, we represent in Figures 18 and 19 the variations of indoor and outdoor climat conditions during a typical days of cooling mode (July 1 until July 30) in the first year of operation for different scenarios.
In the first state, without the HGSHP conditioning system, it was observed that the indoor temperature vary between 37.44 and 41.96 °C, while the indoor relative humidity ranges between 25.48% and 33%, for an outside air temperature and relative humidity ranging between 25.92 and 48.72 °C and 10–57.96%, respectively.
In the second state, with the HGSHP conditioning system, the indoor air temperature ranges between 23.5 and 27 °C while the relative humidity values ranges between 58.31–73.97%, this corresponding to a periodic change in the inlet and outlet RF system water temperature of about 6.95–13.1 and 12.2–16.1 °C, respectively as shown in Figure 14. These results demonstrate the effectiveness of the HGSHPs for cooling in the selected climate that experience hightemperature days.
The simulation total heat absorbed from the building (Q _{ b }) is represented also in Figure 17. We can note from this figure that the absorbed heat can reach a maximum and a minimum values of 5.99 and 10.49 kW, respectively, corresponding to the compressor on/off. These considerable values is mainly due to the high cooling loads of the building example.
Fig. 18 Comparison between indoor and outdoor temperatures in a typical days (July 1 until July 30) for L_{GHX} = 500 m, D_{iGHX} = 0.0363 m and Z = 1 m. 
Fig. 19 Comparison between indoor and outdoor relative humidity in a typical days (July 1 until July 30) for L_{GHX} = 500 m, D_{iGHX} = 0.0363 m and Z = 1 m. 
6.2.3 Energy consumption distribution and COP evaluations
The COP and the operation power consumption of the optimized HGSHP system (including GSHP units, circulating pump in internal circuit and the circulating pump in external circuit) for the first year of cooling operation were shown in Figure 20a and 20b. The average COP of GSHP in cooling season was about 3.92. Based on this result, the energy efficiency ratio (EER) and seasonal energy efficiency ratio (SEER) of the HGSHPs which are especially important indicators in hotdry climates can be calculated as 13.44 and 17.42, respectively.
Fig. 20 The power analysis of the first year HGSHPs operation, (a) the power consumption distribution, (b) the COP distribution for L_{GHX} = 500 W/m, D_{iGHX} = 0.0363 m, m_{wGHX} = 1 kg/s and Z = 1 m. 
7 Conclusions
In this study, the feasibility of improving the performance of a GSHP cooling system coupled with HGHX for the weather conditions of a Saharan environment and more particularly in the southeastern desert of Algeria was evaluated. A thermaleconomic modeling and optimization of the system were analyzed using both MATLAB code and TRNSYS software which were previously validated.
The effects of various parameters such as mass flow rate of circulating water, length and burial depth of the GHX on the performance of the GSHPs were analyzed for a tenyear period in order to obtain the most costeffective design parameters of the HGHX. Furthermore, the effects of the optimum design parameters on the operation of the HGSHPs used for cooling a 95 m^{2} surface area was investigated. The following conclusions are achieved:

According to the undisturbed soil temperature before any heat rejected, the soil thermally stable zone corresponds to depths more than 3 m.

Higher pipe length more than 500 m has no significant effect on the thermal performance of the HGSHPs and is also not economical.

Increasing the burial depth more than 1 m greatly affects the first investment total cost by increasing the price of earthworks and excavation. For every 0.5 m of additional borehole depth the return on investment is deteriorated by 12%, whereas the system COP performance is improved by 4.9%.

The thermal performance coefficient decreases and the electric power consumption increases as the pipe diameters decrease. The small pipe diameters are not practical.

The electric power consumption of the well pump increases sharply when the water velocity in the GHX pipe increases. Lower mass flow rates (1–1.5 kg/s) are more suitable for GSHP operation in this case.

For tenyear economical operations of HGSHPs, suggested values for pipe length, burial depth, pipe diameter and mass flow rate for this case are 500, 1, 0.0363 m and 1 kg/s m respectively.

According to outlet/inlet GHX temperature of buried pipes and soil temperature for 10 years operation, the HGSHPs showed excellent and stable performance. Therefore, the horizontal ground source heat pump with radiant floor as heat distribution system could be recommended for cooling applications in Saharan areas in the long term.

The average value of the COP for ten years operation were found to be 3.89. It represents a high system performance when compared to other traditional techniques.

The use of HGHX coupled GSHP as an air conditioning system with improved values of the mostcost effective design parameters lead to diminished the indoor temperature of about 14.96 °C and increased the relative humidity by more than 40.97% compared to the case without HGSHP cooling system.

As an estimation of the geothermal potential of the Saharan territories, the water temperature drop between the inlet and the outlet of the GHX reach a maximum value ranging between 22 and 22.2 °C, this can be translated to an amount of heat rejected to the ground when the compressor switches on reached 83.6 kW which reflected the importance of exploitable shallow geothermal energy in the studied area.
Based on the results obtained in this study, it could be concluded that the development of the GSHP system is promising and profitable solution for providing building sector with energyefficient cooling in the hot Saharan climates and arid areas, and we recommend by taking into account all the optimized design criteria which could provide some reference on HGSHPs utilization in like these hotdry regions.
Nomenclature
T _{ in−con } : Condenser entering water temperature (°C)
T _{ out−con } : Condenser leaving water temperature (°C)
${{\displaystyle \stackrel{}{m}}}_{wCon}$ : Condenser water mass flow rate (kg/s)
COP: Coefficient of performance in cooling mode
${{\displaystyle \stackrel{}{m}}}_{wEvp}$ : Evaporator water mass flow rate (kg/s)
T _{ in−Evp } : Evaporator entering water temperature (°C)
T _{ out−Evp } : Evaporator leaving water temperature (°C)
T _{ in−GHX } : GHX entering water temperature (°C)
T _{ out−GHX } : GHX leaving water temperature (°C)
${{\displaystyle \stackrel{}{m}}}_{wGHX}$ : GHX water mass flow rate (kg/s)
T _{ in−RF } : RF entering water temperature (°C)
T _{ out−RF } : RF Leaving water temperature (°C)
${{\displaystyle \stackrel{}{m}}}_{wRF}$ : RF water mass flow rate (kg/s)
T _{ s } : Soil temperature (°C)
Q _{ b } : The heat absorbed from the building (W)
Q _{ g } : The heat transferred to the ground (W)
Q _{ Evp } : The heat absorbed from the evaporator (W)
Q _{ Con } : The heat transfer rate from condenser (W)
${C}_{p{H}_{2}O}$ : The fluid specific heat of the fluid (kJ/kgK)
E: The annual power consumption kWh/y
C _{ inv } : The investment cost of GSHPs ($/y)
C _{ elec } : The annual cost of power consumption ($/y)
C _{heatpump} : The cost of heat pump ($)
C _{pump} : The cost of circulation pump ($)
c _{earthworks} : The regional price of the drilling and refilling ($/m^{2})
c _{Pipe} : The regional cost of polyethylene pipe per meter ($/m)
n : The number of operation years (y)
H _{pump } : The pressure drop (Pa)
E _{ tariff } : The tariff of unit electricity consumption ($/kWh)
L _{ GHX } : The pipe length of heat exchanger (m)
NPT: The number of parallel pipe
Z: The depth of the ground (m)
$\stackrel{}{{T}_{s}}$ : The mean value of the ground surface temperature (°C)
A _{ s } : The amplitude of surface temperature
Subscripts
Abbreviation
HDPE: Highdensity polyethylene
HGSHP: Horizontal ground source heat pump
SEER: Seasonal energy efficiency ratio
Greek letters
α _{ g } : Ground diffusivity (m ^{2}/day)
Acknowledgments
This research was partly supported by the basic research project by the “Laboratoire des Procédés Thermiques (LPT)”, “Centre de Recherches et des Technologies de l'Energie (CRTEn)”, Tunisia. And by the laboratory of new and renewable energy in arid and saharan zones − LENREZA, Algeria.
References
 N. Giti, N. Younes, Y. Hossein, Designing and optimization of solar assisted ground source heat pump system to supply heating, cooling and hot water demands, Geothermics 82, 212–231 (2019) [CrossRef] [Google Scholar]
 R. Ghedamsi, N. Settou, A. Gouareh, A. Khamouli, N. Saifi, B. Recioui, B. Dokkar, Modeling and forecasting energy consumption for residential buildings in Algeria using bottomup approach, Energy Build. 121, 309–317 (2016) [CrossRef] [Google Scholar]
 F. Amri, The relationship amongst energy consumption (renewable and nonrenewable), and GDP in Algeria, Renew. Sustain. Energy Rev. 76, 62–71 (2017) [CrossRef] [Google Scholar]
 S. Sanaye, B. Niroomand, Thermaleconomic modeling and optimization of vertical groundcoupled heat pump, Energy Convers. Manag. 50, 1136–1147 (2009) [CrossRef] [Google Scholar]
 J. Luo, W. Xue, T. Hu, W. Xiang, J. Rohn, Thermoeconomic analysis of borehole heat exchangers (BHE) grouted using drilling cuttings in a dolomite area, Appl. Therm. Eng. 150, 305–315 (2019) [CrossRef] [Google Scholar]
 Y. Zhao, Z. Shigang, L. Xun, Costeffective optimal design of groundwater source heat pumps, Appl. Therm. Eng. 23, 1595–1603 (2003) [CrossRef] [Google Scholar]
 S. Sanaye, B. Niroomand, Horizontal ground coupled heat pump: thermaleconomic modeling and optimization, Energy Convers. Manag. 51, 2600–2612 (2010) [CrossRef] [Google Scholar]
 N. Kayaci, H. Demir, Long time performance analysis of ground source heat pump for space heating and cooling applications based on thermoeconomic optimization criteria, Energy Build. 163, 121–139 (2018) [CrossRef] [Google Scholar]
 M. Inalli, H. Esen, Experimental thermal performance evaluation of a horizontal groundsource heat pump system, Appl. Therm. Eng. 24, 2219–2232 (2004) [CrossRef] [Google Scholar]
 M.J. Kim, S.R. Lee, S. Yoon, J.S. Jeon, An applicable design method for horizontal spiralcoiltype ground heat exchangers, Geothermics 72, 338–347 (2018) [CrossRef] [Google Scholar]
 M. De Rosa, F. RuizCalvo, J.M. Corberán, C. Montagud, L.A. Tagliafico, A novel TRNSYS type for shortterm borehole heat exchanger simulation: B2G model, Energy Convers. Manag. 100, 347–357 (2015) [CrossRef] [Google Scholar]
 C. Lyu, W.H. Leong, M. Zheng, G. Chen, S. Ye, Y. Liu, Verification and analysis of a dynamic simulation model of groundcoupled heat pump with solar seasonal heat storage system, Procedia Eng. 205, 3154–3161 (2017) [CrossRef] [Google Scholar]
 Y. Noorollahi, H.G. Arjenaki, R. Ghasempour, Thermoeconomic modeling and GISbased spatial data analysis of ground source heat pump systems for regional shallow geothermal mapping. Renew. Sustain. Energy Rev. 72, 648–660 (2017) [CrossRef] [Google Scholar]
 G. Radioti, K. Sartor, R. Charlier, P. Dewallef, F. Nguyen, Effect of undisturbed ground temperature on the design of closedloop geothermal systems: a case study in a semiurban environment, Appl. Energy 200, 89–105 (2017) [CrossRef] [Google Scholar]
 H. Saibi, Geothermal resources in Algeria, Renew. Sustain. Energy Rev. 13, 2544–2552 (2009) [CrossRef] [Google Scholar]
 A. Abdelali, I.E. Nezli, A.S. Benhamida, Geothermometry application to the Continental Intercalaire geothermal aquifer: a case study of Ouargla region, Energy Procedia 119, 264–269 (2017) [CrossRef] [Google Scholar]
 The national society of electricity and gas in Algeria. Sonelgaz of the East [Google Scholar]
 N. Kayaci, H. Demir, Comparative performance analysis of building foundation Ground heat exchanger, Geothermics 83, 101710 (2020) [CrossRef] [Google Scholar]
 Webpage: http://www.algerie.prixconstruction.info (accessed June 2019) [Google Scholar]
 A. Cacabelos, P. Eguía, J.L. Míguez, E. Granada, M.E. Arce, Calibrated simulation of a public library HVAC system with a groundsource heat pump and a radiant floor using TRNSYS and GenOpt, Energy Build. 108, 114–126 (2015) [CrossRef] [Google Scholar]
 M.H. Benzaama, S. Menhoudj, C. Maalouf, A. Mokhtari, M. Lachi, Experimental and numerical analysis of the energy performance of a water/soil exchanger coupled to a cooling floor for North Africa, Geothermics 80, 8–19 (2019) [CrossRef] [Google Scholar]
 N. Nabiha, H. Majdi, K. Sami, F. Abdelhamid, Energy and exergy analysis of horizontal ground heat exchanger for hot climatic condition of northern Tunisia, Geothermics 53, 270–280 (2015) [CrossRef] [Google Scholar]
 N. Naili, M. Hazami, I. Attar, A. Farhat, Infield performance analysis of ground source cooling system with horizontal ground heat exchanger in Tunisia, Energy 61, 319–331 (2013) [CrossRef] [Google Scholar]
 N. Naili, M. Hazami, I. Attar, A. Farhat, Assessment of surface geothermal energy for air conditioning in northern Tunisia: direct test and deployment of ground source heat pump system, Energy Build. 111, 207–217 (2016) [CrossRef] [Google Scholar]
 S. Kavanaugh, K. Rafferty, Geothermal heating and cooling: design of ground source heat pump systems (ASHRAE, Atlanda, 2014) [Google Scholar]
Cite this article as: Sabrin Korichi, Bachir Bouchekima, Nabiha Naili, Messaouda Azzouzi, Performance analysis of horizontal ground source heat pump for building cooling in arid Saharan climate: thermaleconomic modeling and optimization on TRNSYS, Renew. Energy Environ. Sustain. 6, 1 (2021)
All Tables
The min max average monthly variation of outdoor temperatures and the average monthly variation of relative humidity and wind speed in the year 2019.
The list of optimum design parameters including the results values of elements of the objective function.
All Figures
Fig. 1 Main Algerian geothermal areas [15]. 

In the text 
Fig. 2 Block diagram of the design procedure. 

In the text 
Fig. 3 The model of HGSHP system developed in TRNSYS. 

In the text 
Fig. 4 3D schematic view of the typical building (1): Zoon1 (kitchen), (2), (3), (6): Zoon2, 3, 6 (bedrooms), 4: Zoon4 (bathroom) and (5): Zonn5 (living room). 

In the text 
Fig. 5 Daily cooling and transition periods. 

In the text 
Fig. 6 Hourly coolin loads of the proposed building. 

In the text 
Fig. 7 Variations of soil temperature with depth (m), Ouargla city. 

In the text 
Fig. 8 Schematic diagram of the HGSHP experimental system (modified from [23]). 

In the text 
Fig. 9 Experimental vs numerical coefficient of performance of the GSHP. 

In the text 
Fig. 10 Effect of HGHX length on the objective function and the COP of the GSHP. 

In the text 
Fig. 11 Effect of HGHX diameter on the objective function and the COP of the GSHP. 

In the text 
Fig. 12 Effect of burial depth on the objective function and the COP of the GSHP. 

In the text 
Fig. 13 Effect of mass flow rate on the objective function and the COP of the GSHP. 

In the text 
Fig. 14 Evolution of water temperatures at the external and the internal circuits in (a) the first year period, (b) typical day for L_{GHX} = 500 W/m, D_{iGHX} = 0.0363 m and Z = 1 m. 

In the text 
Fig. 15 Variation of fluid inlet, outlet ground heat exchanger water temperatures in nine years for L_{GHX} = 500 W/m, D_{iGHX} = 0.0363 m and Z = 1 m. 

In the text 
Fig. 16 Variation of average soil temperature surrounding the ground heat exchanger at the first, fifth and tenth year. 

In the text 
Fig. 17 Evolution of the heat injected into ground and the heat rejected from building in (a) the first year period, (b) typical day for L_{GHX} = 500 W/m, D_{iGHX} = 0.0363 m and Z = 1 m. 

In the text 
Fig. 18 Comparison between indoor and outdoor temperatures in a typical days (July 1 until July 30) for L_{GHX} = 500 m, D_{iGHX} = 0.0363 m and Z = 1 m. 

In the text 
Fig. 19 Comparison between indoor and outdoor relative humidity in a typical days (July 1 until July 30) for L_{GHX} = 500 m, D_{iGHX} = 0.0363 m and Z = 1 m. 

In the text 
Fig. 20 The power analysis of the first year HGSHPs operation, (a) the power consumption distribution, (b) the COP distribution for L_{GHX} = 500 W/m, D_{iGHX} = 0.0363 m, m_{wGHX} = 1 kg/s and Z = 1 m. 

In the text 
Current usage metrics show cumulative count of Article Views (fulltext article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 4896 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.