- Email: [email protected]

Thermal and economical analysis of a central solar heating system with underground seasonal storage in Turkey A. Ucar*, M. Inalli Department of Mechanical Engineering, Firat University, 23279 Elazıg˘, Turkey Received 24 April 2004; accepted 20 September 2004 Available online 18 December 2004

Abstract Thermal performance and economic feasibility of two types of central solar heating system with seasonal storage under four climatically different Turkey locations are investigated. The effects of storage volume and collector area on the thermal performance and cost are studied for three load sizes. The simulation model of the system consisting of flat plate solar collectors, a heat pump, under ground storage tank and heating load based on a finite element analysis and finite element code ANSYSe is chosen as a convenient tool. In this study, the lowest solar fraction value for Trabzon (418N) and the highest solar fraction value for Adana (378N) are obtained. Based on the economic analysis, the payback period of system is found to be about 25–35 years for Turkey. q 2004 Elsevier Ltd. All rights reserved. Keywords: Seasonal storage; Solar energy; Thermal and economical analysis

1. Introduction Fast worldwide population growth leads to a quickly increasing energy demand. To maintain the standard of living in industrialized countries and improve the situation in developing countries, energy consumption cannot be avoided and energy can be used * Corresponding author. Tel.: C90 0424 237 0000/5233; fax: C90 0424 2415526 E-mail address: [email protected] (A. Ucar). 0960-1481/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.renene.2004.09.015

1006

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

Nomenclature Ac bo bp CA CE CF cp COP df F FR F/ IT i m MS N Qal QL Qs Qu Rv S (UA)L UL Ta Th Tiref Toref Ts W ðtaÞ h

collector area (m2) collector constant characteristic coefficient of heat pump area dependent cost ($/m2) fixed cost, $ cost of energy from fuel ($/GJ) specific heat of fluid (J/kg K) coefficient of performance of heat pump down payment solar fraction collector heat removal factor collector efficiency factor monthly average daily solar radiation incident on the collector per unit area (MJ/m2) assumed annual interest rate on mortgage mass flow rate (kg/s) solar system performance degrade period of economic analysis design house heat load (W) heat load of building (W) heat loss to the surrounding earth from the storage tank (W) monthly average of useful solar energy (W) resale value of solar system ($) cost per house ($) building loss coefficient (W/K) collector overall energy loss coefficient (W/m2 K) ambient temperature (K) temperature of fluid in heat exchanger (K) inside design air temperature (K) winter design outside air temperature (K) fluid temperature at the inlet to the collector (K) heat pump power average transmissivity absorptivity product collector efficiency

much more efficiently and with a higher share of renewable sources. The long-term storage of thermal energy is to store solar heat from the summer to the winter for space heating. The basic idea of seasonal storage of solar energy is to compensate the seasonal discrepancy between solar energy supply and heating load. The basic storage types investigated were: water tank on the ground, vertical pipes in clay and a rock cavern. Breger et al. [1] developed a comparative analysis of the heat transfer from boreholes

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

1007

and U-tubes using analytical solutions, finite element modeling and the available simulation model. This analysis is used to support the development of a methodology by which the heat transfer of any U-tube configuration can be modeled by appropriately specifying parameters in the borehole storage simulation model. A central solar plant with seasonal storage (CSHPSS), which is under construction in Korea, is simulated using TRNSYS to predict thermal performances and economic aspects by Chung et al. [2]. Their system consists of two arrays of collectors, a medium- sized storage tank and two thermal loads. It is found that TRNSYS prediction for system is that about 39% of the total heating load of 885,000 MJ/year can be provided from the sun. Nordell and Hellstro¨m [3] evaluated performance of a solar-heated low temperature space-heating system with seasonal storage in the ground using the simulation models TRNSYS and MINSUN together with the ground storage module DST. They implied an economically feasible design for a total annual heat demand of about 2500 MWh. It was found that total annual cost of the solar heating system was reduced by about 20% to about 800 SEK MWhK1, which was lower than the best conventional alternative. Pahud [4] analysed a central solar heating plant with seasonal ground storage by dynamic system simulations. A reference system involving a collector area, water buffer storage and ground duct storage some improvements of the system control are also investigated to assess the influence on the overall thermal performances of the system. Reuss et al. [5] investigated the thermal performance of duct systems with vertical heat exchangers. Thermal performance of these systems is influenced by the heat and moisture movement in the area surrounding the heat exchangers. In this study, the combined heat and moisture transport was simulated by using a computational method for temperatures up to 90 8C. The design data of an interseasonal storage system in Greece are compared with the results of two CSHPSS simulation software codes, MINSUN and SOLCHIPS by Argiriou [6]. The comparison showed good agreement and therefore these user-friendly tools can be used with confidence for the design of CSHPSS under Greek weather conditions. Yumrutas¸ and ¨ nsal [7] investigated annual periodic performance of a solar assisted ground-coupled U heat pump space heating system with seasonal energy storage in a hemispherical surface tank using analytical and computational methods. Their computational model is based on a hybrid analytical-numerical procedure. It is found that the earth type has only a small effect on annual energy fractions but a stronger effect on the thermal storage transient temperature and on the annual COP of the heat pump. In another study of Yumrutas¸ and ¨ nsal [8], an analytical solution is obtained for the transient temperature field outside a U hemispherical surface tank by an application of the Complex Finite Fourier Transform and the Finite Bessel Transform techniques. In this study, the effects of geological structure surrounding the tank, insulation thickness and tank size on the water temperature in the tank and also the effect of tank size on the heat pump COP for different types of earth were investigated. Turkey is located in a relatively advantageous geographical position and Turkey has an average 2640 h (7.2 h/day) of sunshine per year and the average solar heat flux exceeds 1311 kWh/m2-year (3.6 kWh/m2-day) annually. Turkey’s solar energy potential has been estimated to be 26.4 million toe (ton oil equivalent) as thermal and 8.8 million toe as electricity. But, direct use of solar thermal is small, 52 ktoe in 1995 [9]. To date solar energy has been utilized only for domestic hot water usage in small scale systems.

1008

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

Fig. 1. Schematic diagram of a central solar heating system with seasonal storage.

Solar heating systems with the seasonal storage have been studied theoretically in the last decade. The utilization of solar energy and its contribution to the economy of Turkey has increased recent years [10]. This study consists of performance and economic feasibility of two types of central solar heating system with seasonal storage. The simulation is presented for four locations of Turkey under different climate conditions. The temperature distribution of the ground is solved by using finite element method by using ANSYSe code.

2. System description As it is seen schematically in Fig. 1, the model system includes flat plate solar collectors, a heat pump, an under ground storage tank and a heating load. Solar energy absorbed by the solar collectors is transferred to storage tank in the ground. The heat pump operates only when the temperature of the water in the tank is in sufficient to keep the house at the required inside design air temperature. Modelling of the system components, including load and the method of solution are given in the following sections [10]. 2.1. Storage tank The selection of storage type is an important part of a system design. Therefore, the storage tank is usually modelled as a cylindrical insulated tank in the literature. In this study, a trapeze type of storage is chosen and compared with the cylindrical type of storage and Fig. 2 shows the schematic diagram of the two types of storage.

Fig. 2. The schematic diagram of the two types of storage.

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

1009

The energy balance for seasonal storage tank is expressed as follows: rs Vs cs

dTs Z Qu K QL K Qs dt

(1)

The collector useful energy gain Qu and heat load of building QL will be discussed in the Sections 2.2 and 2.3. The heat loss to the surrounding earth from the storage tank Qs is given by Qs Z Qis K Qas

(2)

where Qis is the net energy input rate and evaluated as follows: Qis Z Qu K QL C W

(3)

where W is heat pump work. The energy accumulation rate in the storage tank is defined as Qas Z rs Vs cs

dTs dt

(4)

2.2. Solar collector The useful energy gain of the flat plate collectors is calculated by [11] Qu Z hAc IT

(5)

T K UL ðTs K Ta Þ Qu Z Ac FR ½ðtaÞI

(6)

or

is the where Ac is the collector area, FR is the collector heat removal factor, ðtaÞ transmittance, UL is the collector overall loss coefficient, IT is the instantaneous solar radiation incident on the collector per unit area, Ta is the ambient air temperature and Ts is the fluid temperature at the inlet to the collector. The collector heat removal factor FR is given by FR Z

_ p mc = _ pÞ ½1 K eKðAc UL F Þ=ðmc Ac U L

(7)

depend on the constructional details of the where the collector parameters FR, UL and ðtaÞ collector. Table 1 summarizes some details of the collector arrays. The collector areas were chosen 10, 20, 30, 40 and 50 m2. Table 1 Collector parametric data Parameters

Value

Type Collector overall energy loss coefficient, UL (W/m2 K) Collector heat removal factor, FR Average transmissivity absorptivity product, ðtaÞ Collector parameter, bo

Glass cover black paint 4.5 0.95 0.76 0.15

1010

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

Table 2 Load parametric data Parameters

Value

Design house heat load, Qal (kW, per house) Building loss coefficient, (UA)L (W/K, per house) Inside design air temperature, Tiref (K) Winter design outside air temperature, Toref (K)

10 303 294 261

2.3. Heating loads Solar energy is supplied from the storage tank to a building and instantaneous heat load for building is evaluated by using the equation QL Z ðUAÞL ðTiref K Ta Þ

(8)

The heat loss coefficient (UA)L of the building is calculated as ðUAÞL Z Qal =ðTiref K Toref Þ

(9)

where Qal is the annual heating load of the building. If the energy demand of the building is provided by a heat pump, the instantaneous heat load of the building is QL Z WðCOPÞ Z Wbp Th =½Th K Ts

(10)

where bp is the characteristic coefficient of the heat pump, which is in the range of 0.2–0.3 [10]. The solar fraction is defined as the utilized solar heat divided by the total heat demand and calculated as X FZ Qu K Qs =QL (11) Load parametric data inputs are given in Table 2. 2.4. Climate The data for the monthly average solar radiation on a horizontal surface and the monthly average outside air temperature for four different climate locations were ¨ nsal and Dog˘antan [12]. Four different climate locations in Turkey are taken by U shown in Fig. 3. The geographic location of Turkey, being between 36 and 428N latitude, is very advantageous from solar energy point of view and especially, passive utilization. Fig. 4 shows monthly average weather data for those locations.

3. Simulation The transient heat transfer between storage and the surrounding ground is modelled by using a finite element model. The finite element model of cylindrical type storage is used

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

1011

Fig. 3. The four different climate locations in Turkey.

to compute the solution to the time-dependent heat conduction Eq. (12): 1 v vT v2 T 1 vT r C 2 Z r vr vr a vt vz

(12)

In the present study; finite element code ANSYSe is chosen as a convenient tool. The ANSYS program has many finite element analysis properties, ranging from a simple, linear, static analysis to a complex, nonlinear, transient dynamic analysis. In this study, Plane55 Thermal Solid with two-dimensional thermal conduction property has been chosen as element type. The element has four nodes with a single degree of freedom, temperature, at each node. The finite element model of the cylindrical and trapeze storage and the surrounding ground is developed as shown in Fig. 5. The data for monthly average of daily solar radiation and for the monthly average of daily outside air temperature for four different climatically Turkey locations were transferred to ANSYS as data files. An initial water temperature was assumed equal to the deep ground temperature taken as

Fig. 4. Monthly weather data for Elazıg˘, Adana, Istanbul and Trabzon.

1012

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

Fig. 5. The finite element model of the two types of storage and the surrounding ground: (a) cylindrical storage (b) trapeze storage.

10 8C. The input data to ANSYS are data for the monthly outside air temperature, the earth data and storage size data.

4. Economic analysis The economic analysis of solar heating system performed by using the simplified P1 and P2 method [11]. Solar savings S are given by S Z P1 CF QL F K P2 CS

(13)

Table 3 Parameter values for economic analysis [2] Parameter

Value 2

Area dependent cost, CA ($/m , per house) Fixed cost, CE ($, per house) Solar system performance degrade, MS (%) Down payment, df (%) Assumed annual interest rate on mortgage, i (%) Resale value of solar system, Rv ($, per house) Period of economic analysis, N (years)

1200 4000 1 5 5 4000 50

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

1013

Table 4 Properties of the geological structures Earth type

Density (kg/m3)

Thermal conductivity (W/m K)

Thermal diffusivity (m2/s)

Heat capacity (J/kg K)

Clay Coarse gravel Granite Sand

1500 2050 2640 1500

1.4 0.519 3 0.3

1.1!10K6 1.39!10K7 1.4!10K6 2.5!10K7

848 1842 811 800

Fig. 6. Solar fraction and cost per house as a function solar collector area for three load sizes. (a) 1 house load, (b) 50 houses load, (c) 500 houses load.

1014

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

where CS is calculated as follows CS Z CAAc C CE P1 and P2 are calculated by h N i P1 Z ð1 C df Þ=ðdf K iÞ 1 K ð1 C iÞ=ð1 C df Þ

(14)

(15)

and P2 Z 1 C P1 MS K Rv ð1 C df ÞKN

(16)

Fig. 7. Solar fraction and cost per house as a function storage volume for three load sizes. (a) 1 house load, (b) 50 houses load, (c) 500 houses load.

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

1015

Fig. 8. Solar fraction as a function solar collector area for cylindrical storage under various climatic conditions for the 50 houses load.

The payback period Np is h i Np Z ln 1 K ðP2 CS =CF QL FÞðdf K iÞ =ð1 C df Þ =ln ð1 C iÞð1 C df Þ

(17)

Table 3 summarizes some parametric values used in the economic analysis.

5. Results and discussion The solar fraction is defined as the utilized solar heat divided by the total heat demand. The simulations were performed for four different climate locations; Adana, Elazıg˘, I˙stanbul and Trabzon. The covering latitudes of those areas were 37, 38.7, 40 and 418N, respectively. In this article, the load sizes considered were 1, 50 and 500 housing units. Two types of seasonal storage were simulated: trapeze and cylindrical. Four different types

Fig. 9. Solar fraction as a function storage volume for cylindrical storage under various climatic conditions for the 50 houses load.

1016

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

Fig. 10. Cost as a function solar collector area for cylindrical storage under various climatic conditions for the 50 houses load.

of soil, namely, coarse gravel, sand, clay and granite were considered in the calculations and the thermal properties used are given in Table 4. The results in this study are valid for a periodic operation case of the system. The system needs approximately 15–20 years in order to reach the periodic operation regime. In the result of economic analysis the payback is found to be in the range of 25–35 years. Argiriou [6], found the solar fraction value 0.88 under the condition of 167 m2 collector area and 549 m3 storage volume. In the present work by choosing the same tank volume and collector area, the obtained solar fraction is 0.91. Comparing the two studies it is observed that there is nearly a 3% deviation between them. The solar fraction and cost as a function of solar collector area for three loads, calculated using the weather data for Elazıg˘, are given in Fig. 6. The solar fraction and cost increase with increase on the solar collector area for two storage systems with the same volume (420 m3/house). When a trapeze tank system is used, the solar fractions are higher because the surface area of cylindrical tank is larger than that of the trapeze tank.

Fig. 11. Cost as a function storage volume for cylindrical storage under various climatic conditions for the 50 houses load.

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

1017

Fig. 12. Annual variation of the storage water temperature and solar fraction for trapeze tank embedded in different earth types.

1018

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

Fig. 7 shows the solar fraction and cost per house as a function of storage volume for three loads by choosing the same collector area (20 m2/house). Solar fractions and costs increase with increasing storage volume. The higher solar fraction takes place for the trapeze tank, compared with the other system. The cost per house decreases and solar fraction increases with the increase of the house load. Fig. 8 shows the solar fraction as a function of solar collector area in the case of the 50 houses load and cylindrical storage for the four locations of Turkey. The required collector area to achieve a certain solar fraction is reduced as the latitude decreases[6]. Simultaneously, Fig. 9 shows the solar fraction variation with respect to storage volume. The lowest rates are obtained for Trabzon 418N) and the solar fraction value is reached to 100% for Adana (378N). The Storage volume required to achieve a solar fraction of 90% is 21,000 m3 for Adana and 50,000 m3 for Trabzon. Figs. 10 and 11 give the variations of cost per house with respect to the solar collector area and the storage volume, respectively, for 50 houses load. The highest costs were found for Trabzon and the lowest for Adana. It is observed that the required collector area and the storage volume decrease as the latitude decreases and therefore the cost per house is the lowest for Adana. Fig. 12 presents the annual variation of both water temperature in the trapeze type of tank and solar fraction with respect to the collector area under the conditions of various types of soils. When half of the tank is embedded in sand and the other half in coarse gravel, the highest water temperatures and solar fractions can be reached. The lowest solar fractions and storage water temperatures are obtained if half of the tank is embedded in granite and other half of tank embedded in coarse gravel. In this type of thermal system, the most active parameters are collector area, the heat load and the climate. The effects of ground type for the long- term performance of the storage system are negligible. 6. Conclusions In the present study, a simulation model is developed for two types of central solar heating plants with the seasonal storage (CSHPSS). It is found that the solar fraction and cost increase with increasing collector area and storage volume for two types of seasonal storage. The simulation results have shown that solar fraction is greater than cylindrical storage system, when trapeze tank is used. It is found that the solar fraction value for Adana is obtained higher than other three different locations. It observed from the present results that the central solar heating plants with the seasonal storage in Turkey are promising both technologically and economically. However, these projects have never been any applied so far. References [1] Breger SD, Hubbell JE, Hasnaoui HE, Sunderland JE. Thermal energy storage in the ground: comparative analysis of heat transfer modeling using U-tubes and boreholes. Solar Energy 1996;56:49–503. [2] Chung M, Park J, Yoon H. Simulation of a central solar heating system with seasonal storage in Korea. Solar Energy 1998;64:163–78.

A. Ucar, M. Inalli / Renewable Energy 30 (2005) 1005–1019

1019

[3] Nordell B, Hellstro¨m G. High temperature solar heated seasonal storage system for low temperature heating of buildings. Solar Energy 2000;69:511–23. [4] Pahud D. Central solar heating plants with seasonal duct storage and short-term water storage: design guidelines obtained by dynamic system simulations. Solar Energy 2000;69:495–509. [5] Reuss M, Beck M, Mu¨ller JP. Design of a seasonal thermal energy storage in the ground. Solar Energy 1997; 59:247–57. [6] Argiriou AA. CSHPSS systems in Greece: test of simulation software and analysis of typical systems. Solar Energy 1997;60:159–70. ¨ nsal M. Analysis of solar aided heat pump systems with seasonal thermal energy storage in [7] Yumrutas¸ R, U surface tanks. Energy 2000;25:1231–43. ¨ nsal M. A computational model of a heat pump system with a hemispherical surface tank as [8] Yumrutas¸ R, U the ground heat source. Energy 2000;25:371–88. [9] General Directorate of Electrical Power Resources Survey and Development Administration (EIE). ¨ nsal M, Tanyıldızı V. A computational model of a domestic solar heating system with [10] I˙nallı M, U underground spherical thermal storage. Energy 1997;22:1163–72. [11] Duffie JA, Beckman WA. Solar engineering of thermal process. New York: Wiley; 1980. ¨ nsal M, Dog˘antan ZS. Solar Tables, Desing Data for Solar Aided Space Heating System. Middle East [12] U Technical University, Gaziantep Campus; 1980.