- Email: [email protected]
Photovoltaic cogeneration in the built environment
Pergamon
PII: S0038 – 092X( 01 )00005 – 6
Solar Energy Vol. 71, No. 1, pp. 57–69, 2001 2001 Elsevier Science Ltd All rights reserved. Printed in Great Britain 0038-092X / 01 / $ - see front matter
www.elsevier.com / locate / solener
PHOTOVOLTAIC COGENERATION IN THE BUILT ENVIRONMENT MORGAN D. BAZILIAN* , †, FREDERIK LEENDERS**, B. G. C. VAN DER REE** and DEO PRASAD* *SOLARCH, National Solar Research Unit, UNSW, Sydney 2025, Australia **Ecofys Energy and Environment, P.O. Box 8408, NL 3503 RK Utrecht, The Netherlands Received 31 July 2000; revised version accepted 11 December 2000 Communicated by ANNE GRETE HESTNES
Abstract—Building integrated photovoltaic (BiPV) systems can form a cohesive design, construction, and energy solution for the built environment. The benefits of building integration are well documented and are gaining significant public recognition and government support. PV cells, however, convert only a small portion of the incoming insolation into electricity. The rest is either reflected or lost in the form of sensible heat and light. Various research projects have been conducted on the forms these by-products can take as cogeneration. The term cogeneration is usually associated with utility-scale fossil-fuel electrical generation using combined heat and power production. It is used here in the same spirit in the evaluation of waste heat and by-products in the production of PV electricity. It is important to have a proper synthesis between BiPV cogeneration products, building design, and other HVAC systems in order to avoid overheating or redundancy. Thus, this paper looks at the state-of-the-art in PV cogen from a whole building perspective. Both built examples and research will be reviewed. By taking a holistic approach to the research and products already available, the tools for a more effective building integrated system can be devised. This should increase net system efficiency and lower installed cost per unit area. An evaluation method is also presented that examines the energy and economic performances of PV/ T systems. The performed evaluation shows that applications that most efficiently use the low quality thermal energy produced will be the most suitable niche markets in the shortand mid-term. 2001 Elsevier Science Ltd. All rights reserved.
flat plate solar collectors that use PV cells as an integral part of the absorber plate. Research into utilizing waste heat from photovoltaic modules is more than 30 years old. At its essence is the desire to find an elegant solution to increase system efficiencies, while sharing balance-of-system (BOS) costs, and minimizing the size of solar systems (where there are issues of limited roof space). Commercially, PV cells have conversion efficiencies of between 6 and 15%, depending on various technologies. At one sun (1 kW/ m 2 ), this means that 850–940 W/ m 2 is lost as sensible heat or re-radiated into space. The main arenas of research of PV cogeneration thus far have been in PV/ T modules, ventilation systems designed to remove hot air from the PV modules, and specifying light transmission in PV panels in order to produce varying levels of daylight penetration. The evaluation of the research and designs of various PV cogen system types can bring a better understanding of the possibilities for synthesizing these efforts. The complex load profiles and economics in the built environment are primarily being met by custom architectural and engineering design. The widespread adoption of PV cogen
1. INTRODUCTION
PV cells convert a small portion of the incoming solar insolation into electricity. Innovative designs can produce secondary and tertiary energy production from PV modules. There has been research conducted into the various useful energy forms that these by-products can take as cogeneration. These multi-functional solar systems, hereafter referred to as PV cogen systems, can increase the economic viability of building integrated photovoltaic (BiPV) systems. However, the economic and energetic benefits are seldom identified as those of a PV cogen system. Thus, they are not properly designed or evaluated. This paper aims to improve the design and evaluation of PV cogen systems through a state-of-the-art overview and evaluation process. The term PV cogen will be used in this paper to describe any system that uses by-products or unused solar resources from photovoltaic processes. The term PV/ T refers to
†
Author to whom correspondence should be addressed. Tel.: 161-2-9300-8191; fax: 161-2-9385-6374; e-mail: [email protected] 57
58
M. D. Bazilian et al.
systems will require certain technological advancements, outlined later, as well as modular solutions. Installations that have a need for all of the thermal energy produced will be the most efficient.
1.1. Effects of temperature on PV cells The degradation of PV power output with increased temperature is well understood, and will play a role in the design of any PV cogen system. Mono-crystalline and poly-crystalline silicon cells have a negative temperature coefficient of approximately 0.4% / 8C (Wenham et al., 1998). Heat transfer from the modules can be accomplished through radiation, conduction, or convection. Each of these processes lends itself to a specific type of PV cogeneration system. Removing heat from the PV cells also reduces thermal cycles and stresses. 2. APPROACHES
PV cogen systems have been applied at every level of the built environment, from residential to industrial. The economics of the installed systems will be best where there is a need for heating or hot water, or where energy costs are high. Specific climatic niche markets have been identified in some studies (Bloem and Ossenbrink, 1995). Others will be identified here based on a study by Ecofys (Leenders et al., 1999). For the purpose of evaluation, PV cogen systems have been classified into four primary types.
2.1. PV/T modules PV/ T modules, in general, place PV cells on the absorber plate, or act as the absorber plate of standard solar thermal collectors. In this way, the waste heat from the PV panels is directly transferred into air, water, or phase-changing liquids that can be used for a plethora of building or off-site uses (Fig. 1). The synthesis of solar thermal collectors and
Fig. 1. Schematic diagram of Solarwerk’s Spectrum PV/ T module.
solar electric modules has certain advantages over placing the two systems next to each other (Ricaud and Roubeau, 1994 IEA PV/ T Workshop). • Cost savings • Installation cost savings. Since two or more processes of energy production are included in one module, the size of the overall system will be reduced. This can reduce installation time and mounting equipment necessary. • Space saving. Although this is not an issue at present in much of Europe, Australia or North America, there is a space consideration in Japan. As PV technologies further saturate the market, this could prove important in the future. • Balance of system (BOS) savings. Again, since two or more forms of energy can be produced, the BOS components can be shared. PV cells are currently the most expensive component in PV/ T systems. As the price of PV technology continues to decline, the cost savings in the areas previously listed will have a larger impact. Although several water-based PV/ T modules are available from Israel and Germany, there are very few commercially available.
2.2. Use of air flow behind the PV modules The removal of waste heat through the use of convective airflow behind the PV panels has also been researched (Garg and Adhikari, 1998). This approach has been used commercially either by removing heat from roof-mounted or fac¸ade systems. Natural convective flows can be used, or various mechanized ventilation systems utilizing fan-powered pressurization can be installed to move the air. The thermal energy can also be transferred to another medium, such as water. During times of cooling loads, the hot air can act as the basis for a stack effect to remove unwanted hot air. A stack effect can be created by the use of either active or passive systems. In some of the installations studied, the use of flues or chimneys was created between the PV wall and the exterior wall of the fac¸ade to accommodate this effect. It is most beneficial to the economics of the PV cogen system if natural convective cycles can be set up rather than using fans and motors (Fig. 2). The use of double skin facades in both new construction and renovations has helped spur several research projects and installations. The recently finished BP Amoco wall in Trondheim, Norway, will be an excellent source of data (Aschehoug et al., 2000). Rain screen cladding is
Photovoltaic cogeneration in the built environment
Fig. 2. Schematic of the PV cogen at the Ispra JRC test facility (H. Bloem).
another building material devised for use in fac¸ade renovation projects of older concrete buildings (BP Solarex). Although it provides a space for ventilation to cool the panels, the air cannot be harnessed for indoor heating.
2.3. Daylighting applications Many recent BiPV applications have incorporated the benefits of daylighting into their designs. Electric lighting is one of the largest sources of electrical use in most typical commercial buildings. Major studies have been conducted showing the quantifiable gains of daylighting to increase productivity in the workplace while also assisting in lowering energy bills (PG&E). PV panels are easily adaptable for light transmission. Thus, the PV cogen system requires no additional system cost, except for ambient light sensors to optimize the gain from daylighting (Fig. 3). This approach is classified as cogeneration in order to underscore the importance of utilizing an otherwise lost resource in the production of PV electricity. This type of PV cogen system has far more current installations than the other two categories. It is important to qualify these installations as PV cogen and thus receive the economic benefits of daylighting towards the cost of a BiPV system. Not only will it help the economics from a simple payback or life cycle cost analysis, but will allow PV cogen systems to join the current growth in daylighting applications and funding. Likewise, the use of PV modules for shading applications
Fig. 3. Innovative BiPV corridor, using daylighting and ventilation possibilities (Bear Architects).
59
has met with good success commercially. As this application requires only innovative mounting solutions, the additional costs for shading are minimized. The use of daylighting can be achieved by specifying light transmission through BiPV panels. This can be done in either amorphous or crystalline module types. Crystalline or poly-crystalline cells can be spaced apart in a module to cover nearly any desired percentage of glazing. Laser grooving can be applied to amorphous panels in a wide variety of patterns. The majority of these modules are custom products and are, therefore, very expensive (Figs. 4 and 5). Testing and modeling of these systems can be done in a number of ways. Well-accepted daylighting algorithms are included in software such as Radiance. Architectural scale mock-ups and three-dimensional modeling tools can also give valuable insights. The trade-offs between the various types of assessment have been explored (Lien and Hestnes, 2000)
2.4. The multifunctional fac¸ade The notion of the multifunctional or smart fac¸ade is relatively new. The term has been used to describe a building fac¸ade or roof that performs a number of energetic tasks. Strictly speaking, any building envelope is multifunctional. It serves as a weather protector, a fire retardant, a source of light, ventilation, security, and the like. The new multifunctional solar fac¸ade or roof can incorporate PV cogen systems that can produce heat, light, and electricity. These systems may also be separated or not used at all depending on the building load and optimization studies. A recent study conducted in Finland showed the maximum electricity benefit for a south-facing fac¸ade in northern European conditions (Vartiainen et al., 1999). Although the study used the PVs solely for electrical production, the synthesis between the transparent glazing and the PV modules was optimized for electrical production. This is a very
Figs. 4 and 5. Two patterns for allowing light through amorphous PV modules (EPV).
60
M. D. Bazilian et al.
positive step in intelligent building integration. Any multifunctional solar fac¸ade must optimize a tremendous number of variables (Peippo et al., 1999). Optimized fac¸ade design is therefore at the heart of whole building design. 3. RESEARCH
Research activities in PV cogen date back to the late 1970s. Some of the primary research objectives are listed below. This list is used only to give a cursory idea of the more current research trends in PV cogen. (Specific authors are not cited in many cases because the list is necessarily a generalization.) • Thermodynamic modeling and simulation of PV/ T modules. This research created various mathematical models to predict energy production, fluid flows and the like. • Design considerations of PV/ T modules. This research includes specific flat-plate collector design, glazing options, and even mounting and installation issues. • The use of thin films on PV/ T modules. This research is mostly in the area of amorphous silicon and will be covered in its own section. • Optimizing useful energy in multifunctional fac¸ades. These important models seek to maximize the energy output of a building element, while taking into account building design and occupant comfort (Peippo et al., 1999). • Fluid flow modeling behind PV arrays. This kind of investigation seeks to quantify and / or maximize the convective cycles present in the waste heat of PV modules in fac¸ade application. • Comparisons of single and double pass flatplate collector types for PV/ T hot-air applications. • PV/ T building integration. This research has been conducted within a BiPV framework. This includes economic analysis, building material credits as well as mounting and installation. • The use of concentrating troughs with PV/ T modules. There have been a few papers looking at CPC troughs and Fresnel lenses. Most of the models used silicon cells. Some modeled systems that use GaAS and other III–V combination cells. • Combined efficiencies of PV/ T modules. These studies looked at various aspects of efficiency ratings for PV/ T modules (Luque and Marti, 1999). Academic and government funding has sponsored
the bulk of the research in PV cogen systems. This is typical in cutting-edge technologies with small market realization. The majority of these government-funded projects have been in the EU (Joule and Thermie), the USA (PV:BONUS), and Japan (NEDO). Gregory Kiss produced an economic and energy study for NREL (Kiss and Kinkead, 1995) that compares a series of BiPV options for a model building. It concluded that the heat recovered can be up to 20 times the total amount required by the building in a climate like Oakland, CA, USA. Even in a cold climate, like Chicago, IL, USA, the system produced up to three times the building’s total heat requirement. The simple payback, calculated in years from five different geographic locations, declined from nearly 70 years with only the electrical production included, to 57 years when the thermal benefits were included, to 44 years when the thermal and daylighting benefits were included. The paper shows that PV cogen systems can greatly help the economic and energetic appeal of BiPV systems and should be considered. Research on PV/ T modules mainly focuses on the integration of PV cells and solar hot-water heaters for tap-water heating. Integration will yield economic and aesthetic advantages, but will reduce system efficiency. System efficiency of PV/ T systems will decline due to the need of PV modules to have low operating temperatures for high electrical yield. The solar water heater, on the other hand, requires a high operating temperature for high thermal production. Thus, the PV/ T collector will always be a compromise in operating temperature and, therefore, have lower energy yields than the two systems installed separately side by side. The technological challenge, as opposed to market interests, of improving PV/ T systems is the impetus for most research. A substantial amount of research has been conducted to evaluate the airflow behind PV panels. In Japan and North America, this has led to market-ready products and systems, e.g., OM Solar from Sora designs and Solarwall from Conservall. In Europe, this has not been the case. Some projects have been built, such as the Mataro library outside Barcelona (Lloret et al., 1995), and the Elsa building in Ispra (Clarke et al., 1995). Research is being conducted at the JRC in Ispra, and at the Universities of Cardiff and Strathclyde in the UK, among others (Brinkworth et al., 2000). There has been no market penetration of commercially available products however.
Photovoltaic cogeneration in the built environment
The fundamental reason appears to be the differences in climate and insolation. During winter, large areas of Europe are marked by low levels of insolation and low temperature, whereas large areas of Japan and the USA have high insolation and low temperatures that are ideal for this system type. There has been some success with PV air ventilation systems in the recent PV:VENT and INNOPEX projects in Denmark that might lead the way for commercial products in the near term (Pedersen, 2000). Daylighting, multifunctional fac¸ades, and water-based PV/ T systems are applicable in almost any climate, although the specific building functions may differ. 4. AMORPHOUS SILICON IN PV/ T
Research has been aimed at the possibilities for a new generation of PV/ T modules utilizing amorphous silicon cells as the absorber plate (Fig. 6). A study (Ricaud and Roubeau, 1994) stated that ‘ . . . stabilized a-Si:H PV modules behave as if they had a positive temperature coefficient of 0.2% / 8C’. Other studies have agreed that, when heated, a-Si:H modules show an annealing effect that results in a decrease of the Staebler-Wronski effect (Affolter et al., 1996). Amorphous cells thus appear to have excellent heat characteristics that are ideally suited for PV/ T modules. One IEA PV/ T study concluded that there will be a market for PV/ T in the 10 MW range by 2005. It found that amorphous silicon could be ideal for PV/ T use if its efficiencies reach 10% (Affolter et al., 1997). Another benefit of amorphous silicon is that it can be deposited directly on large surfaces and a variety of substrates (Platz et al., 1997). This can lower production costs. The lower cost of a-Si:H modules is, of course, a significant factor as well. However, the currently low efficiency levels (4– 7%) commercially available can detract from
Fig. 6. Temperature cost graph of amorphous vs. mono-crystalline PV cells (Affolter et al., 1997).
61
economic scenarios for widespread use. Very different performance characteristics were witnessed between certain manufacturers, which will make commercial viability even more challenging (Affolter et al., 1997).
4.1. Collateral energy benefits and distributed generation Collateral energy benefits include environmental externalities that are not normally taken into consideration in cost analyses. These energy benefits can be multiplied if the benefits of distributed generation are taken into account. The offset of greenhouse gases, capacity credits, reduced ohmic and reactive power losses, improved reliability, and transmission and distribution savings can be significant (Wenham et al., 1998). Distributed generation is receiving a great amount of attention from the press and the utility industry at both global and national levels. Studying PV cogen in these terms will be useful to its market infiltration. It has the added benefits of reducing fuel consumption as well as electricity. The concepts and complexities of distributed generation are beyond the scope of this paper. 5. BUILT EXAMPLES
Innovations in PV cogen in the built environment should strive to (Leenders and van der Ree, 1999, IEA PV/ T Workshop); • satisfy the architectural design • satisfy the energy requirements of the building • satisfy the client • satisfy the end-user or occupants of the building • create a useful and delightful space to inhabit The Doxford Solar Office Building in the UK uses a ventilation system to remove waste heat from behind the BiPV system. It uses thermally broken, low level, inclined vents. The air can then be used for building needs or vented outside through operable windows. It has a 73-kW array. Specially designed fins placed on the roof help to decrease the negative effect of wind on the convective cycle (Fig. 7) (Lloyd Jones and Watts, 2000). The 25-m south fac¸ade of the Elsa building in Ispra is covered with 505 m 2 of amorphous PV modules rated at 21 kWp. There is an air gap behind the PV array that collects hot air being used in an integrated ventilation system. A series of ‘chimneys’ have been constructed behind the fac¸ade to provide an area for convective air flow. This can charge the mass of the structural wall system as well (Wouters et al., 1998).
62
M. D. Bazilian et al.
Fig. 7. Doxford building (A. Lloyd Weber). Fig. 8. Applebee’s PV cogen schematic (K. Sheinkopf, 1997).
The Test House in Japan was constructed to study the use of PV/ T modules in residential applications. Roof space in Japan is becoming scarce. The system consists of 24 PV/ T panels that produce 3.2 kW electric and 25 kW thermal. The collector area is 50 m 2 (Strong, 1995). The BiPV entryway to the Thoreau Center for Sustainability building in San Francisco shows one of the countless options for daylighting. Crystalline cells have been spaced apart in order to allow light to wash the entry space. Architecturally interesting spaces can be created and possibilities for overheating can be diminished. This type of application is ideally suited to unconditioned spaces, but can be used in conjunction with insulated glass for building skin applications (http: / / www.thoreau.org). The Scheidegger Metallblau building in Switzerland uses PV modules as shading devices, as well as having hot air collected behind the fac¸ade. It has an 18-kW PV fac¸ade with 12 kW of thermal power. The measured efficiency of the combined system is about 40%. The system was designed by Atlantis Energy (Posansky et al., 1994) The Fenster AG factory building, which is also in Switzerland, uses the waste heat collected from its rooftop monitors for the factory’s energy needs. It consists of a 9-kW vertical PV fac¸ade and a 53-kW thermal cogen PV system integrated into the skylights. The thermal energy can be stored in the concrete slabs of the basement during the summer. The system was also designed by Atlantis Energy. In the PV:BONUS Program project at a North Carolina Applebee’s, hot air taken from behind the PV panels is transferred to hot water, which can be used in the restaurant. The roof has an integrated 1.6 kWp PV system. The cost of the PV cogen system was calculated to cost US$54.00 / m 2 more than a conventional roof (Sheinkopf, 1997; Fig. 8). Two projects named Brig and Rigi were installed in the residential sector by Atlantis Energy in Switzerland. The ‘Brig’ hybrid shingle roof has
a peak power of 15 kW electric and 30 kW thermal. The thermal energy is used to heat the shower rooms of the facility. The ‘Rigi’ system has 8.4 kW electric and 12 kW thermal and is also integrated into a typical shingled roof. The rooftops have an air gap behind the BiPV array and the aim is to maximize the flow of natural convection. The air can be used inside, vented out through a ridge vent, or stored. The simplicity of these systems is what is most appealing. A system of this type can use traditional construction techniques without resorting to additional hot air collector systems. These have been in place since 1993 (Posansky et al., 1994). Another good example of PV cogen in the built environment is the Mataro Public Library in Barcelona. Multifunctional PV panels are integrated into the south fac¸ade. The system was designed to produce an optimal PV and thermal output. A ventilation system is used to move the hot air from behind the modules. Some of the modules are semi-transparent, to allow for natural-light penetration (Lloret et al., 1995).
6. PRODUCTS
The growth of the BiPV market and its influence on innovative design and construction integration components has spurred along the growing interest in PV cogen. PV/ T systems are mostly modular and can eventually be mass-produced. Custom architectural products for use with BiPV systems will be more difficult to massproduce. Esbensen Consulting produced an extensive list of existing systems as part of the IEA PVPS Task VII work (Soerensen and Munro, 2000). It shows that there are three currently available PV/ T systems using domestic hot water product technologies, and one using conventional air-collector technology. The products’ actual commercial availability is difficult to ascertain in some instances.
Photovoltaic cogeneration in the built environment
• Up until four years ago, SunWatt Corp., a small American company, produced a concentrator-type PV/ T module. They have ceased production of this unit due to the inability to buy the proper fin extrusions. They are producing a prototype PV still product that uses the evaporation of seawater to cool solar cells. This product will probably be aimed at the developing world. • The Chromagen ‘Multi Solar System’ is sold at roughly US $1940 / m 2 . It uses solar cells integrated into the non-spectrally selective absorber plate. The company reported that although the product is still under testing, units have been sold and are presently commercially available (www.chromagen.co.il) (Elazari, 2000). • The Solarwerk product has been named ‘Spectrum’. It has a peak output of 250 W, a thermal efficiency of 63%, and an electrical efficiency of 12%. It has 180 PV cells mounted onto the absorber plate. The module is sold with a 2.2-m 2 collector area (Solarwerk). • The OM Solar product is more of a wholebuilding approach to residential energy needs than a product per se. The system collects solar-heated air from underneath the roof and channels it either outside, in times of cooling, to a heat-storing concrete slab, or directly into the building, for heating. They operate as designers and contractors of this system in Japan and now in America (Sora Designs). The Task 7 list showed 28 products that were under research or production in its last edition. A number of these are projects that have used custom PV cogen systems. Most are in product development at various stages. Another interesting product is the Conservall PV Solarwall. The current commercially installed PV Solarwall product uses PV power to run ventilation fans that move hot air from the Solarwall product for building pre-heat. The PV panels get cooled by the removal of heat behind the metal cladding, thus, it can tentatively be called PV cogen through a mutually beneficial relationship. Large glass manufacturers, like Flabeg, previously known as Pilkington (DE) and Viracon (USA), are working with PV manufacturers to create modular and custom solutions for architects and engineers. High-performance glazing can be specified for spectral sensitivity or U-value, in addition to the integration of PV cells. • A member of the Flabeg Group has had the OPTISOL PV glazing system commercially
63
available for BiPV applications for nearly a decade. • BP Solarex is in testing and product design of their Powerview glass product, which will be an architectural glass product with integrated PV cells. • Various fac¸ade contractors, e.g. Kawneer (CA), Schuco (DE), and Oskomera (NL), have developed PV fac¸ades. Architectural glass panels use current technologies and understood construction practices. This is a key factor for dissemination of any BiPV system. 7. AREAS FOR FURTHER INVESTIGATION
The groundwork has been set for the commercial viability of PV cogen systems. There is still an ample amount of worthwhile directions for further study. • PV cogen systems need considerable additional real-world testing to peak interest in the market. • Long-term testing of the temperature effects of heat on a-Si cells and crystalline cells integrated on absorber plates, or acting as them, will be necessary. Stagnation temperatures in PV/ T hot-water systems can reach 1608C (Affolter et al., 1997). • Standardized methods for efficiency ratings of combined systems are essential for comparisons. • Recent studies of rural health clinics in the developing world have shown the need for both electricity and thermal energy (Olsen and Jimenez, 1998). ‘ . . . the World Health Organization has been looking for ways to use renewable energy technology for a broader range of applications . . . ’. These broader applications could use waste heat from PV modules or PV/ T modules for use in the sterilization of medical equipment. Heat energy can also be used for pasteurization, water distillation, and, of course, water heating. Systems should be first tested in easily controlled facilities before being exported. • Seasonal heat storage at low temperature needs additional research and monitored built examples. • Concentrating systems can be an attractive option since they can effectively increase cell efficiency. They can thus benefit from a cogen system to remove large amounts of excess heat (Karlsson and Wilson, 1999). • Selectively coated mirrors can be used to
64
M. D. Bazilian et al.
Table 1. Evaluation criteria Criteria
(unit)
Time to market Market potential Investment Building integration Thermal performance Electrical performance Energy consumption Sustainable building Life time Effect on energy performance
year m 2 / yr Euro / m 2 Euro / m 2 GJ / year?m 2 kWh e / year?m 2 kWh e / year?m 2 2 year 2
optimize the incoming light energy. Light with wavelengths longer than the band-gap of the PV cells can be focused on thermal production (Yang et al., 1997). • There is a need for more economic analyses of PV cogen systems. These will be necessarily complex because there is more than one type of energy being produced. It will have to be completed within the context of a whole-building perspective, looking at the entire energy flow. • Optimization tools that can combine the advantages of BiPV along with fluid dynamic or thermodynamic evaluation will be needed to quantify results and aid in system design. • Market size and type studies. Finding niche markets in various climatic situations will be key in the short-term. Research in PV cogen systems has proved to be a positive environment to stimulate dialogue between experts from the solar thermal and solar electric fields. These researchers do not usually coordinate their efforts and have thus lost out on valuable insights. In order for any mass-market inclusion of solar energy systems, there will need to be continued conversations between these parties. 8. EVALUATION METHOD
The evaluation method originated from a growing need in The Netherlands to evaluate and compare various PV cogen systems that were applicable under Dutch conditions. The methodology begins with a definition of the most
important criteria to evaluate a promising hybrid PV/ T system (Table 1). These ten criteria were used to validate eight promising hybrid PV/ T systems. The systems were defined and selected by a panel of experts. The proposed systems are believed to be the most promising systems using today’s technology. They are applicable in moderate northwest European climates and are technically feasible within 5 to 10 years in new building projects. The goal of the study is to show the real technological and market value of the defined systems. Daylighting systems are not examined within this construct, only PV and thermal combinations are assessed.
9. EVALUATED HYBRID SYSTEMS
During the selection of the hybrid systems to be evaluated, priority was given to closely matching supply and demand of heat since the usefulness of the produced heat is the crucial factor in determining the worth of PV/ T systems both economically and energetically. A wide variety of systems types were evaluated. In all cases, the electricity was assumed to be fed to the grid as an ideal storage source. The following eight systems were assessed for applications in The Netherlands:
9.1. PV building fac¸ade Building-integrated PV in fac¸ades offers a relatively simple opportunity to utilise the heat generated in the PV panels, as previously noted. The PV fac¸ade acts as an unglazed PV/ T air collector, which supplies natural ventilation in summer and pre-heated air in winter. In spring or summer, either ventilation or pre-heated air can be supplied, depending on the climate. Applications are foreseen in both industrial and commercial buildings. The Solarwall is a practical and marketready product of this type of system (Fig. 9). It has been primarily targeted at the industrial sector where aesthetic concerns are not high, comfort levels for occupants are not strict, and pre-heating of ventilation air is of some importance. Sound
Fig. 9. Unglazed PV/ T air collector as a ventilated PV fac¸ade in an industrial building.
Photovoltaic cogeneration in the built environment
65
Fig. 10. Unglazed PV/ T air collector for drying tulip bulbs. Fig. 12. Unglazed PV/ T water collector for pool heating.
PV/ T integration into buildings is more difficult to accomplish as the thermal comfort requirements are relatively severe. However, there is ample evidence of this type of system in the market, as previously described. These applications can also take economic advantage of displacing expensive traditional fac¸ade elements. The use of PV systems to reduce heat gains by shading are not included in this evaluation method.
9.5. Heat pump I The pre-heated air of the unglazed PV/ T collector supplies the heat pump with its heat source. The heat pump upgrades the pre-heated air to low temperature space heating. Additionally, a heat recovery unit is installed to reduce heat losses. This system has been demonstrated in Zwaag (NL) since 1998 (Fig. 13).
9.2. Biomass dryer An unglazed PV/ T collector can be used to dry biomass (e.g. tulip bulbs and woodchips). Not all industrial drying processes can be utilised by solar energy due to the relatively low energy efficiency and low temperatures (Fig. 10).
9.6. Heat pump II
An unglazed PV/ T collector can pre-heat the cold water supply (108C). During the summer, auxiliary heating is not necessary (Fig. 12).
Here, an unglazed PV/ T collector is combined with a heat pump and aquifer. In summer, the PV/ T collector is cooled with 5–108C water from the aquifer. While cooling the PV/ T collector, the water is heated to about 208C and is stored in the aquifer to be used as a heat source for the heat pump in winter. In winter, the heat pump upgrades the stored heat for low temperature space heating to about 408C. [During this process, the aquifer is again fed with cold water (5–108C)]. This system offers opportunities to regenerate the heat in the soil when heat pumps are used on a large scale in urban areas and is now being tested (Fig. 14).
Fig. 11. Glazed PV/ T water collector for pool heating.
Fig. 13. Unglazed PV/ T air collector combined with a heat pump and a heat recovery unit.
9.3. Indoor swimming pool To heat the water of an indoor swimming pool to a maximum of 308C, a simply covered PV/ T collector will suffice (Fig. 11).
9.4. Indoor /outdoor swimming pool
66
M. D. Bazilian et al.
Fig. 14. Unglazed PV/ T water collector combined with a heat pump and aquifer.
9.7. Solar hot-water system A glazed PV/ T collector combines directly the functions of PV (electricity) and a domestic hotwater system (hot tap water). The PV panel acts as an absorber in the thermal collector (Fig. 15).
9.8. Solar hot-water system and space heating This is identical to that described in the preceding paragraph, but now the PV/ T collector also supplies heat for space heating, meaning a larger roof area needs to be covered with PV/ T. 10. RESULTS
The systems were evaluated against the ten criteria previously mentioned. The effects of each criterion on the individual systems was quantified for the Dutch circumstances. This resulted in an extended and quantified matrix, with the upper row listing the systems, and the left column listing the ten criteria. Some highlights are listed here. It should be noted that markets other than the Dutch would show different results, although the methodology can be applied in any number of markets.
10.1. Market potential The research showed that the maximum possible market potential in The Netherlands is the largest for the heat pump systems (about
2 ? 10 6 m 2 / year). The second largest market potential exists for the solar water heaters (about 1 ? 10 6 m 2 / year), followed by the PV fac¸ade (about 0.3 ? 10 6 m 2 / year). The swimming-pool applications seem to be a nice ‘niche’ market (about 2–7 ? 10 3 m 2 / year); investments are low and revenues relatively high thanks to low operating temperatures. These systems must compete with other energy systems, including conventional gasfired systems. The real market potential will thus be lower than the maximum market potential described herein.
10.2. Time to market Concerning the time-to-market, the PV/ T combination of PV and solar water heaters is most advanced. The time to market demonstration is zero (PV/ T technology is available), and full market introduction may follow in about 2 years time. In the meantime, experiences and learned lessons can be evaluated and processed. Perhaps the same absorber can be used in unglazed systems, like the swimming-pool applications. The PV fac¸ade is also in the market introduction phase. Performance data and cost / benefit analyses in different climates must prove its validity. The time to full market introduction is about 2 to 3 years. The heat pump systems are still in a premature development phase; new technologies have to be tested and experimented upon (now under development in The Netherlands). The time to market demonstration and the introduction of the heat pump systems will take the longest time.
10.3. Simple pay-out time
Fig. 15. Example of a PV/ T water collector.
Using the data on investment costs, saved materials, electrical and thermal revenues, and their benefits, the simple pay-out time (SPOT) of the various systems could be calculated (Tables 3 and 4) and compared to those of a traditional PV
Photovoltaic cogeneration in the built environment Table 2. Simple pay-out time (reference)
d 5 Combustion efficiency (0.85)
Fac¸ade
Roof
p-Si a
a-Si b
p-Si
a-Si
BIPV (Euro / m 2 ) Yield (kWh / m 2 ?year) Benefit (Euro / m 2 ?year)
660
450
570
430
55
35
80
50
6
4
9
SPOT (year)
110
a
67
112
e 5 Heat of combustion (35 MJ / m 3 ) f 5 Gas price (0.2 Euro / m 3 ) 11. CONCLUSIONS
5.5
63
78
100 Wp / m 2 ; b 60 Wp / m 2 .
panel (Table 2). This calculation is performed for both the p-Si and a-Si PV panels (cost estimation is for the year 2000). Calculation of the SPOT (year) 5 (Co /Es 1 Ts) Co 5 Costs (Euro / m 2 ) 5 Product investment 1 Costs of system integration 2 Material savings Es 5 Electricity savings (Euro / m 2 ? year) 5 a ? b a 5 Net electricity yield (kWh / m 2 ? year) b 5 Electricity price (0.1 Euro / kWh) Ts 5 Thermal savings (Euro / m 2 ? year) 5 (c ? f ) /(d ? e) c 5 Thermal yield (MJ / m 2 ? year)
Using the waste heat and available daylight from PV modules is an elegant idea, and has been successfully installed and operated in a number of commercial and industrial facilities. Using a comprehensive term for its description, like PV cogen, could prove to be a useful addition to the vocabulary of building professionals. From the technological perspective, PV/ T systems are especially suitable for low temperature. For medium-temperature applications, the thermal and electrical yield of the hybrid system is lower than that of two separate systems. The combination of the two will always be a compromise. Hence, a PV/ T system is economically viable when the costs of the reduced energy performance matches the gained costs of production, installation, and mounting. Apart from the economic motive, the uniform appearance of a PV cogen system may provide an important surplus value in terms of enhanced architectural aesthetics. A PV cogen system will also serve to meet the consumer’s desire for solar energy to meet both thermal and electrical loads. This is difficult however since the supply and demand of electrical and
Table 3. Simple pay-out time (a-Si) a-Si
Reference Fac¸ade
Costs 450 (Euro / m 2 ) Electricity 4 saving (Euro / m 2 ?year) Gas saving (Euro / m 2 ?year) SPOT 110
Swimming pool
DHW
Roof
PV fac¸ade
Industry Biomass dryer
HP I (HRU)
HP II (soil)
In
In and out
Tap water
Tap and space
430
220
420
590
645
515
425
680
645
5.5
22.5
3 1.5
80
Domestic
49
5
4.5
5
2
2
170
2
2
4.5
5
4.5
4
5.5
6
7.5
4
52
39
57
80
Table 4. Simple pay-out time (p-Si) p-Si
Reference Fac¸ade
Costs 660 2 (Euro / m ) Electricity 55 saving (Euro / m 2 ?year) Gas saving 6 (Euro / m 2 ?year) SPOT 110
Industry
Domestic
Swimming pool DHW
Roof
PV fac¸ade
Biomass dryer
HP I (HRU)
HP II (soil)
In
In and out
Tap water
Tap and space
570
0–445
560
730
830
655
560
865
830
80
5.5
1.5
9
1.5
4 102
65
0–64
8
8
7.5
8.5
7.5
7.5
2
2
5
5.5
7
3.5
2
2
52
40
60
75
68
M. D. Bazilian et al.
thermal energy are rarely properly matched. Continued research and information exchange between PV and solar thermal experts are necessary, and must occur at an international level. The glazed collectors in swimming-pool applications have the shortest pay-back period through the efficient and cost-effective use of low-temperature heat. The PV/ T fac¸ade might be an attractive niche market as a PV/ T fac¸ade features architectonic possibilities and can displace building materials. Economic and technical performance are strongly related to climatic conditions, building energy loads, system energy production, and the amount of material saved. Domestic hot water PV/ T systems are currently not profitable. However, this product could have large market volumes and is expected to emerge to a mature market within a relatively short period of time, as interest in BiPV and thermal systems in the domestic market is growing fast. The biomass dryer has the longest pay-back period. This is caused by the high electricity consumption of the fan needed to maintain sufficient air flow to dry the biomass. The pay-back period of the heat pump systems was not calculated as these systems yield heat that cannot be used directly. The heat of the PV/ T collector serves as a heat source that must be upgraded by a heat pump to match the real energy demand. Careful design and the education of architects and engineers can foster the use of PV cogen techniques. The combination of PV in daylighting systems or multifunctional fac¸ades was not included in the evaluation, but its advantages are clear. Specifying for daylighting and ventilation schemes in high quality architectural BiPV installations will require the cooperation of many building design and system experts. Daylighting should continue as the most likely mode for PV cogen in the near term in the commercial market. Heat-storage systems could greatly aid the economic viability of PV cogen by more closely matching production and demand, or by being utilized at peak times. PV cogen should be researched and installed in a holistic manner. By taking into account the benefits of electrical, thermal, and daylight production systems, they can appear more attractive, both economically and environmentally. With the growth and interest in BiPV and, more generally, in distributed energy systems and cogeneration, PV cogen could play a large role in future PV installations at every level of the built environment. Acknowledgements—The authors wish to thank: Novem, the
Netherlands Agency for Energy and Environment B.V., for financing part of the performed work. TNO, the Netherlands Organisation for Applied Scientific Research, and ECN, The Netherlands Energy Research Foundation are thanked for their contributions in developing the evaluation method. Thanks are also due to the National Solar Architecture Research Unit, the Australian CRC for Renewable Energy and the Faculty of the Built Environment at the UNSW. A new IEA working group on combined PV/ T concepts is in the process of compiling an extensive resource list on this subject. It will be made available on the group’s web site in due course at http: / / www.task7.org / pvt / pvt home.html ]
REFERENCES Affolter P., Haller A., Ruoss D. and Toggweiler P. (1996) Absorption and high temperature behaviour evaluation of amorphous modules. In Report for Project 56360 /16868 for the Swiss Federal Office for Energy. Affolter P., Gay J., Haller B. A., Althaus H., Ruoss D. and Toggweiler P. (1997) A new generation of hybrid solar collectors. In Report for Project 56360 /16868 for the Swiss Federal Office for Energy. Aschehoug O., Hestnes A., Matusiak B., Lien A., Stang J. and Dagfinn B. (2000) BP amoco solar skin. In Proceedings for EUROSUN 2000, June 19–22, Copenhagen, Denmark. Bloem J. J. and Ossenbrink H. (1995) Thermal aspects of PV integration in buildings. The 13 th European PV Solar Energy Conference, Nice. Brinkworth B. J., Marshall R. H. and Ibarahim Z. (2000) A validated model of naturally ventilated PV cladding. Solar Energy 69(1), 67–81. Clarke J. A., Johnstone C., Strachan P., Bloem J. J. and Ossenbrink H. (1995) Thermal and power modelling of the photovoltaic fac¸ade on the ELSA building, Ispra. The 13 th European PV Solar Energy Conference, Nice. Elazari A. (2000) Building integrated multi pv / t / a solar system roof tile, Eurosun 2000 Preprint, June 19–22, Copenhagen, Denmark. Garg H. P. and Adhikari R. S. (1998) Transient simulation of conventional hybrid photovoltaic / thermal air heating collectors. International Journal of Energy Research 22, 547– 562. Karlsson B. and Wilson G. (1999) MaReCo–CPC for high latitudes. Proceedings ISES 1999 World Congress, Jerusalem. Kiss G. and Kinkead J. (1995) Building integrated photovoltaics: a case study. In NREL Report TP-472 -7574. Leenders F. et al. (1999) Technology review on PV/ T systems, Ecofys no. E21036. Utrecht, The Netherlands. Leenders F. and van der Ree B. G. C. (1999) Photovoltaic / thermal systems. From Workshop on PV/Thermal Systems, 17–18 September, Amersfoot, The Netherlands. Lien A. G. and Hestnes A. G. (2000) Visual studies of transparent PV elements. In Proceedings for EUROSUN 2000, June 19–22, Copenhagen, Denmark. Lloret et al. (1995) The Mataro´ public library: a 53 kWp grid connected building with integrated PV — thermal multifunctional modules. 13 th European PV Solar Energy Conference, Nice, France, pp. 490–493. Lloyd Jones D. and Watts B. (2000) Effective use of building integrated photovoltaic waste heat: three projects. In Proceedings from the 2 nd World Solar Electric Buildings Conference, 8–10 March, Sydney. Luque A. and Marti A. (1999) Limiting efficiency of coupled thermal and photovoltaic converters. Solar Energy Materials & Solar Cells 58, 147–165. Olsen K. and Jimenez A. (1998) Renewable energy for rural health clinics. National Renewable Energy Laboratory Publishing, Golden, CO, USA. Pedersen P. (2000) Cost effective BIPV systems with combined electricity and heat production. In Proceedings for EUROSUN 2000, June 19–22, Copenhagen, Denmark.
Photovoltaic cogeneration in the built environment Peippo K., Lund P. and Vartiainen E. (1999) Multivariate optimization of design trade-offs for solar low energy buildings. Energy and Buildings 29(2), 189–205. Platz R., Fischer D., Zufferey M., Anna Selvan J. A., Haller A. and Shah A. (1997) Hybrid collectors using thin-film technology. In Proceedings 26 th PVSC, 30 September–3 October, Annaheim, CA, USA, pp. 1293–1296. Posansky M., Gnos S. and Coonene S. (1994) The importance of hybrid PV-building integration. In Proceedings of First WCPEC, 5–9 December, Hawaii, pp. 998–1003. Ricaud A. and Roubeau P. (1994) ‘‘Capthel’’, A 66% efficient solar module and the ‘‘Ecothel’’ co-generation solar system. In Proceedings of First WCPEC, 5–9 December, Hawaii, pp. 1012–1015. Sheinkopf K. (1997) PV system with thermal heat recovery. CADDET Renewable Energy Newsletter, ETSU, UK. Soerensen H. and Munro D. (2000) Hybrid PV/ thermal collectors. In Proceedings from the 2 nd World Solar Electric Buildings Conference, 8–10 March, Sydney.
69
Strong S. (1995). The Dawning of Solar Electric Architecture, NREL Publications, Harvard, MA. Vartiainen E., Peippo K. and Lund P. (1999) Daylight optimization of multifunctional solar facades. Solar Energy 6(3), 223–235. Wenham S., Green M. and Watt M. (1998) Applied Photovoltaics, University of NSW Publications, Sydney. Wouters P., Vandale L., Bloem H. and Zaiman W. J. (1998) Combined heat and power from hybrid photovoltaic building integrated components: results from overall performance assessment. Proceedings from the 2 nd World Conference on Photovoltaic Solar Energy Conversion, Vienna. Yang M., Izumi H., Sato M., Matsunaga S., Takamoto T., Tsuzuki K., Amono T. and Yamaguchi M. (1997) A 3 kW PV–thermal system for home use. In Proceedings 26 th PVSC, September 30–October 3, Annaheim, CA, USA, pp. 1313–1316.
PII: S0038 – 092X( 01 )00005 – 6
Solar Energy Vol. 71, No. 1, pp. 57–69, 2001 2001 Elsevier Science Ltd All rights reserved. Printed in Great Britain 0038-092X / 01 / $ - see front matter
www.elsevier.com / locate / solener
PHOTOVOLTAIC COGENERATION IN THE BUILT ENVIRONMENT MORGAN D. BAZILIAN* , †, FREDERIK LEENDERS**, B. G. C. VAN DER REE** and DEO PRASAD* *SOLARCH, National Solar Research Unit, UNSW, Sydney 2025, Australia **Ecofys Energy and Environment, P.O. Box 8408, NL 3503 RK Utrecht, The Netherlands Received 31 July 2000; revised version accepted 11 December 2000 Communicated by ANNE GRETE HESTNES
Abstract—Building integrated photovoltaic (BiPV) systems can form a cohesive design, construction, and energy solution for the built environment. The benefits of building integration are well documented and are gaining significant public recognition and government support. PV cells, however, convert only a small portion of the incoming insolation into electricity. The rest is either reflected or lost in the form of sensible heat and light. Various research projects have been conducted on the forms these by-products can take as cogeneration. The term cogeneration is usually associated with utility-scale fossil-fuel electrical generation using combined heat and power production. It is used here in the same spirit in the evaluation of waste heat and by-products in the production of PV electricity. It is important to have a proper synthesis between BiPV cogeneration products, building design, and other HVAC systems in order to avoid overheating or redundancy. Thus, this paper looks at the state-of-the-art in PV cogen from a whole building perspective. Both built examples and research will be reviewed. By taking a holistic approach to the research and products already available, the tools for a more effective building integrated system can be devised. This should increase net system efficiency and lower installed cost per unit area. An evaluation method is also presented that examines the energy and economic performances of PV/ T systems. The performed evaluation shows that applications that most efficiently use the low quality thermal energy produced will be the most suitable niche markets in the shortand mid-term. 2001 Elsevier Science Ltd. All rights reserved.
flat plate solar collectors that use PV cells as an integral part of the absorber plate. Research into utilizing waste heat from photovoltaic modules is more than 30 years old. At its essence is the desire to find an elegant solution to increase system efficiencies, while sharing balance-of-system (BOS) costs, and minimizing the size of solar systems (where there are issues of limited roof space). Commercially, PV cells have conversion efficiencies of between 6 and 15%, depending on various technologies. At one sun (1 kW/ m 2 ), this means that 850–940 W/ m 2 is lost as sensible heat or re-radiated into space. The main arenas of research of PV cogeneration thus far have been in PV/ T modules, ventilation systems designed to remove hot air from the PV modules, and specifying light transmission in PV panels in order to produce varying levels of daylight penetration. The evaluation of the research and designs of various PV cogen system types can bring a better understanding of the possibilities for synthesizing these efforts. The complex load profiles and economics in the built environment are primarily being met by custom architectural and engineering design. The widespread adoption of PV cogen
1. INTRODUCTION
PV cells convert a small portion of the incoming solar insolation into electricity. Innovative designs can produce secondary and tertiary energy production from PV modules. There has been research conducted into the various useful energy forms that these by-products can take as cogeneration. These multi-functional solar systems, hereafter referred to as PV cogen systems, can increase the economic viability of building integrated photovoltaic (BiPV) systems. However, the economic and energetic benefits are seldom identified as those of a PV cogen system. Thus, they are not properly designed or evaluated. This paper aims to improve the design and evaluation of PV cogen systems through a state-of-the-art overview and evaluation process. The term PV cogen will be used in this paper to describe any system that uses by-products or unused solar resources from photovoltaic processes. The term PV/ T refers to
†
Author to whom correspondence should be addressed. Tel.: 161-2-9300-8191; fax: 161-2-9385-6374; e-mail: [email protected] 57
58
M. D. Bazilian et al.
systems will require certain technological advancements, outlined later, as well as modular solutions. Installations that have a need for all of the thermal energy produced will be the most efficient.
1.1. Effects of temperature on PV cells The degradation of PV power output with increased temperature is well understood, and will play a role in the design of any PV cogen system. Mono-crystalline and poly-crystalline silicon cells have a negative temperature coefficient of approximately 0.4% / 8C (Wenham et al., 1998). Heat transfer from the modules can be accomplished through radiation, conduction, or convection. Each of these processes lends itself to a specific type of PV cogeneration system. Removing heat from the PV cells also reduces thermal cycles and stresses. 2. APPROACHES
PV cogen systems have been applied at every level of the built environment, from residential to industrial. The economics of the installed systems will be best where there is a need for heating or hot water, or where energy costs are high. Specific climatic niche markets have been identified in some studies (Bloem and Ossenbrink, 1995). Others will be identified here based on a study by Ecofys (Leenders et al., 1999). For the purpose of evaluation, PV cogen systems have been classified into four primary types.
2.1. PV/T modules PV/ T modules, in general, place PV cells on the absorber plate, or act as the absorber plate of standard solar thermal collectors. In this way, the waste heat from the PV panels is directly transferred into air, water, or phase-changing liquids that can be used for a plethora of building or off-site uses (Fig. 1). The synthesis of solar thermal collectors and
Fig. 1. Schematic diagram of Solarwerk’s Spectrum PV/ T module.
solar electric modules has certain advantages over placing the two systems next to each other (Ricaud and Roubeau, 1994 IEA PV/ T Workshop). • Cost savings • Installation cost savings. Since two or more processes of energy production are included in one module, the size of the overall system will be reduced. This can reduce installation time and mounting equipment necessary. • Space saving. Although this is not an issue at present in much of Europe, Australia or North America, there is a space consideration in Japan. As PV technologies further saturate the market, this could prove important in the future. • Balance of system (BOS) savings. Again, since two or more forms of energy can be produced, the BOS components can be shared. PV cells are currently the most expensive component in PV/ T systems. As the price of PV technology continues to decline, the cost savings in the areas previously listed will have a larger impact. Although several water-based PV/ T modules are available from Israel and Germany, there are very few commercially available.
2.2. Use of air flow behind the PV modules The removal of waste heat through the use of convective airflow behind the PV panels has also been researched (Garg and Adhikari, 1998). This approach has been used commercially either by removing heat from roof-mounted or fac¸ade systems. Natural convective flows can be used, or various mechanized ventilation systems utilizing fan-powered pressurization can be installed to move the air. The thermal energy can also be transferred to another medium, such as water. During times of cooling loads, the hot air can act as the basis for a stack effect to remove unwanted hot air. A stack effect can be created by the use of either active or passive systems. In some of the installations studied, the use of flues or chimneys was created between the PV wall and the exterior wall of the fac¸ade to accommodate this effect. It is most beneficial to the economics of the PV cogen system if natural convective cycles can be set up rather than using fans and motors (Fig. 2). The use of double skin facades in both new construction and renovations has helped spur several research projects and installations. The recently finished BP Amoco wall in Trondheim, Norway, will be an excellent source of data (Aschehoug et al., 2000). Rain screen cladding is
Photovoltaic cogeneration in the built environment
Fig. 2. Schematic of the PV cogen at the Ispra JRC test facility (H. Bloem).
another building material devised for use in fac¸ade renovation projects of older concrete buildings (BP Solarex). Although it provides a space for ventilation to cool the panels, the air cannot be harnessed for indoor heating.
2.3. Daylighting applications Many recent BiPV applications have incorporated the benefits of daylighting into their designs. Electric lighting is one of the largest sources of electrical use in most typical commercial buildings. Major studies have been conducted showing the quantifiable gains of daylighting to increase productivity in the workplace while also assisting in lowering energy bills (PG&E). PV panels are easily adaptable for light transmission. Thus, the PV cogen system requires no additional system cost, except for ambient light sensors to optimize the gain from daylighting (Fig. 3). This approach is classified as cogeneration in order to underscore the importance of utilizing an otherwise lost resource in the production of PV electricity. This type of PV cogen system has far more current installations than the other two categories. It is important to qualify these installations as PV cogen and thus receive the economic benefits of daylighting towards the cost of a BiPV system. Not only will it help the economics from a simple payback or life cycle cost analysis, but will allow PV cogen systems to join the current growth in daylighting applications and funding. Likewise, the use of PV modules for shading applications
Fig. 3. Innovative BiPV corridor, using daylighting and ventilation possibilities (Bear Architects).
59
has met with good success commercially. As this application requires only innovative mounting solutions, the additional costs for shading are minimized. The use of daylighting can be achieved by specifying light transmission through BiPV panels. This can be done in either amorphous or crystalline module types. Crystalline or poly-crystalline cells can be spaced apart in a module to cover nearly any desired percentage of glazing. Laser grooving can be applied to amorphous panels in a wide variety of patterns. The majority of these modules are custom products and are, therefore, very expensive (Figs. 4 and 5). Testing and modeling of these systems can be done in a number of ways. Well-accepted daylighting algorithms are included in software such as Radiance. Architectural scale mock-ups and three-dimensional modeling tools can also give valuable insights. The trade-offs between the various types of assessment have been explored (Lien and Hestnes, 2000)
2.4. The multifunctional fac¸ade The notion of the multifunctional or smart fac¸ade is relatively new. The term has been used to describe a building fac¸ade or roof that performs a number of energetic tasks. Strictly speaking, any building envelope is multifunctional. It serves as a weather protector, a fire retardant, a source of light, ventilation, security, and the like. The new multifunctional solar fac¸ade or roof can incorporate PV cogen systems that can produce heat, light, and electricity. These systems may also be separated or not used at all depending on the building load and optimization studies. A recent study conducted in Finland showed the maximum electricity benefit for a south-facing fac¸ade in northern European conditions (Vartiainen et al., 1999). Although the study used the PVs solely for electrical production, the synthesis between the transparent glazing and the PV modules was optimized for electrical production. This is a very
Figs. 4 and 5. Two patterns for allowing light through amorphous PV modules (EPV).
60
M. D. Bazilian et al.
positive step in intelligent building integration. Any multifunctional solar fac¸ade must optimize a tremendous number of variables (Peippo et al., 1999). Optimized fac¸ade design is therefore at the heart of whole building design. 3. RESEARCH
Research activities in PV cogen date back to the late 1970s. Some of the primary research objectives are listed below. This list is used only to give a cursory idea of the more current research trends in PV cogen. (Specific authors are not cited in many cases because the list is necessarily a generalization.) • Thermodynamic modeling and simulation of PV/ T modules. This research created various mathematical models to predict energy production, fluid flows and the like. • Design considerations of PV/ T modules. This research includes specific flat-plate collector design, glazing options, and even mounting and installation issues. • The use of thin films on PV/ T modules. This research is mostly in the area of amorphous silicon and will be covered in its own section. • Optimizing useful energy in multifunctional fac¸ades. These important models seek to maximize the energy output of a building element, while taking into account building design and occupant comfort (Peippo et al., 1999). • Fluid flow modeling behind PV arrays. This kind of investigation seeks to quantify and / or maximize the convective cycles present in the waste heat of PV modules in fac¸ade application. • Comparisons of single and double pass flatplate collector types for PV/ T hot-air applications. • PV/ T building integration. This research has been conducted within a BiPV framework. This includes economic analysis, building material credits as well as mounting and installation. • The use of concentrating troughs with PV/ T modules. There have been a few papers looking at CPC troughs and Fresnel lenses. Most of the models used silicon cells. Some modeled systems that use GaAS and other III–V combination cells. • Combined efficiencies of PV/ T modules. These studies looked at various aspects of efficiency ratings for PV/ T modules (Luque and Marti, 1999). Academic and government funding has sponsored
the bulk of the research in PV cogen systems. This is typical in cutting-edge technologies with small market realization. The majority of these government-funded projects have been in the EU (Joule and Thermie), the USA (PV:BONUS), and Japan (NEDO). Gregory Kiss produced an economic and energy study for NREL (Kiss and Kinkead, 1995) that compares a series of BiPV options for a model building. It concluded that the heat recovered can be up to 20 times the total amount required by the building in a climate like Oakland, CA, USA. Even in a cold climate, like Chicago, IL, USA, the system produced up to three times the building’s total heat requirement. The simple payback, calculated in years from five different geographic locations, declined from nearly 70 years with only the electrical production included, to 57 years when the thermal benefits were included, to 44 years when the thermal and daylighting benefits were included. The paper shows that PV cogen systems can greatly help the economic and energetic appeal of BiPV systems and should be considered. Research on PV/ T modules mainly focuses on the integration of PV cells and solar hot-water heaters for tap-water heating. Integration will yield economic and aesthetic advantages, but will reduce system efficiency. System efficiency of PV/ T systems will decline due to the need of PV modules to have low operating temperatures for high electrical yield. The solar water heater, on the other hand, requires a high operating temperature for high thermal production. Thus, the PV/ T collector will always be a compromise in operating temperature and, therefore, have lower energy yields than the two systems installed separately side by side. The technological challenge, as opposed to market interests, of improving PV/ T systems is the impetus for most research. A substantial amount of research has been conducted to evaluate the airflow behind PV panels. In Japan and North America, this has led to market-ready products and systems, e.g., OM Solar from Sora designs and Solarwall from Conservall. In Europe, this has not been the case. Some projects have been built, such as the Mataro library outside Barcelona (Lloret et al., 1995), and the Elsa building in Ispra (Clarke et al., 1995). Research is being conducted at the JRC in Ispra, and at the Universities of Cardiff and Strathclyde in the UK, among others (Brinkworth et al., 2000). There has been no market penetration of commercially available products however.
Photovoltaic cogeneration in the built environment
The fundamental reason appears to be the differences in climate and insolation. During winter, large areas of Europe are marked by low levels of insolation and low temperature, whereas large areas of Japan and the USA have high insolation and low temperatures that are ideal for this system type. There has been some success with PV air ventilation systems in the recent PV:VENT and INNOPEX projects in Denmark that might lead the way for commercial products in the near term (Pedersen, 2000). Daylighting, multifunctional fac¸ades, and water-based PV/ T systems are applicable in almost any climate, although the specific building functions may differ. 4. AMORPHOUS SILICON IN PV/ T
Research has been aimed at the possibilities for a new generation of PV/ T modules utilizing amorphous silicon cells as the absorber plate (Fig. 6). A study (Ricaud and Roubeau, 1994) stated that ‘ . . . stabilized a-Si:H PV modules behave as if they had a positive temperature coefficient of 0.2% / 8C’. Other studies have agreed that, when heated, a-Si:H modules show an annealing effect that results in a decrease of the Staebler-Wronski effect (Affolter et al., 1996). Amorphous cells thus appear to have excellent heat characteristics that are ideally suited for PV/ T modules. One IEA PV/ T study concluded that there will be a market for PV/ T in the 10 MW range by 2005. It found that amorphous silicon could be ideal for PV/ T use if its efficiencies reach 10% (Affolter et al., 1997). Another benefit of amorphous silicon is that it can be deposited directly on large surfaces and a variety of substrates (Platz et al., 1997). This can lower production costs. The lower cost of a-Si:H modules is, of course, a significant factor as well. However, the currently low efficiency levels (4– 7%) commercially available can detract from
Fig. 6. Temperature cost graph of amorphous vs. mono-crystalline PV cells (Affolter et al., 1997).
61
economic scenarios for widespread use. Very different performance characteristics were witnessed between certain manufacturers, which will make commercial viability even more challenging (Affolter et al., 1997).
4.1. Collateral energy benefits and distributed generation Collateral energy benefits include environmental externalities that are not normally taken into consideration in cost analyses. These energy benefits can be multiplied if the benefits of distributed generation are taken into account. The offset of greenhouse gases, capacity credits, reduced ohmic and reactive power losses, improved reliability, and transmission and distribution savings can be significant (Wenham et al., 1998). Distributed generation is receiving a great amount of attention from the press and the utility industry at both global and national levels. Studying PV cogen in these terms will be useful to its market infiltration. It has the added benefits of reducing fuel consumption as well as electricity. The concepts and complexities of distributed generation are beyond the scope of this paper. 5. BUILT EXAMPLES
Innovations in PV cogen in the built environment should strive to (Leenders and van der Ree, 1999, IEA PV/ T Workshop); • satisfy the architectural design • satisfy the energy requirements of the building • satisfy the client • satisfy the end-user or occupants of the building • create a useful and delightful space to inhabit The Doxford Solar Office Building in the UK uses a ventilation system to remove waste heat from behind the BiPV system. It uses thermally broken, low level, inclined vents. The air can then be used for building needs or vented outside through operable windows. It has a 73-kW array. Specially designed fins placed on the roof help to decrease the negative effect of wind on the convective cycle (Fig. 7) (Lloyd Jones and Watts, 2000). The 25-m south fac¸ade of the Elsa building in Ispra is covered with 505 m 2 of amorphous PV modules rated at 21 kWp. There is an air gap behind the PV array that collects hot air being used in an integrated ventilation system. A series of ‘chimneys’ have been constructed behind the fac¸ade to provide an area for convective air flow. This can charge the mass of the structural wall system as well (Wouters et al., 1998).
62
M. D. Bazilian et al.
Fig. 7. Doxford building (A. Lloyd Weber). Fig. 8. Applebee’s PV cogen schematic (K. Sheinkopf, 1997).
The Test House in Japan was constructed to study the use of PV/ T modules in residential applications. Roof space in Japan is becoming scarce. The system consists of 24 PV/ T panels that produce 3.2 kW electric and 25 kW thermal. The collector area is 50 m 2 (Strong, 1995). The BiPV entryway to the Thoreau Center for Sustainability building in San Francisco shows one of the countless options for daylighting. Crystalline cells have been spaced apart in order to allow light to wash the entry space. Architecturally interesting spaces can be created and possibilities for overheating can be diminished. This type of application is ideally suited to unconditioned spaces, but can be used in conjunction with insulated glass for building skin applications (http: / / www.thoreau.org). The Scheidegger Metallblau building in Switzerland uses PV modules as shading devices, as well as having hot air collected behind the fac¸ade. It has an 18-kW PV fac¸ade with 12 kW of thermal power. The measured efficiency of the combined system is about 40%. The system was designed by Atlantis Energy (Posansky et al., 1994) The Fenster AG factory building, which is also in Switzerland, uses the waste heat collected from its rooftop monitors for the factory’s energy needs. It consists of a 9-kW vertical PV fac¸ade and a 53-kW thermal cogen PV system integrated into the skylights. The thermal energy can be stored in the concrete slabs of the basement during the summer. The system was also designed by Atlantis Energy. In the PV:BONUS Program project at a North Carolina Applebee’s, hot air taken from behind the PV panels is transferred to hot water, which can be used in the restaurant. The roof has an integrated 1.6 kWp PV system. The cost of the PV cogen system was calculated to cost US$54.00 / m 2 more than a conventional roof (Sheinkopf, 1997; Fig. 8). Two projects named Brig and Rigi were installed in the residential sector by Atlantis Energy in Switzerland. The ‘Brig’ hybrid shingle roof has
a peak power of 15 kW electric and 30 kW thermal. The thermal energy is used to heat the shower rooms of the facility. The ‘Rigi’ system has 8.4 kW electric and 12 kW thermal and is also integrated into a typical shingled roof. The rooftops have an air gap behind the BiPV array and the aim is to maximize the flow of natural convection. The air can be used inside, vented out through a ridge vent, or stored. The simplicity of these systems is what is most appealing. A system of this type can use traditional construction techniques without resorting to additional hot air collector systems. These have been in place since 1993 (Posansky et al., 1994). Another good example of PV cogen in the built environment is the Mataro Public Library in Barcelona. Multifunctional PV panels are integrated into the south fac¸ade. The system was designed to produce an optimal PV and thermal output. A ventilation system is used to move the hot air from behind the modules. Some of the modules are semi-transparent, to allow for natural-light penetration (Lloret et al., 1995).
6. PRODUCTS
The growth of the BiPV market and its influence on innovative design and construction integration components has spurred along the growing interest in PV cogen. PV/ T systems are mostly modular and can eventually be mass-produced. Custom architectural products for use with BiPV systems will be more difficult to massproduce. Esbensen Consulting produced an extensive list of existing systems as part of the IEA PVPS Task VII work (Soerensen and Munro, 2000). It shows that there are three currently available PV/ T systems using domestic hot water product technologies, and one using conventional air-collector technology. The products’ actual commercial availability is difficult to ascertain in some instances.
Photovoltaic cogeneration in the built environment
• Up until four years ago, SunWatt Corp., a small American company, produced a concentrator-type PV/ T module. They have ceased production of this unit due to the inability to buy the proper fin extrusions. They are producing a prototype PV still product that uses the evaporation of seawater to cool solar cells. This product will probably be aimed at the developing world. • The Chromagen ‘Multi Solar System’ is sold at roughly US $1940 / m 2 . It uses solar cells integrated into the non-spectrally selective absorber plate. The company reported that although the product is still under testing, units have been sold and are presently commercially available (www.chromagen.co.il) (Elazari, 2000). • The Solarwerk product has been named ‘Spectrum’. It has a peak output of 250 W, a thermal efficiency of 63%, and an electrical efficiency of 12%. It has 180 PV cells mounted onto the absorber plate. The module is sold with a 2.2-m 2 collector area (Solarwerk). • The OM Solar product is more of a wholebuilding approach to residential energy needs than a product per se. The system collects solar-heated air from underneath the roof and channels it either outside, in times of cooling, to a heat-storing concrete slab, or directly into the building, for heating. They operate as designers and contractors of this system in Japan and now in America (Sora Designs). The Task 7 list showed 28 products that were under research or production in its last edition. A number of these are projects that have used custom PV cogen systems. Most are in product development at various stages. Another interesting product is the Conservall PV Solarwall. The current commercially installed PV Solarwall product uses PV power to run ventilation fans that move hot air from the Solarwall product for building pre-heat. The PV panels get cooled by the removal of heat behind the metal cladding, thus, it can tentatively be called PV cogen through a mutually beneficial relationship. Large glass manufacturers, like Flabeg, previously known as Pilkington (DE) and Viracon (USA), are working with PV manufacturers to create modular and custom solutions for architects and engineers. High-performance glazing can be specified for spectral sensitivity or U-value, in addition to the integration of PV cells. • A member of the Flabeg Group has had the OPTISOL PV glazing system commercially
63
available for BiPV applications for nearly a decade. • BP Solarex is in testing and product design of their Powerview glass product, which will be an architectural glass product with integrated PV cells. • Various fac¸ade contractors, e.g. Kawneer (CA), Schuco (DE), and Oskomera (NL), have developed PV fac¸ades. Architectural glass panels use current technologies and understood construction practices. This is a key factor for dissemination of any BiPV system. 7. AREAS FOR FURTHER INVESTIGATION
The groundwork has been set for the commercial viability of PV cogen systems. There is still an ample amount of worthwhile directions for further study. • PV cogen systems need considerable additional real-world testing to peak interest in the market. • Long-term testing of the temperature effects of heat on a-Si cells and crystalline cells integrated on absorber plates, or acting as them, will be necessary. Stagnation temperatures in PV/ T hot-water systems can reach 1608C (Affolter et al., 1997). • Standardized methods for efficiency ratings of combined systems are essential for comparisons. • Recent studies of rural health clinics in the developing world have shown the need for both electricity and thermal energy (Olsen and Jimenez, 1998). ‘ . . . the World Health Organization has been looking for ways to use renewable energy technology for a broader range of applications . . . ’. These broader applications could use waste heat from PV modules or PV/ T modules for use in the sterilization of medical equipment. Heat energy can also be used for pasteurization, water distillation, and, of course, water heating. Systems should be first tested in easily controlled facilities before being exported. • Seasonal heat storage at low temperature needs additional research and monitored built examples. • Concentrating systems can be an attractive option since they can effectively increase cell efficiency. They can thus benefit from a cogen system to remove large amounts of excess heat (Karlsson and Wilson, 1999). • Selectively coated mirrors can be used to
64
M. D. Bazilian et al.
Table 1. Evaluation criteria Criteria
(unit)
Time to market Market potential Investment Building integration Thermal performance Electrical performance Energy consumption Sustainable building Life time Effect on energy performance
year m 2 / yr Euro / m 2 Euro / m 2 GJ / year?m 2 kWh e / year?m 2 kWh e / year?m 2 2 year 2
optimize the incoming light energy. Light with wavelengths longer than the band-gap of the PV cells can be focused on thermal production (Yang et al., 1997). • There is a need for more economic analyses of PV cogen systems. These will be necessarily complex because there is more than one type of energy being produced. It will have to be completed within the context of a whole-building perspective, looking at the entire energy flow. • Optimization tools that can combine the advantages of BiPV along with fluid dynamic or thermodynamic evaluation will be needed to quantify results and aid in system design. • Market size and type studies. Finding niche markets in various climatic situations will be key in the short-term. Research in PV cogen systems has proved to be a positive environment to stimulate dialogue between experts from the solar thermal and solar electric fields. These researchers do not usually coordinate their efforts and have thus lost out on valuable insights. In order for any mass-market inclusion of solar energy systems, there will need to be continued conversations between these parties. 8. EVALUATION METHOD
The evaluation method originated from a growing need in The Netherlands to evaluate and compare various PV cogen systems that were applicable under Dutch conditions. The methodology begins with a definition of the most
important criteria to evaluate a promising hybrid PV/ T system (Table 1). These ten criteria were used to validate eight promising hybrid PV/ T systems. The systems were defined and selected by a panel of experts. The proposed systems are believed to be the most promising systems using today’s technology. They are applicable in moderate northwest European climates and are technically feasible within 5 to 10 years in new building projects. The goal of the study is to show the real technological and market value of the defined systems. Daylighting systems are not examined within this construct, only PV and thermal combinations are assessed.
9. EVALUATED HYBRID SYSTEMS
During the selection of the hybrid systems to be evaluated, priority was given to closely matching supply and demand of heat since the usefulness of the produced heat is the crucial factor in determining the worth of PV/ T systems both economically and energetically. A wide variety of systems types were evaluated. In all cases, the electricity was assumed to be fed to the grid as an ideal storage source. The following eight systems were assessed for applications in The Netherlands:
9.1. PV building fac¸ade Building-integrated PV in fac¸ades offers a relatively simple opportunity to utilise the heat generated in the PV panels, as previously noted. The PV fac¸ade acts as an unglazed PV/ T air collector, which supplies natural ventilation in summer and pre-heated air in winter. In spring or summer, either ventilation or pre-heated air can be supplied, depending on the climate. Applications are foreseen in both industrial and commercial buildings. The Solarwall is a practical and marketready product of this type of system (Fig. 9). It has been primarily targeted at the industrial sector where aesthetic concerns are not high, comfort levels for occupants are not strict, and pre-heating of ventilation air is of some importance. Sound
Fig. 9. Unglazed PV/ T air collector as a ventilated PV fac¸ade in an industrial building.
Photovoltaic cogeneration in the built environment
65
Fig. 10. Unglazed PV/ T air collector for drying tulip bulbs. Fig. 12. Unglazed PV/ T water collector for pool heating.
PV/ T integration into buildings is more difficult to accomplish as the thermal comfort requirements are relatively severe. However, there is ample evidence of this type of system in the market, as previously described. These applications can also take economic advantage of displacing expensive traditional fac¸ade elements. The use of PV systems to reduce heat gains by shading are not included in this evaluation method.
9.5. Heat pump I The pre-heated air of the unglazed PV/ T collector supplies the heat pump with its heat source. The heat pump upgrades the pre-heated air to low temperature space heating. Additionally, a heat recovery unit is installed to reduce heat losses. This system has been demonstrated in Zwaag (NL) since 1998 (Fig. 13).
9.2. Biomass dryer An unglazed PV/ T collector can be used to dry biomass (e.g. tulip bulbs and woodchips). Not all industrial drying processes can be utilised by solar energy due to the relatively low energy efficiency and low temperatures (Fig. 10).
9.6. Heat pump II
An unglazed PV/ T collector can pre-heat the cold water supply (108C). During the summer, auxiliary heating is not necessary (Fig. 12).
Here, an unglazed PV/ T collector is combined with a heat pump and aquifer. In summer, the PV/ T collector is cooled with 5–108C water from the aquifer. While cooling the PV/ T collector, the water is heated to about 208C and is stored in the aquifer to be used as a heat source for the heat pump in winter. In winter, the heat pump upgrades the stored heat for low temperature space heating to about 408C. [During this process, the aquifer is again fed with cold water (5–108C)]. This system offers opportunities to regenerate the heat in the soil when heat pumps are used on a large scale in urban areas and is now being tested (Fig. 14).
Fig. 11. Glazed PV/ T water collector for pool heating.
Fig. 13. Unglazed PV/ T air collector combined with a heat pump and a heat recovery unit.
9.3. Indoor swimming pool To heat the water of an indoor swimming pool to a maximum of 308C, a simply covered PV/ T collector will suffice (Fig. 11).
9.4. Indoor /outdoor swimming pool
66
M. D. Bazilian et al.
Fig. 14. Unglazed PV/ T water collector combined with a heat pump and aquifer.
9.7. Solar hot-water system A glazed PV/ T collector combines directly the functions of PV (electricity) and a domestic hotwater system (hot tap water). The PV panel acts as an absorber in the thermal collector (Fig. 15).
9.8. Solar hot-water system and space heating This is identical to that described in the preceding paragraph, but now the PV/ T collector also supplies heat for space heating, meaning a larger roof area needs to be covered with PV/ T. 10. RESULTS
The systems were evaluated against the ten criteria previously mentioned. The effects of each criterion on the individual systems was quantified for the Dutch circumstances. This resulted in an extended and quantified matrix, with the upper row listing the systems, and the left column listing the ten criteria. Some highlights are listed here. It should be noted that markets other than the Dutch would show different results, although the methodology can be applied in any number of markets.
10.1. Market potential The research showed that the maximum possible market potential in The Netherlands is the largest for the heat pump systems (about
2 ? 10 6 m 2 / year). The second largest market potential exists for the solar water heaters (about 1 ? 10 6 m 2 / year), followed by the PV fac¸ade (about 0.3 ? 10 6 m 2 / year). The swimming-pool applications seem to be a nice ‘niche’ market (about 2–7 ? 10 3 m 2 / year); investments are low and revenues relatively high thanks to low operating temperatures. These systems must compete with other energy systems, including conventional gasfired systems. The real market potential will thus be lower than the maximum market potential described herein.
10.2. Time to market Concerning the time-to-market, the PV/ T combination of PV and solar water heaters is most advanced. The time to market demonstration is zero (PV/ T technology is available), and full market introduction may follow in about 2 years time. In the meantime, experiences and learned lessons can be evaluated and processed. Perhaps the same absorber can be used in unglazed systems, like the swimming-pool applications. The PV fac¸ade is also in the market introduction phase. Performance data and cost / benefit analyses in different climates must prove its validity. The time to full market introduction is about 2 to 3 years. The heat pump systems are still in a premature development phase; new technologies have to be tested and experimented upon (now under development in The Netherlands). The time to market demonstration and the introduction of the heat pump systems will take the longest time.
10.3. Simple pay-out time
Fig. 15. Example of a PV/ T water collector.
Using the data on investment costs, saved materials, electrical and thermal revenues, and their benefits, the simple pay-out time (SPOT) of the various systems could be calculated (Tables 3 and 4) and compared to those of a traditional PV
Photovoltaic cogeneration in the built environment Table 2. Simple pay-out time (reference)
d 5 Combustion efficiency (0.85)
Fac¸ade
Roof
p-Si a
a-Si b
p-Si
a-Si
BIPV (Euro / m 2 ) Yield (kWh / m 2 ?year) Benefit (Euro / m 2 ?year)
660
450
570
430
55
35
80
50
6
4
9
SPOT (year)
110
a
67
112
e 5 Heat of combustion (35 MJ / m 3 ) f 5 Gas price (0.2 Euro / m 3 ) 11. CONCLUSIONS
5.5
63
78
100 Wp / m 2 ; b 60 Wp / m 2 .
panel (Table 2). This calculation is performed for both the p-Si and a-Si PV panels (cost estimation is for the year 2000). Calculation of the SPOT (year) 5 (Co /Es 1 Ts) Co 5 Costs (Euro / m 2 ) 5 Product investment 1 Costs of system integration 2 Material savings Es 5 Electricity savings (Euro / m 2 ? year) 5 a ? b a 5 Net electricity yield (kWh / m 2 ? year) b 5 Electricity price (0.1 Euro / kWh) Ts 5 Thermal savings (Euro / m 2 ? year) 5 (c ? f ) /(d ? e) c 5 Thermal yield (MJ / m 2 ? year)
Using the waste heat and available daylight from PV modules is an elegant idea, and has been successfully installed and operated in a number of commercial and industrial facilities. Using a comprehensive term for its description, like PV cogen, could prove to be a useful addition to the vocabulary of building professionals. From the technological perspective, PV/ T systems are especially suitable for low temperature. For medium-temperature applications, the thermal and electrical yield of the hybrid system is lower than that of two separate systems. The combination of the two will always be a compromise. Hence, a PV/ T system is economically viable when the costs of the reduced energy performance matches the gained costs of production, installation, and mounting. Apart from the economic motive, the uniform appearance of a PV cogen system may provide an important surplus value in terms of enhanced architectural aesthetics. A PV cogen system will also serve to meet the consumer’s desire for solar energy to meet both thermal and electrical loads. This is difficult however since the supply and demand of electrical and
Table 3. Simple pay-out time (a-Si) a-Si
Reference Fac¸ade
Costs 450 (Euro / m 2 ) Electricity 4 saving (Euro / m 2 ?year) Gas saving (Euro / m 2 ?year) SPOT 110
Swimming pool
DHW
Roof
PV fac¸ade
Industry Biomass dryer
HP I (HRU)
HP II (soil)
In
In and out
Tap water
Tap and space
430
220
420
590
645
515
425
680
645
5.5
22.5
3 1.5
80
Domestic
49
5
4.5
5
2
2
170
2
2
4.5
5
4.5
4
5.5
6
7.5
4
52
39
57
80
Table 4. Simple pay-out time (p-Si) p-Si
Reference Fac¸ade
Costs 660 2 (Euro / m ) Electricity 55 saving (Euro / m 2 ?year) Gas saving 6 (Euro / m 2 ?year) SPOT 110
Industry
Domestic
Swimming pool DHW
Roof
PV fac¸ade
Biomass dryer
HP I (HRU)
HP II (soil)
In
In and out
Tap water
Tap and space
570
0–445
560
730
830
655
560
865
830
80
5.5
1.5
9
1.5
4 102
65
0–64
8
8
7.5
8.5
7.5
7.5
2
2
5
5.5
7
3.5
2
2
52
40
60
75
68
M. D. Bazilian et al.
thermal energy are rarely properly matched. Continued research and information exchange between PV and solar thermal experts are necessary, and must occur at an international level. The glazed collectors in swimming-pool applications have the shortest pay-back period through the efficient and cost-effective use of low-temperature heat. The PV/ T fac¸ade might be an attractive niche market as a PV/ T fac¸ade features architectonic possibilities and can displace building materials. Economic and technical performance are strongly related to climatic conditions, building energy loads, system energy production, and the amount of material saved. Domestic hot water PV/ T systems are currently not profitable. However, this product could have large market volumes and is expected to emerge to a mature market within a relatively short period of time, as interest in BiPV and thermal systems in the domestic market is growing fast. The biomass dryer has the longest pay-back period. This is caused by the high electricity consumption of the fan needed to maintain sufficient air flow to dry the biomass. The pay-back period of the heat pump systems was not calculated as these systems yield heat that cannot be used directly. The heat of the PV/ T collector serves as a heat source that must be upgraded by a heat pump to match the real energy demand. Careful design and the education of architects and engineers can foster the use of PV cogen techniques. The combination of PV in daylighting systems or multifunctional fac¸ades was not included in the evaluation, but its advantages are clear. Specifying for daylighting and ventilation schemes in high quality architectural BiPV installations will require the cooperation of many building design and system experts. Daylighting should continue as the most likely mode for PV cogen in the near term in the commercial market. Heat-storage systems could greatly aid the economic viability of PV cogen by more closely matching production and demand, or by being utilized at peak times. PV cogen should be researched and installed in a holistic manner. By taking into account the benefits of electrical, thermal, and daylight production systems, they can appear more attractive, both economically and environmentally. With the growth and interest in BiPV and, more generally, in distributed energy systems and cogeneration, PV cogen could play a large role in future PV installations at every level of the built environment. Acknowledgements—The authors wish to thank: Novem, the
Netherlands Agency for Energy and Environment B.V., for financing part of the performed work. TNO, the Netherlands Organisation for Applied Scientific Research, and ECN, The Netherlands Energy Research Foundation are thanked for their contributions in developing the evaluation method. Thanks are also due to the National Solar Architecture Research Unit, the Australian CRC for Renewable Energy and the Faculty of the Built Environment at the UNSW. A new IEA working group on combined PV/ T concepts is in the process of compiling an extensive resource list on this subject. It will be made available on the group’s web site in due course at http: / / www.task7.org / pvt / pvt home.html ]
REFERENCES Affolter P., Haller A., Ruoss D. and Toggweiler P. (1996) Absorption and high temperature behaviour evaluation of amorphous modules. In Report for Project 56360 /16868 for the Swiss Federal Office for Energy. Affolter P., Gay J., Haller B. A., Althaus H., Ruoss D. and Toggweiler P. (1997) A new generation of hybrid solar collectors. In Report for Project 56360 /16868 for the Swiss Federal Office for Energy. Aschehoug O., Hestnes A., Matusiak B., Lien A., Stang J. and Dagfinn B. (2000) BP amoco solar skin. In Proceedings for EUROSUN 2000, June 19–22, Copenhagen, Denmark. Bloem J. J. and Ossenbrink H. (1995) Thermal aspects of PV integration in buildings. The 13 th European PV Solar Energy Conference, Nice. Brinkworth B. J., Marshall R. H. and Ibarahim Z. (2000) A validated model of naturally ventilated PV cladding. Solar Energy 69(1), 67–81. Clarke J. A., Johnstone C., Strachan P., Bloem J. J. and Ossenbrink H. (1995) Thermal and power modelling of the photovoltaic fac¸ade on the ELSA building, Ispra. The 13 th European PV Solar Energy Conference, Nice. Elazari A. (2000) Building integrated multi pv / t / a solar system roof tile, Eurosun 2000 Preprint, June 19–22, Copenhagen, Denmark. Garg H. P. and Adhikari R. S. (1998) Transient simulation of conventional hybrid photovoltaic / thermal air heating collectors. International Journal of Energy Research 22, 547– 562. Karlsson B. and Wilson G. (1999) MaReCo–CPC for high latitudes. Proceedings ISES 1999 World Congress, Jerusalem. Kiss G. and Kinkead J. (1995) Building integrated photovoltaics: a case study. In NREL Report TP-472 -7574. Leenders F. et al. (1999) Technology review on PV/ T systems, Ecofys no. E21036. Utrecht, The Netherlands. Leenders F. and van der Ree B. G. C. (1999) Photovoltaic / thermal systems. From Workshop on PV/Thermal Systems, 17–18 September, Amersfoot, The Netherlands. Lien A. G. and Hestnes A. G. (2000) Visual studies of transparent PV elements. In Proceedings for EUROSUN 2000, June 19–22, Copenhagen, Denmark. Lloret et al. (1995) The Mataro´ public library: a 53 kWp grid connected building with integrated PV — thermal multifunctional modules. 13 th European PV Solar Energy Conference, Nice, France, pp. 490–493. Lloyd Jones D. and Watts B. (2000) Effective use of building integrated photovoltaic waste heat: three projects. In Proceedings from the 2 nd World Solar Electric Buildings Conference, 8–10 March, Sydney. Luque A. and Marti A. (1999) Limiting efficiency of coupled thermal and photovoltaic converters. Solar Energy Materials & Solar Cells 58, 147–165. Olsen K. and Jimenez A. (1998) Renewable energy for rural health clinics. National Renewable Energy Laboratory Publishing, Golden, CO, USA. Pedersen P. (2000) Cost effective BIPV systems with combined electricity and heat production. In Proceedings for EUROSUN 2000, June 19–22, Copenhagen, Denmark.
Photovoltaic cogeneration in the built environment Peippo K., Lund P. and Vartiainen E. (1999) Multivariate optimization of design trade-offs for solar low energy buildings. Energy and Buildings 29(2), 189–205. Platz R., Fischer D., Zufferey M., Anna Selvan J. A., Haller A. and Shah A. (1997) Hybrid collectors using thin-film technology. In Proceedings 26 th PVSC, 30 September–3 October, Annaheim, CA, USA, pp. 1293–1296. Posansky M., Gnos S. and Coonene S. (1994) The importance of hybrid PV-building integration. In Proceedings of First WCPEC, 5–9 December, Hawaii, pp. 998–1003. Ricaud A. and Roubeau P. (1994) ‘‘Capthel’’, A 66% efficient solar module and the ‘‘Ecothel’’ co-generation solar system. In Proceedings of First WCPEC, 5–9 December, Hawaii, pp. 1012–1015. Sheinkopf K. (1997) PV system with thermal heat recovery. CADDET Renewable Energy Newsletter, ETSU, UK. Soerensen H. and Munro D. (2000) Hybrid PV/ thermal collectors. In Proceedings from the 2 nd World Solar Electric Buildings Conference, 8–10 March, Sydney.
69
Strong S. (1995). The Dawning of Solar Electric Architecture, NREL Publications, Harvard, MA. Vartiainen E., Peippo K. and Lund P. (1999) Daylight optimization of multifunctional solar facades. Solar Energy 6(3), 223–235. Wenham S., Green M. and Watt M. (1998) Applied Photovoltaics, University of NSW Publications, Sydney. Wouters P., Vandale L., Bloem H. and Zaiman W. J. (1998) Combined heat and power from hybrid photovoltaic building integrated components: results from overall performance assessment. Proceedings from the 2 nd World Conference on Photovoltaic Solar Energy Conversion, Vienna. Yang M., Izumi H., Sato M., Matsunaga S., Takamoto T., Tsuzuki K., Amono T. and Yamaguchi M. (1997) A 3 kW PV–thermal system for home use. In Proceedings 26 th PVSC, September 30–October 3, Annaheim, CA, USA, pp. 1313–1316.