Local temperatures at the walls of the storage element are obtained from surface heat balances considering instantaneous solar irradiance gains {Gt — Gtrc/) on the front surface of the store, and heat losses coefficients of Ut and to account for thermal losses through

the front and the other sides of the store respectively. Non-slip conditions at the walls are assumed, i. e. и = v = 0.

From these assumptions, local temperatures at the front (x = L, y), back (x = 0, y), upper (x, y = L) and lower surfaces of the storage element result respectively from the

following expressions:

where T„ is the ambient temperature.

Numerical model

The governing equations and boundary conditions are converted to algebraic equations by means of finite-volume techniques with fully implicit temporal differentiation, using bi­dimensional Cartesian grids in a staggered arrangement. Diffusive terms are evaluated using central differences scheme. For convective terms, the SMART scheme [3] is imple­mented using a deferred correction approach.

The domain where the computations are performed and a schematic of the mesh adopted is shown in Fig. 1. The mesh is represented by the parameter n. According to Fig. 1 b, control volumes are used (for example, when it means that the problem is solved on

100*40 control volumes). The size of the control volumes is maintained constant throughout x-direction, while the mesh is intensified near the front and back walls using a tanh-like function with a concentration factor of 1.5, see [7], so as to properly solve the boundary layer. This aspect is indicated in the figure with two solid triangles. The simulated time is discretized using a constant time increment Д!

The resulting algebraic equation systems are solved segregately using a multigrid solver [5] in a pressure-based SIMPLE-like algorithm [6]. For each time step, the iterative conver­gence procedure is truncated once the maximum increment of the variables and equation residuals are bellow to 10^5 and the normalised heat imbalance is lower than 1%.

The cooling system, as already stated, has the purpose of lowering the temperature of the photovoltaic modules, when this becomes excessive, with consequent improvement of performances of the PV field. The system studied provides for the introduction of an adequate air flow, provided by a centrifugal fan from the outside air at its own temperature, in a cavity (15^2 cm gap) obtained by mounting "modular boxes” in aluminium on the back surface of the panel. During the summer the hot air is expelled again to the outside while, in the winter, it can be introduced to the rooms served by the PV plant to integrate with the traditional heating system.

Suitable connecting collectors (diffusers) to the main channel, for the introduction and withdrawal of air, which allow the use of just one fan also when there are more than one string, were studied. The diffusers can be realised in plastic material and are designed to channel the same air flow into each string. Fig. 6a shows the PV field together with details relative to the cooling system, while Fig. 6b shows the assembled PV field.

a) b)

Fig. 6 — PV field and details of the cooling system

Fig. 7 shows the comparison for July 2003 between the efficiency of the PV field, estimated with the Evans formula, with and without cooling system. As the figure shows, in the case when using the cooling system, the efficiency of the field is greater and displays a more even trend: this is due to the intervention of the fan activated by an on-off control system (using a differential thermostat) each time that the difference between the temperature at the back of the module and the outdoor temperature exceeds the At set. The maximum efficiency deviations, sometimes is greater than 31%, whereas the mean deviation is around 18%.

Fig. 8 instead shows the monthly electrical energy trend produced by the PV field with and without the cooling system. The overall energy supplied by the plant, for the 12 months studied, was 4143 kWh for the plant provided with the cooling system and 3441 kWh for the traditional plant. Taking into account the energy absorbed by the fan (112 kWh), the extra energy produced by the PV field is 590 kWh with an increase of 17,2% compared to that produced by the traditional plant.

Figure 9 shows the monthly electric energy absorbed by the fan.

The overall thermal energy released to the air following cooling of the modules, effectively usable and destined to winter heating, is 860 kWh.

The extra cost to be borne for the realisation and installation of the cooling system has been valuated at 37.00 €/m2, 4,8% of the overall plant value.

Considering an electrical energy cost of 15 €cent/kWh and a thermal energy cost of 8 €cent/kWh (natural gas), the simple recovery time to pay off the extra cost of the plant is

6,3 years.

CONCLUSIONS

Using the experimental data obtained from a photovoltaic plant, installed at the Mechanical Engineering Department of the University of Calabria, it has been demonstrated that the temperature increase of the modules significantly reduces the performances of the PV field.

A cooling system to be mounted at the back of photovoltaic modules was studied, which, using a suitable air flow conducted through a set of channels, provides for the attenuation of the effects of the raised temperature of the modules.

Using this system a mean increase has been calculated in the efficiency of the PV field, and therefore of the energy produced, at around 20%.

The economic analysis carried out showed that the simple payback-time to recoup the extra cost of the plant is about 6,3 years.

Laurentiu Fara1,

Romanian Solar Energy Society (SRES),
c/o Physics Department,

“Politehnica” University of Bucharest,
313, Spaiul Independentei, 060032,
Bucharest, Romania
e-mail: laurf@nare. renem. pub. ro

In stock raising the temperature of the water consumed by animals is from 15 to 20°C and the indoor air temperature is roughly 16°C. The air flows velocity for ventilation is roughly 3m/s. The lighting level is from 5 to 60 lx. The drinking water consumption of the personnel is roughly 80 litres/m2*day. These parameters could be reached during spring and autumn seasons using thermal or electrical energy supplied by solar systems with multiple functions.

This work took into account the specific results obtained by various authors for developing both solar thermal, PV/solar thermal and hybrid, which supply heat and electricity. The paper presents an energy and economic planning study concerning a solar hybrid system with multiple functions in stock raising, in the south-western part of Romania (Timisoara region). The system functions are indoor lighting, indoor air heating, air ventilation, water heating for animals’ use, water supply for farm staff.

Statistical experimental data regarding solar radiation and other meteorological parameters in the South-Western part of Romania.

The solar energy quantity which could be converted into other energy forms depends on:

• The number of clear (S) and cloudy days (N)

• Outside temperature (te)

• Solar radiation intensity (G)

The tilting angle towards the horizontal plane of the south-facing collecting surface, determines the component in the collecting plane of solar radiation intensity (GR). The quantities S, N, G, te, were measured hourly during four years by a team of the physics department of Timisoara Technical University. Daily, monthly and seasonal averages of these quantities were calculated. Solar radiation intensity was measured by an auto- compensating differential bolometer. This device is equipped with an equatorial which offers the possibility to measure solar radiation intensity on arbitrarily oriented planes. External temperature was measured with a normal Hg thermometer according to the meteorological norms. The statistical average values for spring and autumn months of the quantities S, N,Z=S+N, <te>m, <GR>m are presented below. The values of GR are given for the cases s=0 and s=60 grd. The quantity <te>seas shows the average value of the outside temperature during the two seasons. The quantity <GR>seas shows the average value of solar radiation during the two seasons.

The input data for SOLSTILL includes measured weather parameters and the measured temperature and relative humidity values of the air entering the still, as shown in Figure 5. The outputs of the simulation consist of the temperature of the water in the still, the cover temperature, the basin temperature, the temperature and relative humidity of the air
leaving the still, the temperature of the air leaving the condenser, the temperature and relative humidity of the air leaving the heat recovery (or the pre-heater), as well as the distillate production from the glass and from the condenser. Figures 6 and 7 show the measured and predicted values of the temperature and relative humidity of the air leaving the still and the pre-heater, respectively. Figure 8 shows the predicted and measured moisture content of the air leaving the still and Figure 9 shows the predicted and measured temperatures of the water in the still. Figures 10 and 11 show the predicted and measured distillate production from the glass and the condenser, respectively.

SHAPE * MERGEFORMAT

SHAPE * MERGEFORMAT

Figure 11. The predicted and measured distillate collected from the condenser of the still

in the open loop mode test.

As shown in figure 6, the simulation is able to predict both the trends in, and the actual values, for the temperature of the air leaving the still quite well. The maximum error is of the order of 5 0C and occurs in the early afternoon. The errors in the relative humidity calculation are more significant with errors ranging up to 20% RH with an average error of 10%. For most of the time, the calculated values are less than the measured values. The results in figure 8 confirm that the simulation program under-predicts the amount of water in the air leaving the still during the middle of the day. At other times, the measured and calculated values of the absolute humidity of the outlet air are very similar, which indicates that the differences observed in RH at these times are due to variations between the calculated and measured air temperature.

2. CONCLUSION

The development of SOLSTILL, a simulation program for estimating the performance of basin type solar stills was described. Models for both the standard free convection solar still and a forced convection solar still with enhanced heat recovery were included in the program. For the conventional free convection systems, the SOLSTILL program also enables simulation of more complex systems with many more parameters compared to the existing models found in the literature. A new model incorporating heat and mass transfer in forced convection solar stills with enhanced heat recovery was described in this paper. The design, fabrication and testing of an experimental system set up to validate SOLSTILL was detailed. The comparison of experimental and simulation results indicated that the program can predict distillate production at an acceptable level of accuracy.

After design and numerical simulation of the Advanced Solar Dryer (ASD) under study, according to design criteria and the numerical model presented in [2], a prototype with a 20 m2 evaporation area was constructed at INETI, for testing under real brine evaporation conditions:

Common greenhouse materials were used in the prototype construction, resting over a concrete base with its entrance aperture due south.

• evaporation pond has been set with a liner having 1.5 mm thickness: High Density Polyethylene film;

• greenhouse structure constructed in galvanized steel tubes;

• greenhouse covered with 0.2 mm thickness 3 layer thermal polyethylene film (PE + Thermal PE + EVA);

• solar chimney constructed with 0.8 m diameter galvanized tube, painted black.

A data acquisition system was installed for continuous monitoring of prototype performance. Measuring points and variables are indicated in the Fig.2.5.

Prototype testing started right after a brine solution was produced with the expected MED effluent saline concentration (56 g/l), on February 12th.

The simulation study is based on a single family house located at WUrzburg, Germany with a heated living area of 128 m2. The roof on which the collectors are mounted is facing south and is inclined by 45°. The building fulfils the requirements according to the German Energy Saving Directive (Energieeinsparverordung: EnEV) resulting in a space heating demand of 9090 kWh or 71 kWh/m2 respectively. The space heating loop is controlled according to the outdoor temperature and the maximum forward / return temperatures are 50/30 °C.

The heat demand for a hot water production of 200 litres per day at 45 °C equals 28 kWh/(m2a) or 3590 kWh/a respectively including the heat losses of a “typical” conventional domestic hot water store. Based on these assumptions the total heat demand for domestic hot water preparation and space heating comes up to 12680 kWh/a. If an oil or gas boiler with an average boiler efficiency of 85% is used, this results in an annual total energy demand of approximately 14900 kWh. The performance of the different systems is assessed on the basis of the fractional energy savings fsav. This quantity describes the percentage of energy that is saved by using a thermal solar system instead of a conventional, none solar heat generation system.

2 Results

By means of a simulation study the influence of different design parameters and store concepts on the fractional energy savings was determined for the reference conditions described above. The results for the so-called advanced storage concepts are shown in figure 1. Additionally two curves for a hot water store with a conventional thermal insulation are plotted as a reference in figure 1. An extensive simulation study about the influence of the hot water store volumes and different collector areas as well as collector types was already published in /1/.

Figure 1 shows, that by using latent or sorption heat stores with an effective store volume of approximately 1 m3 (plus 750 litres for the “conventional” combistore) and collector areas in the range of 45 m2 (flat plate collector) it is already possible to cover more than one half of the heat demand by solar energy. Due to the fact that the simulations were based on idealised assumptions, the store volumes may be twice as large in reality. One main reason for this is the volume of the necessary heat exchanger or condensate tank respectively.

At present latent and sorption heat stores for this kind of application are still in the scientific and industrial development phase. Therefore such stores are only offered by a few companies. However, regarding the long-term developments these technologies should not be ignored.

Compared with latent and sorption stores, hot water stores require larger store volumes. However, the main advantage of hot water stores is that they are based on a well known technology that is already introduced to the market. Due to this they also show at present the largest potential for cost reduction. With further improvements such as vacuum heat insulation an increase in thermal performance is still possible.

The Euronorm [ 1 ] proposes the Multilinear Regression (MLR) for the QDT. The efficiency of the collector is determined by a model containing 6 parameters (1). The six variables of the model have to be determined from the measured data. With the regression, the six collector coefficients can be determined. By minimizing the square errors of the differences between the calculated and the modeled efficiencies of all data pairs the coefficients are determined. As the uncertainties of the primary measurement data (radiation, flow rate, temperatures) join the calculation of the efficiency, a procedure would be useful, which considers appropriate weighting of a data pair in respect to the inherent uncertainty of that data pair.

For example, pyranometers with an offset of approximately 10 W/m2 show a higher relative measuring uncertainty in their lower measurement range than in their upper range. The weighted least square method (WLS) weights “measuring points by lower radiation (<300 W/mF)” less than for higher radiation for the regression process. Articles [ 5 ], [ 6 ] and [ 8 ] show that the MLR regression based on the least square method (LS) is not sufficient enough to determine the collector coefficients with their uncertainties for the SST. This paper shows the advantage of WLS for the QDT. In the following we discuss the different results obtained by the LS and the WLS for the QDT.

01

To reduce costs to a minimum level, making solar systems a viable domestic option, the materials, dimensions, and method of fabrication must be chosen with great care. Therefore, alternative methods of construction need to be looked into.

A typical flat plate collector is made from a number of parts. To simplify the construction and reduce costs an extruded polycarbonate structure is to be incorporated into the design of the flat plate collector. The purpose of the extruded triple walled polycarbonate construction is to substitute for both the glass cover and the absorber plate. Figure 3.1.1, shows a section of the extruded polycarbonate panel.

Both types of thermal collector configurations being looked at in this study are shown as a section in figure

3.1.2 and 3.1.3. Figure 3.1.2 shows the set up of the collector with a black — coated backing between the insulation and the polycarbonate. Figure 3.1.3, shows the collector with an aluminum backing (absorber) between the insulation and the polycarbonate.

Under steady state conditions the energy balance is used to describe the performance of a solar collector. The useful energy output of a collector of area Ac is the difference between the absorbed solar radiation and the thermal loss [1]:

Qu = A [5 — UL (Tpm — Ta)]

The overall heat loss from a solar collector, UL, consists of top heat loss through cover systems and back and edge heat loss through back and edge insulation of the collector can be expressed as:

Ul = Utop + UEdge + UBack (3’1’2

Since polycarbonate material is being used instead of glass, it is necessary to calculate UTop for a plastic glazing, in order to analyse the thermal performance of the solar collector. The net radiation method was used to calculate the top loss coefficient of plastic-covered solar flat-plate collectors. To evaluate the heat loss through the cover system, all of the convection and radiation heat transfer mechanisms between parallel plates and between the plate and the sky must be considered as shown in Figure 3.1.

Analysing each of the variable

Performing an energy balance on the cover, Utop can be easily obtained:

Q£ox? ftp ^"Qcw, to Qcovrfip Qc/Ntrfxt ^Qsoieypr ‘oer fib/т/

QLve’,dsflnf Ьр-с (p T)

Or* =Ш ~7I)

Qpjcss Q, dsthr ^Qovrjisficnr Qphafar

U Qfils-/ tp~ T _TJ

(3.1.11)

When evaluating the back heat loss, the thermal resistances from convection and radiation heat transfer are much smaller than that of conduction. Therefore, it can be assumed that all the thermal resistance from the back is due to the insulation. The back heat loss, Qb, can be obtained from:

Qb — kb Ac (Tpm-Ta)

Lb (3.1.12)

With the assumption of one-dimensional sideways heat flow around the perimeter of the collector, the edge losses can be estimated by

Qe — kg Ae (Tpm-Ta)

Le (3.1.13)

According to EN 12975-2, the following procedures can be applied for the determination of the effective collector capacity (the first three of these are alternative methods for the stationary collector test).

• Calculation from the physical data (with weighting factors for the components),

• determination from the response of the collector to a step-change of the fluid inlet temperature (Annex J.2),

• determination from the response of the collector to a step-change of the irradiance (Annex J.3),

• determination as a fit parameter of the quasi-dynamic collector test.

For details of the procedures see figure 2.

Unfortunately, the results of these procedures are not at all comparable. According to investigations at Fraunhofer ISE and at ISFH, the J.3 procedure leads to values that are
two to four and a half times as high as the calculated value. The most marked difference is found for vacuum-tube collectors of the dewar type ("Sydney” and similar collectors). According to ITW members, the results of the quasi-dynamic test agree quite well with those of the J.3 procedure. In a test at ISFH, the J.2 procedure gave a value that is even slightly below the calculated one.

These differences influence the thermal gains of the collector, which decrease with increasing capacity. For a dewar-type vacuum-tube collector, tested at ISFH, the calculated capacity was 9 kJ/m2K, while the result of the J.3 procedure was 40 kJ/m2K. The difference of calculated yearly collector yields1, caused by the difference of capacities, amounts to 9 %, which is quite remarkable.

Figure 1: Comparison of energetic balances in the 1-node and 2-node models.

The building examined is a small house, as shown in fig.1. It is made up of nine rooms dislocated on two floors with a heated area of 163 m2. The first floor is over a not heated basament, while the second floor is located beneath a not heated attic.

Radiant floor has been made by drowning a copper pipe with inner diameter of 12 mm in the slab, using a step of 15 cm. The heat transfer fluid, at low temperature and with a 0.112 kg/s flow rate, flows inside. Three localities of Italian territory have been considered: Cosenza, Rome and Milan. Table I shows for each city latitude, period of heating and outside mean air temperature in such period.

Table I — Latitude, days of house heating and outside mean temperature in the period of heating for the three cities.

The thermal calculation to respect the Law 10/91 has determined thickness of insulator (k = 0.045 W/mK) in dispersing external walls equal to 2 cm for Cosenza, 4 cm for Rome and 6 cm for Milan. The values of solar radiation and outside air temperature used in simulations, have been generated by the specific TRNSYS type 54 supplying hour values for each locality, starting from their correspondents monthly average data, so as to model a typical meteorological year (TMY) [4,5]. The values of beam and diffuse radiation on the horizontal plan have been obtained by the Reindl’s relations [6]. They use solar altitude angle and hourly clearness index as calculation parameters. The projection on tilted surface has been carried out by employing isotropic sky model [7]. Radiant floor has been simulated by a finite difference code [8,9] developed in TRNSYS environment dividing the floor into adiabatic segments, imposing inlet temperature to every segment equal to the outlet temperature from the previous segment. Such a code, linked to the Type 56 [10], describing the heated environment with its walls by separating valuation of inside and outside radiative and convective heat exchange, can analyse in detail the interaction between radiant floor and the heated environment. The type of control plays an important role in such heating systems characterized by great thermic inertia [11]. A control strategy has been chosen expecting inlet temperature to be variable in function of outside air temperature associated to a ON/OFF control on inside temperature in the dead band 19-21°C. Inlet temperature, which is generally different in each room, assures in absence of solar radiation an inside air temperature of 20 °C when in the adiacent rooms there is the same temperature. Inlet temperature assumes the following mathematical expression:

tiniet = k(20 — tae)+ 20 (1)

where the “k constant factor only depends on the mean heat loss coefficient of the room dispersing walls. To estimate domestic hot water requirement we have considered that each person consumes 50 liters of water at the temperature of 50 °C. We have considered the presence of 4 people and the temperature of water coming from the adduction circuit

of 10 °C, and we have determined in these conditions a 33.512 MJ daily energy requirement and a annual energy requrement of 12232 MJ. Tables II and III show, for the three localities considered, in the case of combined control, monthly and seasonal thermal energy requirement, monthly and seasonal mean inlet temperatures, calculated when the pump is on. Inlet temperature, between 23 °C and 31°C, shows that combined control in the radiant floor allows to use a low temperature fluid; moreover low temperatures supplying radiant floor permits to obtain, under the same conditions of tank’s storage volume, a greater usable stored energy.