An important yardstick for renewable energy technology evaluation is the cost of one ton of the fuel carbon (dioxide) avoidance by the specific technology. However, there is a fundamental dilemma here: How to compute the avoided carbon amount per one kWhe that need not be generated from fossil sources. Is it by assuming, say, a 40% efficiency conversion of the fuel (heat of combustion) to electricity? Why not 55%. Perhaps 30%? Of course, any assumption will produce a different result for the avoided carbon amount. The problem requires an agreed reference efficiency or standard. In case we find that we need more than one standard, this should be well reasoned. Whatever the standard applied, it should always be transparent.

Fig. 1, drawn after the diagram pattern of Geyer [2] illustrates the Fuel Consumption Ratios (FCR) for model thermal power-systems of various conversion efficiencies by the thin solid lines with varying slopes. In the figure the 60% conversion Combined Cycle (CC) is taken as a baseline standard for reference, thus having an FCR of 1 (shown by the vertical scale on the right) when the power plant has operated for the full 8760 hours of the year (full yearly

SHAPE * MERGEFORMAT

kHrs

Fig. 1. The relative fuel consumption (hence emissions) ratio and the
green energy fraction of systems over full load hours.

9 30% fuel in SEGS solar hours.

*50% fuel in SEGS solar hours.

load). The 60% CC baseline seems to serve as the recommended standard reference for large power plants at sites where natural gas may be available. The CC is a practical, efficient power plant of available technology at present. One of the less efficient generators is represented by the 30% conversion line, which shows an FCR of 2 (the inverse efficiency — ratio with respect to the 60% standard) for the full year operation. It reflects the relative excess fuel consumption (hence emissions) of the 30% system as compared to the 60% (baseline standard). As the amount of fuel (in terms of the heat of combustion) is 1.667 kWht (per 1 kWhe electricity) for the 60% conversion, it will double to 3.333 kWht for the 30% conversion. The avoided carbon in weight (grams) for each case is derived through the FCR value and the particular fuel stoichiometry (chemical composition-based accounting) of the respective fuel. It is of significance to recognise that the avoided carbon amount per one kWhe (green electricity) output is not explicit without a clear decision on the baseline or reference standard [3]. By setting the reference standard, we can resolve the above — mentioned problem of how to compute the avoided carbon amount per one kWhe electrical energy output. This concept is an essential key point for analyzing environmental parameters of hybrid systems. Green Energy Fraction (GREF)

For achieving the goals of climate protection, many kinds of solar energy systems have been developed and a variety of design strategies have emerged to enhance marketing prospects. The blending of fuel with solar has been invoked in order to enhance capacity, smooth power production and exhibit reduced costs for the mixed electricity output. The value of (1-fCr) defines the Green Energy Fraction (GREF) (shown on the left vertical scale) which represents the avoided-fuel fraction of the energy output of a system. It signifies the green (CO2-free) portion of the system energy output. This is the clean or green energy, which explicitly contributes to climate protection, which we aim at. We want this green energy not to be lost upon blending with fuel energy. Both the (GREF) and (1-FCR) of a system must be based on the same reference standard.

In system sizing, the designer goal is to maximise reliability and minimise installation cost. The Loss Of Load Hours (LOLH) is a suitable parameter to assess the reliability: LOLH represents the number of hours per year whenever the inverter is switched off by the charge regulator, in order to allow the charge of batteries by the PV modules. High reliability and low installation cost represent two opposed requirements, hence a trade off is needed, in particular about PV power and battery capacity.

Energy management is a logical algorithm which determines the energy flow in the system from Pv generator to load in order to use energy as efficiently as possible. The State of Charge (SOC) of batteries plays a key role for the energy management [3], [4]. To find a true optimal solution, it is necessary to take into account both system design (component sizing) and control strategy (proper set points) [5]. In the system under study, the charge regulator implements an operation with two set points at p = 30% and 100% for the SOC. Really the charge regulator measures the battery voltage as an indicator of SOC. A hysteresis band between 30% (corresponding to 22.2 V) and 50% (25 V) is allowed at the aim of charging partly the batteries, while around 100% (28.2 V) an equalisation charge by Pulse Width Modulation (PWM) current is included. In such a way, the periodic boost charge stirs the electrolyte, levels the cell voltages and completes the chemical reactions.

Since 1999 development work has taken place with an aim to integrate a large amount of PV (30 MWp or 300.000 m2 PV panels) in a whole city, Valby in Copenhagen to cover 15% of the yearly electricity demand by solar energy in the year 2025. . (See also www. solivalby. dk where a visualisation of where PV can be integrated is shown).

Key actors in this activity have been the Urban Renewal Copenhagen Company, the Copenhagen Energy electricity utility, the energy specialist company Cenergia and the municipality of Copenhagen.

In figure 1 there is an illustration of the size of this PV area and how the use of PV fits into an overall green accounting activity for Valby and Copenhagen. In figure 2 there is an illustration from Copenhagen Energy on the electricity peak shawing effect that only 15 MWp will have on the electricity use in Valby in June.

Due to the widespread use of CHP based electricity production in Denmark PV has a very positive effect because you can limit the operation of the CHP plants in the summer where the need for heat is low. In figure 3 there is shown examples of already realised PV projects in Valby and Copenhagen.

The main idea of working with PV integration for a whole city is to ensure good PV integration solutions so there will be developed a public support for the widespread use of PV instead of an opposition due to bad architectural solutions.

Another interesting aspect of this approach is that PV can be introduced as part of an overall sustainable building solution also aiming at general building qualities and energy savings.

In connection to this an activity of developing and exhibiting a PV assisted CO2 neutral test house in Valby was seen as a very relevant activity.

The modeling of pellets stoves, burner and boiler has been realized with a new TRNSYS — component type 210 that, unlike other models, takes the dynamic behavior during the start and stop phase into account. Type 210 has been developed for pellet burners and pellet stoves and can be even used for boilers where no internal hot water preparation needs to be modeled. The model is calculating the fuel consumption, combustion air flow and the exhaust gas flow and provides data for the delivered energy to the ambient, the heating circuit connected to the water jacket and to the exhaust gas. The model calculates also the air leakage losses when the burner is not in operation, the number of starts and stops and the consumed electricity.

The model separates the pellet heater into two main thermal masses. Mi represents the part of the stove that transfers the heat to the ambient and m2 representing the water jacket of the stove/boiler. Fuel and combustion air entering the stove, combust and form a combustion gas that transfers heat first heat to m-i and then to m2 before leaving the stove. Heat transfer coefficients define the heat transfer between the hot air mass flow, the thermal masses, the ambient air and the fluid in the water jacket.

For the two stoves and the integrated burner a verification has confirmed the correctness of the identified parameters. Simulation tests with parameter for the pellet boiler showed good energetic accordance to the measured data, but no exact verification has been performed thus this boiler is being considered as a generic boiler.

System 3 uses type 210 as a pellet burner, where all heat except the heat losses from the burner itself is transferred to the flue gas before entering the air to liquid heat exchanger of the combistore. Consequently the total flue gas losses for this system need to be calculated separately using the exhaust gas temperature of the air to liquid heat exchanger outlet.

Qflto, = mfl ■ CPfl ■ (Tohxb — TrooJ (1 )

where m’fl is during the combustion phase calculated by type 210 and after the burner is out of operation by equation 4 using the temperature difference between the gas leaving the heat exchanger in the store and the outdoor temperature. The leakage losses have been determined by:

Qleak = | m fl ‘ Cp fl ‘ (Tihxb Tohxb ) I for Tohxb >Tihxb (2).

A modification has been accomplished for the model of system 3 where the factory settings for the placement of the temperature sensor of the pellet burner have been adapted. The simulations showed that the sensor was placed too far in the top of the store causing a delayed start of the burner and thus giving in the meanwhile the electrical heater the possibility to heat up the store. To prevent the electrical heater turning on during normal operation the sensor was placed lower and also the set temperature of the electrical heater was reduced to 55 °C, the same value as for system 4. Moreover the maximum heating power has been simulated with the lower summer adjustment (12 kW).

In the model of system 4 the control settings for the boiler pump have been changed so that almost all the heat of the boilers water volume is transferred to the combistore once it is heated.

For the simulation a one zone building model has been used, which implies that the results for system 1 are only relevant if good heat distribution can be achieved from the stove to the whole building. The losses from the store and, in the case of systems 3 and 4, also from the boiler, are not used as heat input to the building model. The domestic hot water load has been modeled with the load profile developed by Jordan and Vajen (2002) assuming a daily hot water demand of 200 liter. All four DHW-stores and combistores have been model with TRNSYS type 140 (Druck and Pauschinger, 1996).

a) Solar air-conditioners

Today, in Serbia and Montenegro there are about 300,000 air/conditioners installed, each of an average power of 3 kW. That means that total electric power of these devices is about 900 MW! Thus, this is relatively largest market, with certain perspective for further expansion.

Solarisation of air-conditioning devices has an advantage base on the fact that requirements for space air conditioning coincide with insolation intensity. Disadvantage lies in large space requirements for solar modules (20-30 m2), which make their installation difficult and the same time increase the price of such solar air-conditioners.

Therefore an estimate of the dynamics of implementation of such systems at Yugoslavian market starts from moderate 44 kW during the first observed year (about fifteen demonstration systems). With anticipated drop in prices of solar cells and increase of requirements for air-conditioning during summer, and with announced high increase of prices of electric power, installed power of solar air conditioning devices would be rapidly increased (110 kW in 2007., 225 kW in 2008., etc.) (Table 4a.).

According to the data from the above table, in two last observed years (2009 and 2010.) installed power of solar air-conditioners would be stabilized at an annual level of about 350 kW.

b) Solar refrigerating plants

For application of solar cells in Balkans region, the following is interesting: mini refrigerating plants for freezing of meat and easily perishable goods in agriculture, as well
as refrigerators-trailers of lower carrying capacity for transport of frozen products and soft drinks.

Due to large number of mini refrigerating plants in Serbia that have not been statistical included (which are irreplaceable for numerous public feasts in the open), in further estimate we shall assume their total power of about 60 MW!

As an average power of a mini-refrigeration plant we shall adopt 2 kW. Unlike solar air — conditioner requiring large space for solar modules, this problem can be simple solved with refrigerating plants by optimum installation of movable plates of solar cells on flat roofs of mini chambers for freezing.

Because of this advantage and the fact that mini refrigerating plants are means for making profit, in further analysis it is assumed that these will have larger share at the market than air-conditioners (Table 4b).

PV-NORD is a Combined RTD Northern Dimension project supported by the European Commission, DG Energy & Transport, under the Fifth Framework Programme, thematic programme Energy, Environment and Sustainable Development. PV-NORD is the first EU-funded research project ever to focus on the Northern Dimension (the Nordic and Baltic countries, Poland and so on). Sixteen partners from five member countries share a budget of 2.8 million euro. The EU finances the project with app. 1.1 million. The project will run for three years and started on January 1st, 2002. NCC, a Swedish construction company co­ordinates the work. Please visit the project website www. pvnord. org.

Wilson Rickerson

Center for Energy & Environmental Policy
University of Delaware
Newark, Delaware, USA

Performance contracting, or contract energy management as it is called in the UK, refers to the practice of financing energy services based on the savings stream that those energy services are projected to generate. While performance contracts have not traditionally incorporated renewable energy systems (Goldman et al., 2002), a number of recent contracts have incorporated photovoltaic (PV) systems as part of an overall building energy service strategy. This paper presents the results of a survey of US PV performance contracts and describes how, when examined in aggregate, PV performance contracts represent an important share of new US installations. This paper will also report on the diffusion of the PV performance contracting model from the US Federal government to other sectors.

Methodology

Three sources of information were used in preparing this paper: (1) a review of the literature on performance contracting, solar energy, and energy service companies[44] (ESCOs); (2) a review of PV performance contract project case studies posted on the Internet; (3) telephone interviews with government and industry stakeholders.

PV performance contracting has emerged only in the last four years, and therefore little published literature exists. A few documents released by the Renewable Energy Policy Project (REPP) discuss the theory of bundling PV with energy efficiency in performance contracts (Stronberg & Singh, 1998; Eckhart, 1999; Singh 2001), but they do not address the subject empirically. There is a considerable amount of information available from the Federal Energy Management Program (FEMP), the US agency charged with improving the energy performance of Federal buildings, but PV performance contracting activity at the state, municipal, or private levels is not addressed

To determine the extent of PV performance contracting in the US, 32 interviews were conducted targeting Federal, state, and municipal governments, 8 ESCOs, and 3 solar contractors. Data for 18 projects completed in the period 2000-2003 were gathered. Follow­up calls were made to the ESCOs, solar installers, and government officials to verify that the data were complete and accurate. The nameplate capacities of these projects were then aggregated by year and compared to figures on total annual grid-tied Us installations through 2003 (Maycock, 2004).

The thermal properties of air make it far less efficient as a coolant medium than water [24]. This implies that more parasitic power will be needed to achieve the same cooling performance. Hence, air is a less favourable option in many cases. Detailed information on the design of forced air heat sinks can be found in [24].

The microchannel heat sink is a concept well suited to many electronic applications because of its ability to remove a large amount of heat from a small area. Tuckerman and Pease [26] were the pioneers who first suggested the microchannel heat sink, based on the fact that the convective heat transfer coefficient scales inversely with the channel width. Two major drawbacks to the microchannel heat sink are a large temperature gradient in the streamwise direction and a significant pressure drop that leads to high pumping power requirements. A numerical optimization that minimises the thermal resistance subject to a specified pumping power is presented by Ryu et al. [27]. Harms et al. [28] conclude that heat transfer performance in microchannels can be increased by decreasing the channel width and increasing the channel depth. Developing laminar flow is
found to perform better than turbulent flow due to the larger pressure drop associated with turbulent flow. Owhaib and Palm [29] show that in the laminar flow regime, the heat transfer coefficient is largely independent of channel diameter, while in the turbulent regime, smaller channels are clearly better. Introducing alternating flow directions can reduce the streamwise temperature gradient in the microchannel heat sink. Missagia and Walpole [30] describe a single layer counter flow technique. Vafai and Zhu [31] suggest using two layers of counter-flow microchannels, which is shown to significantly lower the streamwise temperature gradient compared to a one-layer structure. Chong et al. [32] optimised the counter flow principle for single and double layer channels for both designs for laminar and turbulent flows, and found that laminar flow was to be preferred over turbulent for both cases. The manifold microchannel heat sink (Figure 5), in which the coolant flows through alternating inlet and outlet manifolds in a direction normal to the heat sink, has been modelled and optimised by Ryu et al. [33]. Because the fluid spends a relatively short time in contact with the base, a more uniform temperature distribution across the surface is achieved.

Very low thermal resistances can be achieved through the use of impinging liquid jets. The impinging jets are capable of extracting a large amount of heat because of the very thin thermal boundary layer that is formed in the stagnation zone directly under the impingement, and that extends radially outwards from the jet. However, the heat transfer coefficient decreases rapidly with distance from the jet. To cool larger surfaces, it is therefore desirable to use an array of jets. If measures are taken to ensure the flow from different jets does not interact in such a way that they lower the overall heat transfer, impinging jets are predicted to be a superior alternative to microchannel cooling for target dimensions larger than the order of 0.07 x 0.07 m2 [34]. Webb and Ma [35] give an extensive overview of the literature available on liquid impinging jets.

By allowing the coolant fluid to boil, the latent heat capacity of the fluid can accommodate a significantly larger heat flux and achieve an almost isothermal surface. Although any comprehensive heat transfer textbook such as [36] will give an introduction to forced convection boiling, two-phase flows are complicated to model. The most important parameter in forced convection boiling is the critical heat flux (CHF), defined as the point at
which enough vapour is being formed that the surface is no longer continuously wetted. To achieve maximum cooling, one wants to run the system close to the CHF, but never above. High velocities, large subcoolings, small diameter channels and short heated lengths are known to increase the CHF. Two-phase flows may be a good option for the cooling of photovoltaic cells when the heat fluxes are high. The saturation temperature of water can be brought to 50 °C at a pressure of 0.13 bar [36]. To avoid pressurised systems, other working fluids may be used eg. Vertrel XF [37]. Ghiaasiaan and Abdel — Khalik [38] give an extensive literature review of two-phase flow in microchannels, which includes a thorough description of flow regimes in horizontal and vertical channels, correlations for pressure drops, forced flow subcooled boiling and CHF. Hetsroni et al. [37] describes a two-phase microchannel heat sink that keeps heated surface at a temperature of 50-60 °C, a temperature highly suited for photovoltaic purposes. The working fluid is Vertrel XF, which has the desired saturation temperature and is dielectric, so that it can be brought into contact with the active electronics. The study was performed at relatively low heat fluxes (< 60 kW/m2). Inoue et al. [39] study the use of boiling in confined jets (Figure 6) to cool a very high heat flux (near 30 MW/m2) in a fusion reactor. This system proposes an innovative way of preventing flow interaction between neighbouring jets, and at the same time preventing splash of water from the violent boiling that may occur at the surface under these conditions.

2 Conclusion

Cell cooling is an important factor when designing concentrating photovoltaic systems. The cooling system should be designed to keep the cell temperature low and uniform, be simple and reliable, keep parasitic power consumption to a minimum and, if possible, enable the use of extracted thermal heat.

With single-cell geometries, research shows that passive cooling is feasible and the most cost-efficient solution for concentration values of up to 1000 suns provided the cells and lenses are kept small.

Linear concentrators can also be cooled passively, but the heat sinks tend to get very intricate and therefore expensive for concentration values above 20 suns. A heat pipe based solution is one way to increase the passive cooling performance. Different ways of active cooling by water or other coolants have also been found to work well and should be considered for concentration levels above 20 suns.

For densely packed cells, active cooling is the only feasible solution. At high concentrations, the high heat flux makes a low contact resistance from cell to cooling system extremely important. Recent options such as microchannels or impinging jets generally prove to be good solutions. Microchannels are particularly promising because they have the option of being incorporated in the cell manufacturing process. Forced convection boiling give the possibility of uniform-temperature cooling at extremely high heat fluxes, but a coolant other than water is generally needed in order to keep the cells at the desired low temperature.

Academician Dmitrii Strebkov,
post-graduate Pavel Litvinov,
candidate of technical science Eduard Tverjanovich
(All-Russia Scientific Research Institute of Electrification of Agriculture)

Now the cost of modules of solar batteries is high enough. It is possible to lower this cost (in 1,4^1,6 time) using concentration of sunlight and solar cells with two-sided photosensivity. In order to prevent complication of a construction of the photoelectric converter with concentrating modules [1, 2], in the All-Russia Scientific Research Institute of Electrification of Agriculture (ВИЭСХ) the conceptual direction on development of photo­electric modules with concentrators only for stationary installation is accepted (without devices of tracing behind a seen position of the Sun on a firmament).

Principle of functioning stationary U-shaped cylindrical parabolic concentrators with the receiver located in a focal plane the following. Such concentrators assume use of photoelectric converters with a two-sided effective surface and have the big concentration in comparison with compound cylindrical concentrators. The profile of the U-shaped asymmetrical stationary concentrator [3, 4] with a continuous reflecting covering (fig. 1) is formed by two branches of parabola AB and CD, unrolled concerning common focal point F on angles ai and a2 accordingly at an aperture angle of the stationary concentrator (сн+а2) and connected among themselves on circle BC. In such concentrator under condition of |5|<a1, |6|<a2 where 5 — the angle of current declination of the sun, all rays go on receiver OF located below of a focal point of the concentrator.

SHAPE * MERGEFORMAT

At the same time, stationary concentrators with planar mirror surfaces as have shown researches, have the major losses of radiation and geometrical concentration, than the stationary concentrator of an ideal profile with a continuous reflecting surface. It is connected, first of all, to partial loss of the radiation overflowing the receiver of sunlight. Besides the major nonuniformity of functioning of the stationary planar mirror concentrator within one year is observed at an arrangement of mirrors on the inner side of a pattern with an ideal rated profile.

The increase of height of the receiver the planar mirror stationary concentrator though allows to use a solar energy more full, however leads to significant decrease of geometrical concentration. Good power parameters are observed at an arrangement of mirrors on a tangent to lateral aspect of a pattern with an ideal profile.

In other important parameter influencing functioning of the stationary U-shaped concentrator, the select of an expansion angle a is. The profile of the concentrator represented on fig. 1, has an aperture angle 2a. At 5<-a, or 5>a the sun within all day remains outside of a visibility range of the concentrator. At — a<5<a, but 5, close to +a the concentrator works short time. An operating time of the concentrator according to the introduced circuit at various angles a at latitude of Moscow is introduced on fig. 3.

Thus, the more aperture angle of the concentrator, the is more time of its work in one year. However, at increase of an aperture angle, geometrical concentration Kr0 decreases because of the breadth on midship of the stationary concentrator is reduced. From the graph (fig. 3) it is visible, that the greatest operating time t the U-shaped stationary symmetric concentrator within day is observed in days of the spring and autumn equinox, in the similar image geometrical concentration of Kr0 (3) changes also. A character of association of size of an aperture angle 2a (or an expansion angle a) from an uptime of the concentrator within day t and geometrical concentration of Kr0 is seen.

As a result of computer modelling of functioning of the U-shaped stationary concentrator power performances (fig. 4) were obtained.

The graph (fig. 4) is represented for latitude ф=56.5° (Moscow), however character of associations is fair and for all latitudes. At modelling it was supposed, that the concentrator has optical efficiency t=0.8; it was oriented on latitude of region and worked all-the-year — round. From fig. 4 it is visible, that at annual use of the concentrator in days with 5<-a and 5>a the stationary U-shaped symmetric concentrator generally remains disabled. It is observed under condition of a<23.5°.

Researches of function of capacity of the module by means of the computer have shown, that the stationary concentrator at an expansion angle a =23.5° in days, the close to a summer and winter solstice, has a dip in a power generation. As a result of this it is impossible to consider an expansion angle of the stationary concentrator a=23.5° optimum. The maximum power generation within one year falls at the stationary concentrator with an expansion angle a=27.5°. In this case the power generation is more as contrasted to a power generation the stationary concentrator with an expansion angle a=23.5° on 9%.

The carried out computer simulation of operation of the U-shaped stationary concentrator can be used for a select of such geometrical performances of the stationary concentrator, as an optimum aperture angle, a breadth on a midsection of the concentrator, its depth, etc. Energy performances can be the useful by optimization of the listed above geometrical sizes at a known time in use of the concentrator within day, and also phase of functioning of the concentrator in one year. The time in use of the concentrator within day and phase of functioning within one year is determined in turn by performances of a sink.

The literature

1. Г. Раушенбах. Справочник по проектированию солнечных батарей. М.: Энергоатомиздат, 1983 р. 57-60.

2. J. C. Minano, A. Luque, J. Parada. Recent results of non-tracing photovoltaic concentrators. Invited paper at the Fourth Sunshine Workshop on Crystalline Silicon Solar Cells. Makuhari, Japan, 1992.

3. A Luque ed., Adam Hilger. Solar Cells and Optics for Photovoltaic Concentration. — Bristol, UK, 1989, pp. 381-395.

4. J. C. Minano. Static concentration. International Journal of Solar Energy, №6, 1988, pp. 367-386.

More than 1000 grid connected PVSs are working in Saxony today. Of those, fewer than 3% have an output of more than 10kW. The annual yields of PVS that were recently constructed is at 800 kWh/kW — about 100 kWh/kW above the yield of the PvS constructed in the early 90s.

The systematic monitoring of the PVSs has proven itself successful in the 1000-roof program [7], but has since then not been financed. The re-start of a systematic monitoring — program seems to be very useful for analysing the properties and the behaviour of the new introduced components and to contribute to higher yields. Through the regulation of payment of the renewable energy law, the yields of PV power plants is now of high interest to plant owners. The today available technology of transferring PVS output data remotely should be used for obtaining further increases in yield.

The first PV power plants of MW-class will become operative in Saxony still in 2004. The 1-MW station in Meerane is about to start operations, and south of Leipzig (near Espenhain) a 5-MW plant is being constructed. The total PV-output in Saxony with therefore more than double in 2004 (reaching a total of approximately 10 MW).