Impact of compact solar domestic hot-water systems on
the peak demand of a utility grid in Brazil

Samuel L. Abreu1, Juan Pablo de L. C. Salazar1, Sergio Colle1
1LABSOLAR — Solar Energy Laboratory / Federal University of Santa Catarina —
Florianopolis — SC — Brazil
Wilson Reguse2

2Centrais Eletricas de Santa Catarina — CELESC (Santa Catarina state utility)
samuel@emc. ufsc. br, phone +55 4 8 331 9379, fax + 55 4 8 3317615

Introduction

A particular characteristic of the electric energy consumption in Brazil is the widespread use of electric showerheads and the resulting peak demand between 18h and 21 h. Over 90% of the residences in Brazil have electric showerheads. Studies have shown that electric showerheads represent approximately 23% of a household’s energy consumption and this fraction increases to around 35% of the total demand during the peak hours for low-income consumers (Prado and Gongalves, 1998). Electric showerheads are very cheap, usual prices lie under US$30 in Brazil, have a nominal power between 4kW and 8kW and are very efficient in terms of energy conversion. All these aspects contribute to the large scale use of electric showerheads for water heating among low-income consumers. Furthermore, showering is the only use of hot-water by this class of consumers in Brazil. Therefore, Compact Domestic Solar Hot-Water Systems — CDSHWS, cheaper and easy to install when compared to conventional solar hot-water systems, may be an economically attractive alternative to supply large scale hot-water usage, with the benefit of reducing the peak demand on the utility grid.

Utilities in Brazil are obliged by regulation laws to supply electric energy to low — income consumers. However, the associated costs of energy generation and distributions are heavily affected by the power of the electric showerheads, making investments in almost all cases economically unviable. Januzzi and Schipper (1991) estimate that the marginal expansion costs lie around US$ 1,500.00 per kilowatt. (falta ref para inserir). This scenario leads to the conclusion that the utility can account for the large scale installation of CSDHWS in its investment policy. In other words, the utility can share the cost of the solar heaters, which lies around US$ 300,00, with low — income electricity consumers.

In a previous work, Salazar et al. (2003) optimized seven parameters of a CSDHWS using peak demand and total cost as constraints. The optimized parameters were: collector aperture area, storage tank volume, heater power, electric showerhead power, set-point temperature of the storage tank, mixing valve temperature and collector slope. The chosen optimization procedure was successful, but the lack of information on hot-water consumption profiles is a limitation on the reliability of the predictions.

Colle et al. (2003) carried out the economical optimization of the CSDHWS storage tank insulation thickness. The optimization showed that life cycle cost savings are sensitive to insulation costs, when preheating of the storage tank during the morning early hours is required in order to avoid the expected peak demand. This optimization concern should be taken into account, in order to minimize the cost of CSDHWSs.

To study the effects of CSDHWSs on the peak demand, ninety low-income consumers from a housing unit were chosen to have their showerhead electric energy consumption monitored. Sixty consumers were equipped with CSDHWSs, while the remaining consumers served as a reference case. The electric energy
consumption of the showerheads was continuously measured, providing the profiles from which the results of this paper are derived.

The thermodynamic properties of the water are considered in this study. The velocity and is assumed to be zero at the beginning. The operation pressure in the tank is taken 1 atmosphere and the hot water temperature is assumed to be 320 K when the water usage started. The time between the water is started to be used and the water usage was finished is assumed as calculation region. This time is taken as 30 minute and the problem solved as unsteady.

2.2. The numerical method

The velocity and temperature distributions inside the tank have obtained by using the computer code [12]. Three dimensional unsteady solutions are taken by using implicit method in segregated solver. The discretization is achieved by using standard method for pressure and by using first upwind method for momentum, energy, turbulent kinetic energy and turbulent dissipation rate equations. The SIMPLE (semi-implicit method for pressure-linked equations) algorithm was used for flow calculations.

The convergence is very important in numerical calculation by using the code. Solutions were assumed to converge when the following convergence criterion is satisfied by every dependent
variable at every grid point in the computational domain

pnew — pold

¥

Where (^) in general could be any dependent variable. In this study, ^=10’3 for continuity, momentum. Turbulent kinetic energy and turbulent dissipation rate equations. This value is 10^6 or energy equations.

To under-relaxation factors are assumed as 0.3, 0.7, 1, 0.8, 0.8 and 1 for pressure, momentum, energy equations, turbulent kinetic energy (k), turbulent dissipation rate (s) and turbulent viscosity (pt) as, respectively. This factor has assumed as 1 for body force and density.

The CSDHWS works in a single phase thermosyphon mode, consisted of a single glazed flat-plate collector and a horizontal storage tank equipped with a resistor located immediately above the collector, as shown in Fig. 1. The system can be easily accommodated on the rooftop and integrated with existing piping (see Fig. 2). An additional electric showerhead, with continuously adjustable power, provides extra heat input. Therefore, auxiliary energy can be added to the system either in the storage tank or in the electric showerhead, but in the present analysis only the electric showerhead was used. The system is also equipped with a thermostatic mixing valve at the storage tank outlet pipe, which prevents scalding. The compact system is only used for showering purposes. Table 1 shows the technical characteristics of the system. The solar collector was tested according to European flat-plate collector test standards (Muller-Steinhagen, 2002).

hundred consumers from a housing unit for low-income families (monthly income from US$250.00 to US$500.00) were interviewed using a questionnaire based on a model suggested by Vine et al. (1986). Ninety families were selected according to the similarity with a standard hot-water consumption profile and sixty of them received a CSDHWS. The groups are named as follows:

(i) Group A — sixty consumers with CSDHWSs.

(ii) Group B — thirty consumers without CSDHWSs.

Fig. 2 shows the systems installed on the rooftops. The buildings were financed by Caixa Economica Federal (Federal Brazilian Savings Bank) under a leasing contract.

The power of the electric showerheads of all ninety consumers was averaged over five minute intervals. The consumers without the CSDWHS were used to characterize the typical energy consumption profile of a group of low-income consumers. The comparison between the two groups was used to estimate the solar fraction provided by the solar systems.

The cold water enters and exit pipes in radial direction are in the same axis as symmetrical condition. So, higher percentage of the cold water entering to the tank would be directed towards to cold water exit channel (T-i) from the tank. At the end, the cold water can not produce a lot of vortex inside the tank and destroying effects into thermal stratification inside the tank would be decreased. The hot water exit channel (T3) is generally located at the top of the tank. So, the hot water can be supplied as long as possible because of the positions of the hot water exit channel.

2. Discussion

The effect of the cold water velocity into the thermal stratification has been presented to obtain higher thermal stratification. So, the water velocity has varied between 0.1 to 1 m/s for finding optimum velocity in order to obtain better stratification. Figure from 2-1 to 2-12 represent the temperature distributions of the tank with and without the obstacle with different water velocities. This work is the initial work to investigate optimum velocity to try in experimental set­up.

In order to take the optimum thermal stratification, firstly smooth tank is considered as in figure 2-1 to 2-6 for different velocities. There is a little thermal stratification at the upper part of the tank in smooth tank but this temperature differences is not very high enough. The better thermal stratification has obtained in figure 2-5 has 0.8 m/s water velocity in smooth tank models. The hot water (T2) and cold water (T4) has contact in all axial cross-sectional area when they enter the tank. The rotations of the hot and cold water velocity vectors occur. The hot water stratification has been destroyed by cold water in this condition. In order to keep higher thermal stratification, the axial contact area between cold and hot water must be decreased and cold water mustn’t be directed towards upper part of the tank. Therefore, the obstacle is placed into the tank to decrease contact area.

The thermal stratification area and thickness are higher in tank with obstacle rather than tank without obstacle. T3 and T-i temperature distributions according to the tank model must be considered for choosing the tank model has better performance of thermal stratification. The T1 must be lower and T3 must be higher value as possible as to get better stratifications.

T-i and T3 values have presented in figure 3 for smooth tank for several water velocities. Both of these temperatures are increased with the increase of the water velocities. The differences of these two temperatures are 2 oC at 0.1 m/s velocity and 5 oC for 1.0 m/s.

SHAPE * MERGEFORMAT

Vk=0.8 m/s, V?=1 m/s, Si= 200 mm, di= 200 mm

Vk=1 m/s, V?=1 m/s, Si= 200 mm, di= 200 mm

Figure 2. z-r Temperature distribution inside the tank in the z-r plane

The temperature differences of T1 and T3 values are presented in figure 4 for tank with obstacle for different velocities. The temperature differences are also increased with the increase of the velocity. But, this increment is higher than the smooth tank. This difference is 27 oC at 0.1 m/s velocity and 34 oC for 1.0 m/s. The differences id increased from 7 to 13.5 times rather than smooth tank. This is proved that the use of the obstacle is necessary to get the higher thermal stratification.

The best velocity is found as the Vk=0.8 m/s for all investigated cases. The temperature differences between T1 and T3 is 4.887 K and 30.928 K for smooth and obstacle placed tank for Vk=0.8 m/s.

In figure 5, T1 and T3, T2 and T3, T1 and T4, T3 and T4, T2 and T1 temperature distribution are

presented for different velocities for smooth tank. The temperature differences are very small for this type as 2.5 oC. T1-T3 has increased with the increase of the water velocity. This is desirable thing for thermal stratification. If (T2-T3) is high this will help the increase of the performance of the heater and storage tank. This value is also decreased with the water velocity. (T1-T4) is desired to be as small as possible. This value is nearly 20^25 oC for smooth tank. This is not as small as desired. This value has also decreased with the water velocity.

The temperature difference (T3 and T4) is desired to be high. This value is nearly 22^23 oC for smooth tank as seen in figure 5. This value has increased with the water velocity increment. (T1 — T4) is also desired to have small values. This value is nearly 4^7 oC for smooth tank. This is good value for this condition. This value increased while the velocity increases.

The different temperature differences versus by water velocity has presented in figure 6. The very helpful parameter to supply high water to usage is the (T2-T1). This value is higher 1.5 to 2 times in obstacle placed tank rather than smooth tank. The difference between T3 and T4 is higher 1.5 times in obstacle placed tank than smooth tank. The higher value of this parameter is very important for solar energy heating systems. T3 and T1 temperature differences are also desired to be high for solar energy heat storage tank and evaluate the efficiency of the water heater by solar. This difference is 33 % higher in obstacle placed tank as seen in figure 6. This value is increased with the increase of the water velocity in obstacle placed tank and this value is decreased in smooth tank. The temperature difference between T2 and T3 is desired to be lower for obtaining the higher thermal stratification. This value is nearly 20 oC and 15 oC for smooth and obstacle placed tank, respectively. This is the advantage of the obstacle. The temperature differences of water between the water return to the storage tank and water

coming from the main line is important to improve heat efficiency of the storage tank. This value is desired to be as low as possible. This value is nearly 5 oC for obstacle placed tank. This value is also the same in smooth tank but it increases with water velocity. This increment is lower in obstacle placed tank.

320 318 ^316 ^314 >-312 310 308

0, 1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1

Vk [mis]

Figure 6. T3 and T1 temperature distribution for non obstacle tank

5. Conclusions

The numerical analysis of the effects of water flow rate into the thermal stratifications in the cylindrical hot water storage tank has been presented. The optimum water velocity value has been found to obtain higher thermal stratifications. The water velocity Vk=0.8 m/s is found to be better velocity to obtain higher thermal stratifications for the tank with and without obstacle.

T3-T1 temperature differences are important parameters to evaluate the thermal stratification. This value is very close as 3^5 oC for smooth pipe and 27^34 oC for tank with obstacle. As a result, the tank with obstacle has nearly 7^13.5 times higher thermal stratification compared to the smooth one. This value has increased with the increment of the water velocity for both types. T3 — T1 value has 4.887 and 30.928 K for smooth and obstacle placed type tank.

The temperature difference between T2-T3 is desired to be high to improve heat storage tank and heater performance. This value has decreased with the increment of water velocity.

T3-T4 is very important parameter to determine thermal stratification. This value has changed between 22^33 oC for smooth tank and between 33^39 oC for obstacle placed tank. This value is 33% higher in obstacle placed tank rather than smooth tank. This is the advantage of using obstacle in the tank. This value has increased with the increment of water velocity for both types.

T1-T4 is desired to be low. This value is nearly 20^25 C and 4^7 oC for the tank with and without obstacle, respectively. Using obstacle has improved this value. This value has not highly changed with both velocities.

T2-T1 is 1.5^2 times higher in obstacle placed tank. This value has not highly changed with the water velocity.

T2-T3 is desired to be low to obtain higher thermal stratification. This value is nearly 20 oC and 15 oC for the tank with and without obstacle, respectively. This is a very important advantage of using obstacle in the tank. This value has decreased with the increase of water velocity. This is undesirable condition for thermal stratification.

Acknowledgement

It is the pleasure of the authors to acknowledge TUBITAK for their sponsorship and collaboration with the University of the Notre Dame.

The impact of the use of CSDWHSs on the peak demand of a utility grid was determined for a group of low-income consumers. The energy consumption profiles of the groups are an average of the individual profiles; therefore they represent a variety of electric energy consumers.

The peak demand still remains, even with solar heating, but its use can reduce the peak by 60% on a monthly average basis of the hourly values. Using the power recorded in 5 minute interval averages, a reduction of 47% was achieved for the days where the highest values occurred. However, the obtained results were derived using only one month of data of a specific housing unit, therefore, results presented here are far from being conclusive. Data measurement will continue until one year is completed.

The estimated solar fraction for this period was 58%, but there is a high variability among the results for different families.

The ongoing research will take into account the determination of typical individual consumption profiles and the relative contribution of each profile on a group of low-income families. These consumption profiles will provide the basis for simulation of CSDHWSs conjugated to showerheads. The theoretical results will be validated against experimental data collected in the same housing unit. The validated results will be used to evaluate the impact of CSDHWSs on the peak demand of urban utility grids for other locations in Brazil.

An additional measure to further reduce peak demand on days of low solar irradiation may be electric preheating in the storage tank using weather forecast information. In this case, better thermal insulation of the storage tank as well as improved tank design may be necessary in order to maintain the stored water at acceptable temperature levels over long periods. The authors are presently investigating an algorithm that uses weather forecast information as input to predict the need of storage tank preheating.

Acknowledgements

The authors are indebted to CELESC for funding the present research, under contract FAPEU / P&D CELESC — ANEEL 032/435, and also to Caixa Economica Federal for authorizing the use of the housing unit for research purposes.

References

Colle, S., Glitz, K. L. Z., Salazar, J. P., and Abreu, S. L., Cost optimization of low-cost solar domestic hot water systems assisted by electric energy, Proceedings of the ISES — International Solar Energy Society 2003 Solar World Congress, Goteborg, Sweden, 2003

Jannuzzi G. M., and Schipper L. The structure of electricity demand in the Brazilian household Sector. Energy Policy (19) 879-891, 1991.

Muller-Steinhagen, H., Test report — Thermal performance of solar collector, Acc. To EN 12975 — 2. 2001, n. 02COL273, Institut fur Thermodynamik und Warmetechnik, 2002.

Prado R. T.A., and Gongalves O. M. Water heating through electric shower and energy demand. Energy and Buildings (29) 77-82, 1998.

Salazar, L. C. J. P., Abreu S. L., Borges, T. P. F., Colle, S. and Reguse W., Optimization of a compact solar domestic hot water system for low-income families with peak demand and total cost constraints, Proceedings of the ISES — International Solar Energy Society 2003 Solar World Congress, Goteborg, Sweden, 2003

Vine, E., Diamond R. and Szydlowski R., Domestic hot water consumption in four low — income apartment buildings. Energy 12, 459-467, 1986.

Mevlut ARSLAN1, Necdet ALTUNTOP1, Veysel OZCEYHAN1,

1 Dept. of Mechanical Engineering, Erciyes University, 38039 Kayseri, Turkey

Abstract

The numerical analysis of thermal stratification in the cylindrical hot water storage tank with using different kind of obstacles has been presented. The heat transfer is carried out by forced convections in the tank. To obtain higher thermal stratification inside the tank, optimum obstacle type has been determined between 30 different kinds of obstacles. The forced convection effect into the temperature distribution inside the tank has also obtained for different obstacle types. The temperature relations between water exit from the tank and, the water enter to the collector and the water return to the collector have presented according to obstacle type in the graphs. The temperature differences of these waters have also presented to determine thermal stratification according to the obstacle types and angles. The temperature distributions between tank with using obstacle and smooth tank have also been analyzed to compare to get high thermal stratification. At the results, Obstacle type 11 has been found that it has best thermal stratification inside the tank between all investigated cases in forced convection.

1. Introduction

Solar energy applications are widely used in industry. Water heating is one of the solar energy applications. Energy storage is much more important where the energy source is intermittent, such as solar energy. In solar energy storage tank, the temperature of the hot water in the tank stars to decrease while the use of the hot water from the tank starts because of the cold water entering from the main line. The hot water and cold water would mix each other and the hot water temperature will decrease. In present study, using different kinds of obstacles have been proposed to obtain high thermal stratification in the storage tank.

Some researcher presented some works about there fields as; ALIZADEH, has investigated the thermal behavior of a horizontal cylindrical storage tank both experimentally and numerically. He used one dimensional the Turbulent Mixing Model and Displacement Mixing Model in numerical calculations [1].

AL-NIMR has solved and presented some mathematical models to determine the effect of the different design parameters on the thermal stratification within the tank and the time required by tank to supply water within a specified outlet temperature [2].

MISRA has analyzed the thermal stratification both theoretically and experimentally in hot water storage tank for the thermo siphon effect in solar water heating systems. He has given the analytical expressions to obtain temperature distributions in the tank [3].

HELVA ad MOBARAK have investigated the effect of the amount of the hot water using into the temperature distribution in solar water heating storage tank [4].

HARIHARAN and BADRINARAYANA have analyzed the thermal stratifications numerically and experimentally in the hot water storage tank. They have observed that stratification improves with increasing AT and water flow rates [5].

Mo and MIYATAKE have carried out the transient numerical analysis for the thermal stratifications in the storage tanks. They have used turbulence model (k-s model). They have presented the effect of exchange cold water with hot water into the thermal stratifications [6].

YOO and KIM have presented analytical solution to model describing the charging process of stratified thermal storage tank with variable inlet temperature [7].

EAMES and NORTON have investigated the effect of the tank geometry into the thermal stratification for sensible heat storage for low Reynolds number. They have presented the effect of inlet and outlet port locations on store performance [8].

In present study, numerical analysis of using the different obstacles into the thermal stratification has been presented in the storage tank. The different kinds of obstacles are placed in the cylindrical tank to get the best performance for thermal stratifications inside the tank between all investigated cases. The water was used as fluid.

E. Erdila, M. Ilkanb, F. Egeliogluc, and L. B. Y. Aldabbagh c

Electrical Engineering Dept., Eastern Mediterranean University, Magusa, Mersin 10, TURKEY

School of Comp. and Tecnology., Eastern Mediterranean University, Magusa, Mersin 10, TURKEY

Mechanical Engineering Dept., Eastern Mediterranean University, Magusa, Mersin 10, TURKEY

INTRODUCTION

Solar thermal collectors are commercially widespread in many countries. Cyprus, an Eastern Mediterranean island, has relatively high rates of solar insolation throughout a year. Summer days are rather warm and prolonged and the winter is fairly mild. The daily average sunny periods range from 5.5 -12 hours, [1] through the seasons, are shown in Fig. 1. The daily average global radiation is minimum during the months of December and January, at about 2.3 kWh/m2 (Fig. 2). The maximum is about

7.2

kWh/m2 in the months of June and July. Annual averages of daily sunny periods

Fig. 2 Daily sunny hours for each month (1981-1991)

and daily global radiation are about 9 hours and 5 kWh/m2 respectively.

Solar energy has been widely utilized in Cyprus for at least the past 40 years, [2]. People have enjoyed using solar energy for domestic water heating at no fuel cost. The average daily energy consumption of a typical household in N. Cyprus is about 7 kWh,

[3] as shown in Fig. 3. This amount of energy can be produced in electrical form, with a 1600 Watt PV system occupying an area of about 10 m2, [4].

The photovoltaic (PV) modules available in markets are now well developed. The trend is to generate even more power per m2 at a lower cost. The increase in demand and lowering of prices makes them even more attractive and relatively large systems are now installed at moderate prices. The growing worldwide awareness of the benefits of using clean energy has greatly increased the sales of PV systems by many folds in the last 10 years, [5].

C* Fax No.: +90392 3653715, loay. aldabagh@emu. edu. tr

Generally, depending on the type of solar cell used, some of the 70-95% of the collected energy is not converted to electricity. Hence, large area solar modules absorb considerable amounts of solar radiation that also generate excessive heat. This excessive heat leads to lowering of module efficiency, [6, 7]. Its found that the open — circuit voltages decreases by about 0.4% per centigrade (o C) for silicon cell [8], and hence the fill factor also decreases with increasing temperature. Many systems have been designed to remove this excessive heat and to utilize it as a source of energy for some other applications. This however requires some modification of the module structure to allow circulating a cooling medium with an aim to extract the excessive heat. Usuallly, the heat transfer fluid is either air or water. Use of water requires more extensive modifications to enable water-tight and corrosion free construction.

Such a hybrid module may combine the electricity generating function of a photovoltaic module with that of a hot-water collector, a system for space heating and any other low-temperature process. Performance of many different kinds of hybrid systems is extensively studied by many researchers [7, 9, 10, 11].

Generally, the cooling operation is applied at the rare of the solar modules. However, extraction of heat at the rare of the solar module may not be very efficient due to the often used reflective layer at the rare of the module. Such systems, require a complex design and fixures at the rare, where mounting frames are usually located.

The hybrid module we are studying is a variant of conventional systems with the cooling operation being applied at front cover of the solar module (Fig. 4). The photovoltaic modules are equipped with a transparent window at the front surface. The heat extracting fluid is circulated through a small gap between the front plate of the module and the attached window, by solar pumping. This arrangement may not provide an optimum method of cooling, but it is simple and practical to apply even to

I

Qin Qout

Fig. 4 Structure of the hybrid system.

an existing installation of modules. The attached window and the water-tight seal can easily be removed if and when not required, without replacing the module and its fixtures. The window at the front surface, off course, negatively effect the electrical conversion efficiency by absorbing/reflecting some of the incident radiation.

The cross-sectional view of the used mechanical model for the thermal stratifications for numerical analysis is shown in Fig. 1.

Boundary layer equations were used to determine the temperature and velocity distributions in the flow field. The analysis was based on the unsteady, three-dimensional continuity, momentum and energy equations.

The assumptions are used in the present study as:

— The flow is unsteady, turbulent and three-dimensional,

— The thermal conductivity of the tube sheet material does not change with temperature

— The tube material is homogeneous and isotropic.

Three dimensional continuity, momentum and energy equations are solved numerically. The upwind and central difference method used for convections and diffusions, respectively [9].

The hot water temperature is desired to keep constant as long as possible in the upper part of the tank. Therefore, different obstacles are placed into the tank in order to supply higher thermal stratifications. The schematic view of the obstacle type and geometries are shown in figure 2. The details of these geometries are shown in figure 3. Obstacle types and tank matches are

The water temperatures are assumed 333, 320 and 285 for water exit from the tank to usage, water in the tank and water coming from the main lines, respectively.

V.

The hybrid system consisted of a PV module, where a glass plate window (jacket) was layed over the front surface, a circulating pump and a heat exchanger. The glass plate was temporarily attached and could be removed at any time without much effort. A fluid, distilled water, was circulated between the front surface of the PV module and the jacket layed on the front surface.

The incident solar radiation first passed through the glass jacket and the layer of circulating water before reaching the module. The high transmittance of glass and water layers for the visible radiation, ensured that the silicone cells of the PV module would receive most of the visible radiation to generate electricity without much sacrifice in power. Infra-red part of the incident radiation is essentially absorbed by the glass-water and module structure combination leading to warming of the solar cells. It is this excess heat that the hybrid unit is designed to remove, by circulating water, and utilize in domestic applications. While the circulating water removes the heat and leads to cooling of the module, at the same time it provides a source of warmed water for preheating applications in a domestic use.

A standard commercial module, M, rated at 55 Watt peak, with dimensions 130x47x5 cm was used. The jacket, J, had the diamensions of 130x47x1.5 cm (Fig. 4) and was placed over the module surface with a gap, about 0.4 cm to allow flow of water. The PV module was tilted 45 degrees to horizontal and faced south for maximum solar insolation, I. Initially pure water, Q, was circulated at the rate of 36 liters per hour. An identical PV module was used as a control and measurements were performed on both the hybrid and the control modules at the same time and location.

A data acquisition card was used to record a range of data. Meteorological information published by the Statistics office[1] was used. The electrical characteristics of both modules and thermal performance of the hybrid system were measured. The data listed below were recorded at various times from 9.00-17.00 hrs. each day and the average values were plotted. The temperature of the circulated water, at input/output Tin, Tout, the ambient temperature, Tamb, and surface temperatures, Tsc, Tsh, on the control and hybrid modules respectively were recorded. All temperatures were recorded in degrees centigrade. The short-circuit currents, Iscc, Isch and open-circuit voltages, Voc, Voh of both control and hybrid modules respectively were recorded. The maximum operating power of each module is defined as P = Voc Iscc FF (1)

The fill factor, FF, was taken as 0.7, which is typical of single crystal silicone cells.

The electrical energy, Ec and Eh, of control and hybrid module respectively were calculated using numerical integration. Hence, the area under the power-time curve, was taken at various one-hour periods.

Thermal energy collected by the circulating water in one-hour period, can be defined as Qw = Q Cp AT (2)

where Q is the volune flow rate of the water in lt/hr, Cp is the specific heat at constant pressure (4187J/kg. °C), and AT is the temperature difference (Tout — Tin ).

The obstacle types, have supplied higher thermal stratification, have been determined numerically by using computer simulation. In this situation, the aim is supply hot water as long times as possible with high thermal stratification between the temperatures T4 (cold water enter to the tank from main lines) and T2 (hot water supplied by solar collector). Different kinds of obstacle as in figure 3 with different assembly as in figure 2 have been placed in the tank in order to obtain this higher thermal stratification. These obstacles are placed in the tank with suitable angles and coordinates to decrease contact area between cold and hot water in the tank. T4 flow fibers don’t have to destroy hot water gradient. However, the hot water entering the tank (T2) would destroy the stratification. Therefore, hot water velocity is taken as 0.2 m/s for forced circulation. The flow rate of the hot water from tank to usage must be equal to the cold water flow rate entering to the tank. Therefore, the cold water velocity is assumed as 1 m/s.

Vortexes have occurred in the tank because of the mixing of two different fluids. These two fluids would hit each other towards to the wall and obstacle surfaces. Therefore, calculations are carried out by using wall functions method and standard k — s turbulent models with FLUENT [10] code.

The cold water enters and exit pipes in radial direction are in the same axis as symmetrical condition. So, higher percentage of the cold water entering to the tank would be directed towards to cold water exit channel (T1) from the tank. At the end, the cold water can not produce a lot of vortex inside the tank and destroying effects into thermal stratification inside the tank would be decreased. The hot water exit channel (T3) is generally located at the top of the tank. So, the hot water can be supplied as long as possible because of the positions of the hot water exit channel.