Series of numerical calculations have been carried out in order to analyse transient heat transfer during melting and solidification of the technical grade paraffin used in experimental investigations. Some of the computational results are illustrated further.

Fig. 6 shows radial dimensionless temperature distribution at an axial location of X = 12.73 in different dimensionless times during melting of PCM for laminar HTF flow with Re = 657. Melting of the PCM has occurred non-isothermally within a dimensionless temperature interval 0 to 0.42. The regions of HTF, tube wall and a PCM are indicated in the figure.

Melting of the paraffin starts on the HTF tube wall surface and expands inside the PCM storage tank. As melting front progresses, the temperature curve moves upward. The fluid velocity profile reaches a steady state quickly, while the temperature profile never reaches a fully developed state due to the moving melting front. This clearly approves that the use of empirical correlations for the convective heat transfer can result in a significant error.

Radial dimensionless temperature distribution at an axial location of X = 12.73 in different times during PCM solidification, for laminar HTF flow with Re = 657, is shown in Fig. 7. Solidification has occurred isothermally. Solidification fronts in different times are the intersections of the dimensionless temperature 0 = 0 and the corresponding temperature curves. Solidification of the PCM starts on the HTF tube wall surface and spreads inside PCM tank. Temperature curve moves downward due to a solidification front progression.

The propagation of the solidification front is shown in Fig. 8. The solidification front moves in the axial direction of the PCM container faster than in the radial direction. Due to the relatively large Prandtl numbers (small thermal conductivity) of the water as HTF, a large amount of heat is carried downstream, while a relatively small amount of heat is transferred directly to the paraffin as PCM upstream. The solidification zones are indicated in the figure. At dimensionless time t= 10909, the PCM is mainly in the solid phase.

Spatial dimensionless temperature distributions of HTF, tube wall and PCM inside the latent storage unit in different times during PCM solidification are shown in Fig. 9.

CONCLUSION

A numerical and experimental study of transient phase-change heat transfer during charging and discharging of the shell-and-tube latent thermal energy storage unit, with HTF circulating inside the tube and PCM filling the shell side, has been performed. Numerical predictions coincide quite well with the experimental results. Non-isothermal melting and isothermal solidification of technical grade paraffin, which has been used as PCM, have been observed by experimental investigations. Unsteady temperature distributions of HTF, tube wall and PCM have been calculated numerically. The results of numerical analysis underline that HTF velocities reach a steady state condition quickly, while temperatures change with moving of the melting/solidification interface, so it is necessary to treat the phase-change and fluid flow and heat transfer as a conjugate problem and solve them simultaneously as one domain. The usage of empirical correlations in expressing a convective heat transfer should be avoided. Due to the relatively large Prandtl numbers of the water as HTF, a large amount of heat is carried downstream, while a relatively small amount of heat is transferred directly to the paraffin as PCM upstream. The developed numerical procedure could be efficiently used for the simulation of transient thermal behaviour during charging and discharging of a latent thermal energy storage unit. Obtained numerical results provide guidelines for its design optimisation.

[1]

From the experimental data the receiver module efficiencies were derived based on incident flux measurements. Measured efficiencies ranged between 68% and 79%. The pressure drop through the receiver cluster was about 120 mbar, which corresponds well with the design data. The solar fractions reached up to 70%; the remaining contribution comes from fuel combustion. The SOLGATE solarized turbine efficiency, at nominal conditions of 230 kWe, was about 20%. The maximum receiver outlet temperature reached 960°C before stopping the tests due to the necessary gas turbine maintenance. Test data for the day with the highest air outlet temperature is given here as an example. About 45 heliostats were used at a solar irradiation of about 770 W/m2 . Air temperatures in the receiver modules were increased from about 300°C compressor discharge temperature to about 960°C in the high temperature receiver outlet. By mixing with the bypass air stream the air temperature dropped to 800°C which is the limit of the current combustor.

The gas turbine system performance for design conditions is shown in Fig. 6.

The electrical power production at the 960°C situation was about 190 kWe (with 130 kWe due to solar contribution). The estimated overall thermal efficiency of the receiver cluster was 77% for 800 °C and 70% for 950 °С.

Layout, Optimization and Performance Calculation

Modern computer based simulation models have been developed, adapted and validated to analyse the performance of solar-hybrid gas turbines in commercial system size,. The design of the optical part of the tower system (concentrator field arrangement and size, secondary acceptance angle, receiver aperture and orientation and tower height) can be cost-optimized using an adapted version of the HFLCAL code [5]. For the annual

performance calculation of the thermal power system the simulation environment TRNSYS with the model library STEC is used [7], [8].

Results for two industrial gas turbine systems as potential solar-hybrid prototype plants are presented here:

• Solar Mercury 50 — recuperated single shaft gas turbine. ISO rating 4.2 MW, thermal

efficiency 40.3%

• PGT 10 — gas turbine with bottom cycle. ISO rating 16.1 MW, thermal efficiency

44.6%

The solarization adds a receiver cluster directly before each combustion chamber for solar preheating of the compressed air. The receiver exit temperature under design conditions is 800°C or 1000°C. The receiver design temperature rules the maximum solar share. Data is presented for the site Daggett (California, USA, 34.9°N, annual DNI 2790 kWh/m2).More details will be published in [9]. Table 1 summarizes the cost-optimized layout of the prototype plants. Fig. 7 shows the layout of the solarized gas turbine plants. Each receiver zone consists of a group of single receiver modules connected in parallel. Receivers are subdivided into low-temperature (up to 600°C), medium-temperature (up to 800°C) and high-temperature modules (up to 1000°C). According to their temperature level, receiver zones are located in the low, medium and high flux region of the focal spot.

The annual performance of the prototype plants was calculated using a typical meteorological year on hourly basis. The results for day time operation are summarized in Table 2. The solar incremental electricity is the basis for the other incremental figures of merit. It is defined as the amount of net electricity produced by the solar-hybrid plant compared to the pure fossil reference plant using the same amount of fuel.

The incremental solar share varies according to the receiver inlet and outlet temperatures between 17.8% and 52.5% for daytime operation. The incremental solar to electric efficiency is the fair basis for comparison with other electricity generating technologies like PV. Values of 15% to 19% are reached with the prototype plants analyzed here. These

values will be even better when higher power levels with more efficient cycles are considered.

The performance of the collector field is affected basically by the physical properties of the absorber coating. In order to maximize solar gains and minimize heat losses, a high solar absorbtance and a low emissivity at working temperature is to be achieved.

Schematic structure and spectral reflectance of SCHOTT absorber coating

SCHOTT teamed with Fraunhofer ISE to develope a cermet absorber coating that is produced in a reactive sputtering process. The coating has a solar absorptance of 96% and an emittance at 400°C of 12%.

Tests

The first thermomechanical prototypes have been installed in June 2003 in the Eurotrough [4] at Plataforma Solar de Almeria, Spain. A field test of 100 thermomechanical prototypes has been started in Oct. 2003 at KJC, USA and is running up to now without failures. A new field test with 200 receivers will start at KJC in June 2004.

Outlook

In the follow-up project PARFOR the development activities are focused on field tests, further improvements for higher operating temperatures and cost reduction. This project is supported by the German Ministry for the Environment, project partners are the Flagsol GmbH, the German DLR and the Fraunhofer ISE.

[1]

Calculations were carried out with the simulation model Mantlsim developed at the Technical University of Denmark, [10], [3], [4] and [11]. The two tested low flow systems with the data given in Table 1 are taken into calculation. Weather data of the Danish Test Reference Year [12] is used in the calculations.

The daily hot water consumption is 4.6 kWh, corresponding to 100 l water heated from 10°C to 50°C. Hot water is tapped with an energy quantity of 1.53 kWh three times each day: 7 am, 12 am and 7 pm. The required hot water draw-off temperature is 50°C.

The thermal performance of the system with the two draw-off levels was calculated with one draw-off pipe at the very top of the tank and with different positions of the second draw-off level.

2.2.1 SIMULATION RESULTS

Fig. 3 shows the calculated yearly net utilized solar energy of the system as a function of the relative position of the second draw-off level and as a function of the auxiliary set point temperature.

Fig. 3. Calculated net utilized solar energy of the SDHW system as functions of the position of the second draw-off level and of the auxiliary set point temperature.

Fig. 4. Calculated extra percentage net utilized solar energy of the system by using two draw-off levels as functions of the position of the second draw-off level and of the auxiliary set point temperature.

Fig. 4 shows the calculated yearly percentage increase of the net utilized solar energy by using a second draw-off pipe as functions of the relative position of the second draw-off level and the set point temperature for the auxiliary energy system. If the relative position of the second draw-off level is at the top of the tank, the system is identical to the standard system with only one draw-off level at the very top of the tank. The thermal advantage of using a second draw-off pipe is strongly influenced by the auxiliary set point temperature. The higher the set point temperature the larger the advantage. If the auxiliary set point temperature is only 0.5 K higher than the required draw-off hot water temperature, the extra thermal performance by using a second draw-off level is about 1%. If the auxiliary set point temperature is 15 K higher than the required hot water temperature, the extra thermal performance by using a second draw-off level is about 13%. The second draw-off level is best placed in the middle of the tank.

Figs. 5 and 6 show the calculated yearly net utilized solar energy of the system and the calculated extra percentage net utilized solar energy of the system by using two draw-off levels as a function of the position of the second draw-off level for four different daily hot water consumptions: 50, 100, 160 and 180 l. Hot water is tapped by means of three daily draw-offs with the same draw-off volume: At 7 am, 12 am and 7 pm. Hot water is tapped at 45°C at 7 am and at 7 pm, while hot water is tapped at 40°C at 12 am. During all hours the top auxiliary volume is heated to 50.5°C. The draw-off temperatures are realistic, since hot water is not used at the same temperature level in practice. Further, in practice the set point of the auxiliary energy supply system is often 5-10 K higher than the required hot water draw-off temperature. The figures show that the net utilized solar energy is increased by about 6% by using two draw-off levels. Again, the second draw-off level is best placed in the middle of the tank. It should be mentioned, that there is a need for development of an advanced control system before solar tanks in practice can supply the consumers with different draw-off temperatures.

Fig. 5. Calculated yearly net utilized solar energy of the SDHW system as a function of the position of the second draw-off level for different daily hot water consumptions.

Fig. 6. Calculated extra percentage net utilized solar energy of the SDHW system by using two draw-off levels as a function of the position of the second draw-off level.

1.2 mol of 1, 2, 4-Triazole (Kanto Chemicals Co. Ltd., 98.0 %) was dissolved in 2 mol of piperidine (Kanto Chemicals Co. Ltd., 98.0 %). 0.4 mol of titanium tetraisopropoxide (Kanto Chemicals Co. Ltd., 97.0 %) and 0.4 mol of 1-amino-2-propanol (Kanto Chemicals Co. Ltd., 99.0 %) were mixed directly in a four necked round bottom flask equipped with thermometer, condenser and drying tube. 0.4 mol of Urea (Kanto Chemicals Co. Ltd., 99.0 %) was dissolved in 2 mol of N, N-Dimethylformamide (DMF, Kanto Chemicals Co. Ltd., 99.5 %). The solution was poured into the flask. Then, the mixture was heated by oil bath at 85 °C for 2 h. The solution was evaporated, so that corresponding TiOxNy precursor content became about 8 %.

Most collectors are equipped with special openings for enabling a controlled ventilation. Low — cost collectors or collectors with wooden frames are not really tight and allow considerable uncontrolled ventilation. Both contribute to changes of the micro-climate.

A procedure for measurement of the ventilation rate must take into account both. The working group Materials for Solar Thermal Collectors (MSTC) of the Solar Heating an Cooling Programme (SHCP) of the International Energy Agency (IEA) developed a method for Ventilation rate measurements and tested it by performing a round robin test of the same collector in several laboratories: IBE (DK), TNO (NL), SPF (CH), ISE (D). The method was to measure the air-flow-rate through an additional opening in the back-plane of the collectors while applying a positive or negative pressure difference in the order of some Pascal between collector and ambient. The results of the different labs are shown in figure 3. The good

agreement proved the applicability of the procedure. The variation of the ventilation rate of different commercial collectors is shown in figure 4. The ventilation rate was normalised by relating it to the volume of the collector in order to enable a comparison.

The function of pressure and flow rate follow the expected parabolic shape (figure 4a). The ventilation rate was defined as the flow-rate in collector-volumes per hour at a pressure — difference of 1 Pa. The collectors with the high flow-rates were untight wooden constructions. The values are usually not reached in operation. The flow-velocity measured by a micro­anemometer in front of a ventilation hole of a controlled ventilated collector during stagnation conditions (displayed in figure 5) follows the air temperature measured in the gap between absorber and glazing. The total air exchange during this sunny winter day was about 7 volumes per day. The air-exchange during night-time was nearly 50% of the day-time, with reversed direction, of course, because of the relative cooling in the clear nights. This might cause accumulation of moisture in the collector, when the collector temperature is below the dew-point and/or the insulation material was dried before.

Figure 4: Measurement of the ventilation rate of different collectors

Energetic behaviour of solar system based on facade-integrated collector has been investigated through a computer simulation. Simulation was aimed to characterise performance of facade solar system for hot water preparation in block of flats building and to obtain information on effect of facade collector on building performance.

Computer simulation was performed using Transient System Simulation Program — TRNSYS [6]. TRNSYS model for integrated facade collector was composed from a multizone model and the solar system model with thermal interconnection between them. Only a one zone in the central part of the considered building facade was modelled to investigate the block of flats building performance with facade collector. TRNSYS model used for simulations is shown in Figure 7. Facade construction was divided to two surfaces, one of them has been coupled to collector absorber (absorber temperature is identical with the last layer surface temperature).

Solar systems (facade, roof) were modelled as conventional ones — collector connected to storage tank with stratification. Facade solar collector with slope 90° was modelled as thermally coupled to building facade as described above. Roof solar collector was modelled separately with slope 45°. Thermal characteristics of the collectors were obtained from detailed simulation through KOLEKTOR model. Standard parameters of hot water were used (heating from 12 °C to 55 °C, max. temperature 85 °C). Solar systems have been modelled with high flow forced circulation (100 l/h/m2).

Two facade types were investigated for application of facade collector. First, middle-weight facade is common to panel block of flats buildings based on 27 cm ceramzit-concrete panels. These types of construction represent a wide range of buildings in large housing estates. Second, heavy-weight facade based on 45 cm brick represents the old buildings antecedent to the panel housing. Both types should be renovated with respect to construction problems, indoor comfort and energy consumption. In the model, applications with overall thermal resistance R =1,3 and 6 m2K/W for building envelope were considered. Windows with heat insulating glazing were used (U =1.7W/m2K). Overall surface area of zone facade is 9.0 m2, the window area is 3.2 m2, the wall is 6 m2. Splitting of the wall into two surfaces allows changing the collector / facade area ratio for parametric analysis.

Simulations were performed for different cases which can be considered in decision­
making for building renovation. Simulated cases were: panel wall — base case (envelope insulation with/without roof collector)

— integrated case (envelope insulation with facade collector) brick wall — base case (envelope insulation with/without roof collector)

— integrated case (envelope insulation with facade collector)

Parametric analysis for different facade construction resistances R, collector field surfaces Ac, required solar fractions and orientations were performed. Test reference year for Prague (TRY_Prague) was used as a climate database in the system and building simulations. Principal observed parameters for the building behaviour were energy consumption in winter season, overheating characteristics (inside temperatures, PPD values) in summer season and possible temperature-induced risk in construction. For the solar system, specific annual solar gains qs, u, solar fraction f system efficiency n and unutilizable energy qs, nu due to collector stagnation were the required outputs.

Results

Results of parametric simulations of solar fraction achieved by investigated solar systems (roof, facade) are shown in Figure 8. The solar fraction is plotted against the specific area of collector field Ac/Vaku. Interesting solar fraction values for facade collectors result from higher insulation levels (R = 3, 6 m2K/W). While the facade collector area should be increased by 30% compared to roof collector (45°) area to achieve 60 % solar fraction, for solar fraction above 70 %, the required facade collector area is comparable or lower.

However, roof collector gets to much more higher levels of stagnation conditions, which lead to possible operation problems and material degradation. Vertical position of the facade collector results in a well-balanced useful solar gains profile and very low level of unutilizable energy gains in comparison with the roof collector case.

In the Figure 8, the unutilizability factor bnu is introduced and plotted. Unutilizability factor is defined as a ratio of solar energy gains available from solar collector, but not used due to upper temperature limits (Tmax) in the storage tank, to total available solar energy gains Qs from collector field (solar system gains Qs, u utilized for water heating + unutilized Qs, nu).

b

Comparison of solar fraction f, specific solar system gains qs, u and solar system efficiency П, annual profiles for roof and facade solar system (R = 6 m2K/W) at annual solar fraction 60 and 70% is shown in Figure 9. Facade solar collector orientation impact on system gains and achievable solar fraction is shown in Figure 10 (compared with roof collector with south orientation).

Interaction of facade solar system with building has been investigated for winter (from October to April) and summer (from June to August) season. Performance analysis through collector-zone coupled modelling has been done for two usual building types, middle-weight (panel) and heavy-weight (brick). Results for winter season were put in the Table 1. With increasing heat insulation level, the heat gains caused by facade collector tends to be negligible.

South-oriented buildings often suffer with overheating problems in summer season. Temperature of the building envelope raises and considerable heat gains through the facade and windows could contribute to space overheating. Facade collector, due to good level of thermal insulation and absorber temperatures kept under 70 °C (low level of collector stagnation as resulted from system simulation), doesn’t cause notable temperature increase inside the building. Average temperature Tinside and PPD value (predicted percentage of dissatisfied) inside the zone adjacent to facade with collector were derived from frequency histograms for summer season and these are shown in Figure 11 in dependence on collector/facade area ratio (solar fraction 60 % and 70 %). Panel wall case, due to lower storage capability (middle-weight wall), results in higher average temperature values than brick wall case (heavy-weight wall). Application of facade collector thermally coupled to the wall moves the average temperatures no more than 1 K higher.

23.0

Fig. 11 Average temperature Tinside and PPD in the zone with facade collector in summer

season (June to August)

Figure 12 shows the temperature profiles in the facade collector-building construction (middle-weight ceramzit panel) during the typical summer day from 8 am to 8 pm. There are individual modelled layers outlined in the figure. The solar system heats the storage tank and during the day the absorber temperature raises up to 70 °C. Thermal insulation layers are affected by absorber, the first layer extremely. It should be made of materials capable to withstand high temperatures (up to 150 °C), e. g. mineral wool. Next layers (ceramzit panel) are kept at moderate temperatures with minimal variation during the day. These layers are mainly affected by the temperature variation inside the building. Brick wall case behaves in similar way with lower variations in indoor temperatures with respect to higher inertia.

The solar ventilation system described above consists of four main components: the slates, the PV module, the DC motor/fan combination and the duct. The flow rate in the system is governed by the interaction of the three latter components. Furthermore, since maximising the daily volume of air necessitates using a system with minimal resistance (i. e. smallest length of duct), the optimisation methodology depends primarily on the coupling between
the PV module and DC motor/fan combination. For a fixed length of duct, the maximum flow rate of air is obtained at the motor/fan maximum possible speed.

1.1 Photovoltaic module model

The I-V characteristic of the PV module can be described by the following equation, which is derived from the equivalent circuit described by Applebaum[3]

V + I ■ R S

I = I G — I о ( Є " — 1) (1)

where I is the PV module output current (A), IG is the light-generated current (A), I0 is the diode reverse saturation current (A), V is the PV module’s output voltage (V), RS is the series resistance of the solar cell (Q) and A is a curve fitting parameter (V).

This equation and the parameters included in it (i. e. IG, I0 and A) are valid at a given irradiance and temperature. Many methods are available so that the I-V characteristic of the PV module can be adapted to different levels of irradiance and module temperature. This work makes use of a new method which has been previously investigated by our research group at Napier. For a given irradiance (G, W/m2) and PV module temperature (Tmod, °C), the module short circuit current, its maximum power, its voltage and current at maximum power and its open circuit voltage are calculated. The parameters I0, A and IG are then calculated and substituted in Eq. 1 for an I-V characteristic which is valid at these conditions of G and Tmod.

Eduardo Rincon Mejia, Fidel Osorio Jaramillo and Fernando Vera Noguez

Facultad de Ingenierfa de la UAEMex
Cerro de Coatepec, C. U., 50130 Toluca, Mdxico
Tel.: + (52) (722) 214 08 55; Fax: + (52) (722) 215 45 12;

INTRODUCTION

This paper describes a new and simple multi-compound solar concentrator (MCC), which can be installed with easy in solar arrays or in single flat-plate solar collector if the absorbers can resist temperatures up to 140° C without damage, increasing very significantly its performance with a very modest investment. For a pair of flat collectors, the MCC consists essentially of two parabolic mirrors of the CPC type at the extreme sides of the collectors, and other pair of smaller elliptic fitted between them. At the bottom side of the array a flat mirror is placed.

This way, the following benefits are achieved: the gathering of solar energy increases due to the augmentation of the total area of aperture of the array, the stagnant and operation temperatures are both increased due to the concentration of solar radiation, the mean thermal efficiency is increased also, and the driving force for thermo-siphon convection is almost doubled. All the forgoing effects yield a very much high performance of the system, increasing its economical efficiency too. A variation of this MCC consists of a pair of flat mirrors (placed instead the parabolic ones of the first option) and a pair of parabolic mirrors (instead of the elliptic pair of first option) between the flat collectors. This option is simpler, cheaper and easy to implement, but the increase in performance is smaller. Nevertheless, it could be a very good election for inexpensive flat collectors.

In Mexico more than 100.000 m2 of solar flat collectors with metallic absorbers of copper or aluminum for water heating for domestic, industrial or services applications has been installed. These solar collectors are reliable, safe and very economically efficient; most of them costs less than 200 dollars for m2 installed, they do not need pumps due to the thermo-siphon effect and the storage and labor of installation is so cheap that the total investment is recovered in about three years or less, while their time of life are more than 10 years. However, their performance can be boosted if a simple and cheap multi-compound concentrator, tailored to the size of the available commercial flat collectors is implemented to the arrays. This can be made if the absorbers resist temperatures up to 140° C without damage.

It is expected that the present development would contribute to the massive the use of solar collectors in Mexico and other developing countries. The main limitation of these solar concentrators is that they cannot be implemented for many flat collectors with plastic absorbers, because they are degraded due to the high temperatures and UV deteriorative effects.

NOMENCLATURE

a Width of a solar flat collector

b = 2a + e Width of an array of two flat collectors

c1, c2, cn Adjusting coefficients for the function Ta (t)

Ac Aperture area of each solar collector

B Linear coefficient of thermal losses (W/m2 ° C)

C Quadratic coefficient of thermal losses (W/m2 ° C)2

Cg Geometric solar concentration (non-dimensional)

CPC Compound Parabolic Concentrator, placed at the extremes of the array

cec Compound Elliptic Concentrator, placed between two solar flat collectors

e Gap between two flat collectors

exc Eccentricity of the elliptic mirrors

F (t) Acceptation function

G Total irradiance (W/m2)

Gm Maximum irradiance in a day (in Mexico, about 1000 W/m2)

MCC Multi-compound solar concentrator

N Day length (yearly mean value: 12 hours)

Pu Thermal useful power (W)

Qu Gathered energy (useful heat) per m2 of flat collector (W/m2)

Qu Useful power (W)

t Instantaneous time (seconds)

ta Dawn instant

Ta Ambient temperature (° C)

Ti Fluid inlet temperature (° C)

Ta (t) Ambient temperature as a function of time

Greek symbols

Angle modifier function Efficiency variable (m2 ° C/W)

Thermal efficiency for flat collectors without MCC Thermal efficiency of flat collectors with MCC

Mean thermal efficiency in a given period Maximum thermal efficiency of flat collectors Angle of incidence of beam radiation Acceptance half-angle of the MCC Specular reflectance of the mirrors Angular parameter or coordinate Truncation angle (maximum value of t)

Together with an evaluation of the outcome of this project, a special aspect is the comparison of centralized and decentralized solar domestic hot water systems for terraced houses. Alongside the classical, self-sufficient ‘one-family house’ type of hot water central heating system, exists the possibility to connect the solar units of terraced houses together and to collect the heat in a central tank. The hot water is then directed to each individual house via a simple network of pipes. The production and distribution of heat for the heating system is also done via the central network.

The objective of the project, i. e. whether centralized systems for the combined supply of several houses with a solar collector area of 20-60 m2 can be an appropriate alternative to standard solar systems for each individual house was thoroughly investigated. The project offered optimal conditions for this investigation, as both system types were put directly

alongside each other so that both would be used under the same conditions (weather, orientation, demands made on the system).

2 Method

In addition to the technical and primary energy assessment of the systems with the help of a comprehensive measurement program and parallel computer simulation tests, a further important criterion of a hot water system, i. e. the economic implications involved, were also investigated. Not simply the initial investment, but the annual total costs with regard to the investment made, as well as the running costs incurred, were taken into account. As an alternative to the centralized freshwater storage system as used in Gelsenkirchen, the performance of other centralized systems on the market were also simulated, not just from the point of view of energy efficiency but in terms of economy. The simulation program used was MATLAB-Simulink®-Toolbox”CARNOT”/1/, a program specially developed by the Solar Institute Julich for researching conventional heating units and likewise innovative thermal solar systems. The results were generalised in order to be applicable to other solar settlements. However, the presence of influences which are not directly quantifiable should also be taken into account.

3 Results

As a result of the project, a research paper entitled „Advice on the Planning of Solar Hot Water Systems in Housing Settlements" was produced /2/. Following on from this, the most important findings both from the real-life comparisons made in Gelsenkirchen as well as the continuing generalized research done on the basis of computer simulations and economic analyses, were documented as a guideline for general use.