Optimization of a Ground Source Heat Pump System Using ... - METU

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

Optimization of a Ground Source Heat Pump System Using Monte-Carlo Simulation KUMUDU GAMAGE, MOSLEM YOUSEFZADEH, ERAY UZGÖREN, YASEMIN MERZIFONLUOĞLU UZGÖREN Sustainable Environment and Energy Systems Middle East Technical University Northern Cyprus Campus Kalkanli, Güzelyurt, Mersin 10 [email protected], [email protected], [email protected], [email protected]

Abstract: Ground source heat pump (GSHP) systems provide an alternative energy source for residential and commercial space heating and cooling applications by utilizing the favorable temperature profile at a certain depth under the ground. However, over-sizing or under-sizing of GSHP can lead to increase in installation cost or reduction in the performance of the ground heat exchanger. This paper proposes a framework to integrate ground thermal properties and building loads to optimize the length and number of boreholes for a campus dormitory environment. This study considers the long term temperature changes of the ground due to thermal imbalances and the uncertainty in the installation and the operational cost of GSHP system. A case study has been done for a dormitory at Middle East Technical University Northern Cyprus Campus. The results show that as the contribution of ground load into space heating increases, the net present value for operating a GSHP system over 10 years of period is also increases. The least economic break-even point as opposed to the current fossil fuel based space heating system is occurred approximately within 2 and half years of installation of GSHP system when covering the 60% of the total heating load of the building. Therefore, GSHP can be an alternative technology for the space heating applications for a campus dormitory in the Northern Cyprus. Keywords: Renewable energy, ground Source Heat Pump (GSHP), heating, ventilating and air-conditioning (HVAC), Northern Cyprus, temperature response factor 1. INTRODUCTION Ground source heat pumps (GSHP), also called geothermal heat pumps, provide significant benefits to the space heating and cooling applications due to its less energy consumption and the reduction in CO2 emissions. Therefore, ground source heat pump technology has been gaining attention in the renewable energy research field. Nevertheless, its usage is still limited as the installation cost is very high compared to the existing conventional space heating and cooling appliances. Lack of reliable and commonly accepted method to design a GSHP system has lead this technology to become less popular in the world. Hence, it is necessary to develop methodology to optimize a GSHP system minimizing the installation expenditures and maximizing savings compared to the conventional heating and spacing system. Within the last decade, number of studies has been conducted by several researchers to identify the optimal design parameters of GSHP systems in terms of thermodynamic

995

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

performance and as well as cost effectiveness of such systems. A GSHP system is consisted with ground loops, or in another words ground heat exchanger (GHX), heat pump system and heat distribution system at the conditioned space. GHX is consisted with series of closed loop pipe buried in the ground. Based on the arrangement of the pipes either buried in shallow trenches or vertical boreholes, they are characterised as horizontal loop ground heat exchangers and vertical loop ground heat exchangers. However, vertical loop ground heat exchangers are more efficient than the horizontal one as it is exposed into constant temperature profile in the ground and also it is suitable for areas where the property area is limited. In the vertical loop ground heat exchangers, boreholes are drilled in the soil and Upipes are inserted. Then the remaining space is filled with a material called grout to enhance the heat transfer from the pipes to soil and prevent the infiltration of the rain water into the borehole. Several boreholes can be connected to form a network of heat exchangers to enhance the usage of the ground volume. The design parameters of the ground heat exchangers (GHX) are mainly depended on the climate, ground thermal and hydraulic properties and characteristics of the building. Sayyaadi et al. [1] has suggested a multi-objective optimization method for a vertical loop ground source heat pump system using an evolutionary algorithm. Their optimization was mainly depended on the energy and exergy analysis. Garber et al. [2] has proposed a methodology to evaluate financial risks of over-sizing a GSHP system. The analysis was not conducted for parameterizing the GSHP design with multiple borehole cases. Nagano et al. [3] has developed a design and performance prediction tool for a vertical loop GSHP system using Infinite cylindrical heat source theory and it was implemented for small house in Sapporo, Hokkaido. The payback period for covering the increased installation cost of the installed GSHP system was about 9 years against an oil boiler and AC system. Sanaye and Niroomand [4] have introduced an optimum design process for ground source heat pump system which consists of thermal modelling of the system and choosing the optimal design parameters. However, long term effect of the thermal interactions of boreholes was not considered in the model. Li et al. [5] showed that ground temperature around the U-tube increases due to the excess heat ejected into the earth in summer and using a new multi-function ground source heat pump system can balance the ground temperature increase. Also, a number of numerical and analytical models engaged in simulating the ground temperature and performance of the ground heat exchanger can be found in the literature Yang et al. [6], Lee et al. [7], Nam et al. [8]. The present study develops a methodology to obtain optimal design parameters of a GSHP system. The objective is to design a system with the minimum break-even point which is computed considering the capital installation cost and operational cost savings from consuming less conventional energy compared to conventional HVAC system. Uncertainty in electricity prices, diesel prices capital and installation cost over the time horizon of analysis is taken into account. Four decision variables are considered in the analysis including length, number of boreholes, spacing between boreholes and contribution of the GSHP system as a ratio of the ground load into total heating load of a building.

996

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

The study first develops a model to calculate heating and cooling load of a building using OpenStudio software. Then, design parameters for GHX are calculated by a model developed on cylindrical source theory considering the long term effect on the ground temperature caused due to thermal imbalances. Further, an optimization model is proposed to maximize the net benefit value of installing GSHP compared to existing conventional space heating and cooling system. Using the developed model, a case study is established for a dormitory building at METU NCC campus, which is located in Northern Cyprus to investigate the economic feasibility of the proposed GSHP system by carrying out Monte Carlo simulations. 2. PROPOSED FRAMEWORK First, the framework starts with the building energy simulation model. This model calculates the heating load for any given building. Weather data, location, number of occupants, occupancy schedules, construction materials and heating and cooling set points are loaded into the model created in the OpenStudio software. The calculated building loads, soil and fluid properties and borehole characteristics are used as an input into the Ground heat exchanger (GHX) design model. By applying the cylindrical source theory and thermodynamic principals, the GHX Design model generates a set of suitable design parameters for a GHX system including number of boreholes, length and the spacing of the boreholes for different ratios of ground loads into the total heating loads of a building. The GHX design model is developed accounting the long term ground temperature changes in the bore-field region after years of operation of GHX. Given the set of suitable design parameters calculated from the GHX design model, the optimization model determines the optimal number of boreholes, length and spacing for a GHX system for each value of ratio to maximize the net profit of GSHP installation for a 10 year time horizon considering the uncertainty of electricity prices, diesel prices, borehole installation cost, heat pump cost and cost of water circulation pump. Finally, a break-even point analysis is carried out to find the ratio that covers the installation cost of the optimize GSHP system from the saving of operation of GSHP compared to an existing conventional space heating system. Fig. 1 shows that the proposed framework. 2.1. BUILDING ENERGY SIMULATION MODEL Building load calculation is vital when determining the GHX design parameters. The amount of heat rejection and extraction to and from the ground are depended on the building loads. In this study, open source software called OpenStudio is used to model the building loads. It is a collection of software including SketchUp, EnergyPlus and Radiance. SketchUp is used to model the three dimensional view of the building envelop. EnergyPlus is used for the building energy simulations. EnergyPlus required two types of input data such as weather data including basic location, latitude, longitude, time zone, elevation, peak heating and cooling design conditions, building description data such as geometry of the building, construction materials, internal load objects such as people, lights, Luminaries, electric equipments, gas equipments, steam equipments, and water use equipments, collections of schedules for

997

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

building activities or elements and number of thermal zones, cooling and heating set points and HVAC equipments.

Fig. 1 Overview of the proposed framework Summary of the building cooling and heating load simulations steps can be given as follows: 1. Create 3D EnergyPlus geometry using plug in for Google SketchUp. 2. Assign the space types (E.g. Medium Office, Hospital, Secondary school, etc.) to the spaces built by SketchUp.

998

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

3. Assign the spaces into thermal zones. When building is zoned, several factors to be considered such as their usage, occupancy, activity level of occupants, exposure to the sun (interior zone, exterior zone), etc. Most buildings have more than one zone. 4. Add the location and the weather data into the OpenStudio model using EPW Weather Files 5. Modify the existing construction types (materials, thickness, and conductivity values). 6. Add HVAC equipments into each zone. 7. Modify the predefined space parameters such as temperature set points and internal loads. 8. Calculate the conduction, convection, infiltration, radiation heat gain and heat losses using the basic thermodynamic principals. 2.2. GHX DESIGN MODEL Among the various methods, the most popular ones are the line source theory and the cylindrical source theory. Both methods give the radial temperature distribution of the ground as a function of dimensionless time. The current paper uses the borehole sizing equations given in the Philippe et al. [9]. It uses algebraic correlations to the cylindrical heat source solution originally developed Carslaw et al. [10], based on large number of calculations, to calculate the ground thermal response due to three thermal pulses (yearly, monthly and hourly). Further, borehole thermal interfereness is accounted using another correlation developed by Bernier et al.[11] to evaluate the temperature penalty as a variable that take in to account the real response of the ground where multiple boreholes are used. All the equations were derived assuming heat transfer in the ground occurs only by conduction, no moisture migration, no underground water movements, constant ground heat loads, uniform soil thermal properties and uniform undisturbed ground temperature. For a multiple borehole system, characteristic total length for a constant ground heat loads can be given by equation (1),

L  (qh Rb  q y R10 y  qm R1m  qh R6h ) /(Tm  (Tg  T p ))

(1)

where, L is the total length of the borehole system in case of number of boreholes is more than one. q h ,. q y and qm represent peak hourly, yearly average and highest monthly heat load transfer from ground. R10 y , R1m and R6 h represent effective ground thermal responses for three successive thermal pulse corresponding to 10 years, one month, and six hours respectively. Rb is the borehole thermal resistance. Tg is the undisturbed ground temperature. Tm , is the mean fluid temperature which is equivalent to average of the heat pump inlet and outlet temperature. T p , is the temperature penalty for long term interference to correct the undistributed ground temperature in case of presents of the multiple boreholes. Equations (2) can be used to calculate the thermal response of the ground due to three thermal pulses [9]. R  (1 / k ) f ( , rb )

(2)

999

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

where,  is the thermal diffusivity of soil, k is the thermal conductivity of soil. rb is the radius of the borehole. The function f is a simplified fitted curve to analytical solution of the cylindrical source theory. The equation (3) represents the formula for f -function and the correlation coefficients for f 6 h , f 1m , and f 10 y corresponding to R6 h , R1m and R10 y can be found in the Philippe et al [9]. However, the correlation coefficients are valid if the thermal diffusivity of the soil is in between 0.025 m2/day and 0.2m2/day only. f  a0  a1rb  a2 rb  a3  a4 2  a5 ln( )  a6 ln( ) 2  a7 rb  a8 rb ln( )  a9 ln( ) 2

(3)

Temperature Penalty T p can be obtained from equation (4). For a single borehole case, T p is equivalent to zero [9].

Tp  (q y / 2 k L) F (t / t s , B / H , NB, A)

(4)

H , represents the depth of each borehole; B represents the borehole spacing, NB , is the number of boreholes. A is the aspect ratio (number of boreholes in the longer direction over number of boreholes over the other direction). t s is the characteristic time H 2 / 9 . The function F can be defined using the equation (5) and the corresponding coefficients can be found in Philippe et al. [9]. The corresponding coefficients in the Philippe et al. [9] is valid only if the range of number of boreholes is 4 to 144 , aspect ratio is in between 1 to 9 and the B / H is in 0.05 to 0.1 and ln(t / t s ) is in the range of -2 to 3. These ranges are considered as the constraints of the model. 36

F   bi  ci

(5)

0

Borehole thermal resistance is calculated by equation (6).

Rb  Rg  ( Rconv  R p ) / 2

(6)

Rconv  1 / 2 rp,in hconv

(7)

R p  ln(rp,ext / rp,in ) / 2 k p

(8)

Rg  1 /(4 k g )[ln(rb / rp,ext )  ln(rb / Ls )  (k g  k ) /(k g  k ) ln(rb /(rb  ( Ls / 2) 4 ))] 4

4

(9)

where Rconv , R p and Rg are respectively, the convective resistance inside the tube, the conduction resistance for each tube and the grout resistance. hconv is the film convection coefficient, rp ,in and rp,ext are the inner and outer radius of the pipe, k p is the thermal conductivity of the pipe material, k g is the thermal conductivity of the grout, k is the ground thermal conductivity, and Ls is the shank spacing. Fig. 2 illustrates the overview of the GHX design model. First, for a given percentage of the total heating load, length of borehole is calculated assuming the heating load is supplied only

1000

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

by a single borehole (set T p  0 ). In the presence of multiple boreholes, iterations are necessary as penalty temperature is depended on H which is at the beginning an unknown value. For each number of boreholes, best arrangement of them was decided, considering having a maximum distance between boreholes. Then, from obtained arrangements, the distance between boreholes as well as the aspect ratio is found. Then, taking these as secondary inputs into the model, borehole depth H is calculated by dividing the total length by number of boreholes. Taking the calculated borehole depth as an input, a new T p value is calculated. The procedure is continued until a minimum penalty temperature and the constraints defined at the above are satisfied.

Fig. 2 Overview of the GHX Design Model

1001

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

2.3. OPTIMIZATION MODEL The main objective of the optimization model is to determine the optimal number of boreholes, their lengths and spacing for each ratio. The net present value (NPV) calculation is conducted considering the uncertainty of electricity prices, diesel prices, borehole installation cost, heat pump cost and cost of circulation pump. A Monte Carlo simulation is carried out for the NPV analysis by taking into account of the uncertainties. Finally, a break-even point analysis is carried out to find the ratio that covers the installation cost of the optimize GSHP system (for each ratio) from the saving of operation of GSHP compared to an existing conventional space heating system. 2.3.1. Problem formulation The Net present value (NPV) is calculated for varying levels of the ratio (r), i.e., 0.1, 0.2, …, 1, over a time horizon T. NPVr = PVSOPr - ICr , r = 0.1, 0.2,...,1

(11)

where, PVSOP is the present value of savings (TL) from operating GSHP over a conventional boiler and IC is the total installation cost (TL) of the GSHP system. The IC includes the borehole installation cost (BIC), heat pump installation cost (HPIC) and water circulation pump installation cost (WPIC). ICr = BICr + HPICr + WPICr

(12)

Borehole installation cost (BIC) is assumed to be constant per unit length with the depth increases. BICr = NBr x Hr x UBIC

(13)

where, NB is the number of boreholes required, H is the length of each borehole (m) and UBIC is the borehole installation cost per unit length (TL/m). Heat pump installation cost (HPIC) is mainly depended on the ratio of the heating load obtained from the GHX. HPICr = NHPUr x HPUC

(14)

where, NHPU is the number of heat pump units required; HPUC is the heat pump unit cost (TL/unit). Number of heat pump units required is depended on the input power required to drive the

1002

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

GSHP system and the capacity of a single heat pump available in commercially. Calculation of the NHPU is given by the equation (15). NHPUr = (rq/COP) x (1/1000) x (1/CPHPU)

(15)

where, q is the total heating load (W), CPHPU is the capacity of the heat pump unit (kW/unit). Water circulation pump installation cost (WPIC) can be defined by equation (16). WPICr = NWPUr x CPWPUr x WPUC

(16)

where, NWPU is the number of water circulation pump, CPWPU is the capacity of a water circulation pump (kW) and WPUC is the water circulation pump unit cost (TL/kW). NWPU is chosen by equalling it to the number of boreholes in the shortest length of the area selected for installation of the boreholes. (17) NWPUr =br where, b is the number of boreholes in the shortest length of the area of borehole installation. CPWPU is given by equation (18). CPWPUr = (∆p x V x 1/η)/ NWPUr

(18)

where, ∆p is the pressure difference across the system (kPa), V is the volumetric flow rate (m3/s) and η is the efficiency of the water pump. To calculate the present value of savings from operating GSHP system (PVSOP), following equation can be used, PVSOPr =

T

 SOP ,

r j

/(1  iR ) j , j=1,2,…..,T

(19)

j 1

where, SOP is the annual saving from operating GSHP system over conventional boiler for space heating (TL). SOPr,j = FCCr,j – ECGSHPr,j

(20)

where, FCC is the fuel cost of the conventional boiler (TL), ECGSHP is the electricity of operating GSHP system (TL). Calculation of fuel consumption cost of the boiler (FCC) is given in the following equation, FCCr,j = COFj x FCMr

(21)

where, COF is the fuel cost per litre (TL/L) and FCM is the fuel consumption of the conventional boiler (L). FCM can be calculated as follows, FCMr = rq x NH x ND x (1/HV)

(22)

1003

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

where, NH is the number of operating hours of the space heating appliance (hrs), ND is the number of days required heating (days), HV is the heating value of the fuel. Electricity consumption cost of GSHP system (ECGSHP) can be calculated by equation (23). ECGSHPr,j = COEj x ECMr

(23)

where, COE is the unit electricity cost (TL/kWh), ECM is the electricity consumption of the GSHP system (kWh). ECMr =( rq/COP) x NH xND x (1/1000)

(24)

where, COP is the coefficient of performance of the GSHP system. Break-even point for covering the installation cost of GSHP system from savings from operational cost compared to conventional boiler is given in the following equation, ICr =

n

 SOP ,

r j

/(1  iR ) j

(25)

j 1

Where, n is the break-even point (years). 3. RESULTS and DISCUSSION The building energy simulation model has been implemented in a case study for a campus dormitory building located in Middle East Technical University, Northern Cyprus Campus, Kalkanli, Guzelyurt area. The dormitory is a five story building consisted of student rooms, kitchens, bathrooms and full control on heating and cooling. As the students are on summer vacation, dormitory building is unoccupied during the months of June-September. Therefore, calculation of cooling load is not required. Input parameters for the building load calculation model for the dormitory building are presented in Table 1. The three dimensional view of the dormitory building modelled using OpenStudio software is illustrated in Fig. 3. Table 2 shows the simulated peak hourly, peak monthly and yearly average heating load for the dormitory building. The results show that peak hourly heating load is occurred in a day on February while the peak monthly heating load is happened on January. The GHX design model was implemented for the same building. First, a maximum area for installation of the borehole was selected. The area was selected such that distance from the building to the boreholes will be the least. This will cause less piping and hence the less cost. The length and width of the selected rectangular area is 70 m and 14 m respectively. Building heating loads calculated using the building load calculation model was taken as the preliminary input. Characteristic of the soil was obtained from the Geological map of Cyprus, revised in 1995 [12]. The soil in the campus area is mainly consisted of sandstone, sandy, marls, limestone and biocalcarenites. Thermal conductivity, thermal diffusivity and the heat capacity for soil in the

1004

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

campus area were obtained in correspondence with experimental measurements conducted in boreholes installed at a various location of the Cyprus by Florides, G. A., et al. [13]. Table 3 presents the first set of input parameters into GHX design model for designing a hypothetical ground heat exchanger system for the campus dormitory building. The undisturbed ground temperature of the Cyprus is approximately 220C [14]. According to the ASHRAE guidelines, entering water temperature to the heat pump should be less than the undisturbed ground temperature - [5,10] 0C [15]. Therefore, entering water temperature to the heat pump was taken as 14.50C. Simulations are carried out for the GHX design model and validated using the example given in the Philippe et al. [9] and Bernier et al. [11] for the multiple borehole cases. The length, number of boreholes and the spacing between the boreholes were calculated for different percentages of the building heating load using the equations described in the section 2.2. Table 1 Input parameters for building energy simulation model Parameters Values or Source Geometry AutoCad plan weather data EPW Weather Files Number of students 300 Materials similar to block B Total floor area (m2) 7681 Number of zones 5 0 Heating Set point Temperature ( C) 22

Fig. 3 Modeled 3D view of dormitory building using SketchUp The GHX design model was simulated for different borehole configurations with bore spacing as 3m, 4m 5m, and 6m for each ratio. 130 possible configurations were found for the

1005

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

installation of boreholes within the limited area after filtering the invalid configurations according to the model described above. Table 2 Simulated heating load for dormitory building Heating load value (W) peak hourly ground load

268,900

monthly ground load

104,110

yearly average ground load

4,212

It was found that, when adding the boreholes into the longest direction in the area by removing them in the shorter direction while keeping the total number of boreholes constant, borehole wall temperature drop is reduced compared to that of previous configurations. Hence it leads to a shorter total length in the borehole system (Table 4). Also, as the number of boreholes increased in the longer direction while keeping the number of boreholes in the other direction as constant, temperature drop at the boreholes is increased to an optimum value and then decreased. Therefore, for each ratio of the ground load, finding the best configuration that leads to decrease in the temperature drop should be found. The Fig. 4 shows that, how the penalty temperature effect the total length of the borehole system when the GSHP contribution to the total heating load is only 20%. The least cost configuration for each ratio was found by analysing the configuration that gives the least value for the product of the number of boreholes into the depth of the borehole. The Table 5 depicts the simulated result for the best configurations for each ratio. Also, it was noticed that if the ground temperature is less than that of the Cyprus like 100C (approximately near to the ground temperature in Japan [3]), required length for the same number of boreholes with the same configuration will be significantly longer (Fig. 5). Further, analysis was carried out to find the borehole depth for the same best configurations when the thermal diffusivity is higher than that of the Cyprus. It was found that the thermal diffusivity significantly effect on the borehole depth and the penalty temperature (Fig. 6). The optimization model described in the section 2.3 was implemented. The least cost configuration of the borehole system for each ratio was taken as the input into the optimization model. To calculate the GSHP installation cost and savings from operational cost, several cost parameters were needed. As the required length for borehole system for each ratio was calculated considering a penalty temperature for 10 year period, the economic life time to compute the savings from operational cost also taken as 10 year period. The minimum and maximum values for unit cost of these parameters are depicted in the Table 6. Different sources are used to obtain the values for these cost parameters and they are listed in the Table 6. Besides the parameters listed in the Table 6, electricity cost and diesel prices for next 10 year period is also required. A regression analysis was conducted to find the trend of electricity and diesel prices increases over the past 10 years and forecasted the same for the next 10 year period accordingly.

1006

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

A Monte Carlo method was used to generate sample of 2000 cost parameter sets in order to account the uncertainty in the cost values for installation cost, electricity and diesel prices. To generate the sample data for the electricity and diesel prices, normal standard distribution was used to create different set of forecasted prices for the next 10 years as changes in these prices are not necessarily uniform. The assumption of uniform distribution of the capital installation cost was used to provide a common base for comparison between different installation cost scenarios. When calculating the water circulation pumps installation cost, efficiency of the water pumps (η) was taken as 90%. To calculate the heat pump installation cost and the electricity consumption of the GSHP system, COP value was assumed as 4. Fuel consumption of the boiler was calculated taking the heating value (HV) of the diesel as 9,271 Wh/L. Table 3 First set of input parameters to the GHX design model Ground properties Thermal conductivity (Wm-1K-1) Thermal diffusivity (m2day-1) Undisturbed ground temperature (°C) Fluid properties

1.25 0.025 22

Thermal heat capacity (Jkg-1K-1) Mass flow rate (kgs-1kW-1) Heat pump inlet temperature (°C) Borehole characteristics Borehole radius (m) Pipe inner radius (m) pipe outer radius (m) Grout thermal conductivity (Wm-1K-1) Pipe thermal conductivity (Wm-1K-1) Shank spacing (m)

4200 0.074 14.5

Convection coefficient (Wm-2K-1)

r 0.2 0.2 0.2

0.054 0.0137 0.0167 1.00 0.45 0.0471 1000

Table 4 Penalty temperature for different configuration of the boreholes B NB a b A Tp L0 L H 3 24 6 4 1.5 -0.67 1239 1337 56 3 24 8 3 2.6 -0.64 1239 1333 55 3 24 12 2 6 -0.31 1239 1283 53

Present value of savings from operating GSHP system and break-even point calculations were carried out by assuming 12% interest rate. Net present value and borehole installation cost for each ratio for the best configuration of the borehole system is shown in the Fig. 7 and break-even point calculations are given in Fig. 8. It can be seen that as the ratio increases, borehole installation cost increase up to the ratio of 0.5 and a suddenly decline in the 0.6 ratio

1007

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

-0.66 -0.67 -0.68 -0.69 -0.7 -0.71 -0.72 -0.73 -0.74 -0.75 -0.76 -0.77

Total borehole length (m)

1,354 L

1,352

Tp

1,350 1,348 1,346 1,344 1,342 1,340 1,338 1,336 23

28

33

38

Tp(0 C)

and then it increases dramatically with the increase in ratio. This sudden decline is cuased due to length decline in the 0.6 ratio compared to that of 0.5 ratio. However, net present value is incresed with ratio, since when the contribution from the GSHP increses, operation cost is reduced. Thats, why optimal design ratio is become 0.6 when considering the break-even point calculations. Within the 10 year economic life time, GSHP system with all different ratios covers the capital installation cost.

43

Number of boreholes

Fig. 4 Total borehole length and temperature penalty for different number of boreholes when r =0.2 Table 5 The best configurations for borehole system for each ratio r B NB a b A H L 0.1 6 10 5 2 3 64 636 0.2 3 32 16 2 8 36 1138 0.3 4 36 18 2 9 41 1460 0.4 4 36 18 2 9 55 1962 0.5 4 36 18 2 9 68 2455 0.6 3 63 21 3 7 33 2033 0.7 3 63 21 3 7 39 2426 0.8 3 63 21 3 7 45 2812 0.9 3 63 21 3 7 51 3188 1.0 3 63 21 3 7 57 3551

1008

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

Depth of the borehole (m)

90 80 70

60 50 40 30

20 0

10

at 0C 10 0C H atH 10

220CC HHatat22

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

ratio

Fig. 5 Borehole depth for the best configuration for each ratio with different ground temperatures 250

Borehole depth (m)

α =0.025 m2/day α =0.025 m2/day

α =0.15 m2/day α =0.15 m2/day

200 150 100 50 0 0.1

0.2

0.3

0.4

0.5

0.6

ratio

0.7

0.8

0.9

1

Fig. 6 Borehole depth for best configuration for each ratio at different thermal diffusivity values in the soil

1009

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

Table 6 The minimum and maximum values of the cost parameters Parameters Borehole Installation cost (TL/m) Min Value Max Value

Values

Source

74.23 148.47

[2] [2]

Water-water heat pump cost (10 kW per one piece) Unit cost (TL/piece ) Min value 2,185.40 Max Value 17,483.20

[16] [16]

Water pump cost unit pump cost (TL/kW) Min Value Max Value

[1] [1]

708.07 1000

1.00 Net present value

0.90

Net present value (Millions TL)

0.80

Borehole installation cost

0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 0.1

0.2

0.3

0.4

0.5

0.6

ratio

0.7

0.8

0.9

1.0

Fig. 7 Net present value and Borehole installation cost for beast configuration for each ratio

1010

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

1.50

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

NPV(Million TL)

1.00

0.50

0.00 0

1

2

3

4

5

6

7

8

9

10

-0.50

time (Years)

Fig. 8 Break-even point for beat configuration for each ratio 4. CONCLUSION To find optimal configuration of the boreholes for a GSHP system, a new methodology has been developed. The methodology is consisted with three sub models such as building energy simulation model, GHX design model and optimization model. Each model was implemented for a campus dormitory building located at Middle East Technical University, Northern Cyprus campus. The results showed that as the number of columns increased by decreasing the rows for the same number of boreholes for given matrix of borehole system, the temperature drop at each bore is reduced. Hence required length for the borehole system is reduced. Also, an interesting phenomenon was observed for the penalty temperature when increasing the number of columns in a matrix of bore system, while keeping the raw constant. The temperature drop is increased (penalty temperature value is decreased) up to optimal value and then decreased. Therefore, the simulations should be conducted for different configurations of the borehole system to find the least length for a borehole system. Also, net present value calculations show that as the contribution of the GSHP system to the heating load of the building increases, net present value also increased. The least break-even occurred for the 60% of the GSHP contribution to the system which was 2 and half years. Also, every ratio was able to break-even the capital cost within the

1011

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

economic life time. The main reason for this is the Cyprus has higher ground temperature, higher thermal conductivity in the soil. Therefore, Cyprus has a good potential for the ground source heat pump technology. Acknowledgements Dr. Murat Farioğlu, Dr. Bertuğ Akintuğ, and Mr. Mustafa Ozan Uçar from METU, NCC provided valuable data for this paper. References [1] H. Sayyaadi, E. H. Amlashi, and M. Amidpour, “Multi-objective optimization of a vertical ground source heat pump using evolutionary algorithm,” Energy Convers. Manag., vol. 50, no. 8, pp. 2035–2046, Aug. 2009. [2] D. Garber, R. Choudhary, and K. Soga, “Risk based lifetime costs assessment of a ground source heat pump (GSHP) system design: Methodology and case study,” Build. Environ., vol. 60, pp. 66–80, Feb. 2013. [3] K. Nagano, T. Katsura, and S. Takeda, “Development of a design and performance prediction tool for the ground source heat pump system,” Appl. Therm. Eng., vol. 26, no. 14–15, pp. 1578–1592, Oct. 2006. [4] S. Sanaye and B. Niroomand, “Thermal-economic modeling and optimization of vertical ground-coupled heat pump,” Energy Convers. Manag., vol. 50, no. 4, pp. 1136–1147, Apr. 2009. [5] S. Li, W. Yang, and X. Zhang, “Soil temperature distribution around a U-tube heat exchanger in a multi-function ground source heat pump system,” Appl. Therm. Eng., vol. 29, no. 17–18, pp. 3679–3686, Dec. 2009. [6] W. B. Yang, M. H. Shi, and H. Dong, “Numerical simulation of the performance of a solarearth source heat pump system,” Appl. Therm. Eng., vol. 26, no. 17–18, pp. 2367–2376, Dec. 2006. [7] C. K. Lee and H. N. Lam, “Computer simulation of borehole ground heat exchangers for geothermal heat pump systems,” Renew. Energy, vol. 33, no. 6, pp. 1286–1296, Jun. 2008. [8] Y. Nam, R. Ooka, and S. Hwang, “Development of a numerical model to predict heat exchange rates for a ground-source heat pump system,” Energy Build., vol. 40, no. 12, pp. 2133–2140, 2008. [9] M. Philippe and M. Bernier, “Vertical Geothermal Borefields,” ASHRAE J., 2010”

1012

7th International Ege Energy Symposium & Exhibition June 18-20, 2014 Usak, Turkey

[10] H. S. Carslaw and J. C. Jaeger, Conduction of heat in solids. Oxford [Oxfordshire]; New York: Clarendon Press ; Oxford University Press, 1986. [11] M. A. Bernier, A. Chahla, and P. Pinel, “Long-Term Ground-Temperature Changes in GeoExchange Systems,” ASHRAE Trans., vol. 114, no. 2, pp. 342–350, Oct. 2008. [12] “Geological map of Cyprus, revised in 1995.” [13] G. A. Florides, P. D. Pouloupatis, S. Kalogirou, V. Messaritis, I. Panayides, Z. Zomeni, G. Partasides, A. Lizides, E. Sophocleous, and K. Koutsoumpas, “The geothermal characteristics of the ground and the potential of using ground coupled heat pumps in Cyprus,” Energy, vol. 36, no. 8, pp. 5027–5036, Aug. 2011. [14] G. Florides and S. Kalogirou, “Ground heat exchangers—A review of systems, models and applications,” Renew. Energy, vol. 32, no. 15, pp. 2461–2478, Dec. 2007. [15] L. Lamarche, G. Dupré, and S. Kajl, “A new design approach for ground source heat pumps based on hourly load simulations,” month, vol. 16, pp. 0–02, 2008. [16] “Brine to Water Heat Pump Water Heater (CGS-95) [Online]. Available: http://sprsun.en.made-in-china.com/product/OKHJlLGUgRWb/China-Brine-to-WaterHeat-Pump-Water-Heater-CGS-95-.html. [Accessed: 28-Mar-2014].

1013