Final Report EFP project 1713/00-0014 Photovoltaic/Thermal Solar ...

Final ReportEFP project 1713/00-0014

Photovoltaic/Thermal Solar Collectors andTheir Potential in Denmark

Miroslav Bosanac, Bent Sørensen*, Ivan Katic,Henrik Sørensen**, Bruno Nielsen**, Jamal Badran

Danish Technological Institute, Solar Energy CentreGregersensvej, DK 2730 Taastrup, tel. +4572202486, fax +4572202500, email:[email protected]

*Novator Advanced Technology ConsultingØstre Alle 43D, DK 3250 Gilleleje, Tel. +45 48361540, fax +4548361541, email:[email protected]

**Esbensen Consulting Engineers Ltd.Vesterbrogade 124B, DK-1620 Copenhagen V, Tel +45 33267304, Fax +45 33267301,email: [email protected]

Copenhagen, 21 May 2003

EFP reportCombined photovoltaic/thermal solar collectors and their potential in Denmark - 2 -

Table of Contents:

1. Introduction to solar co-generation.................................................................................. 42. Literature Study................................................................................................................. 5

2.1 Definition of PV/T collector designs ...................................................................................62.1.1 Water PV/T collectors ................................................................................................6

2.1.2 Air PV/T collectors ........................................................................................................73. Market survey of PV/T solar collectors ............................................................................ 8

3.1 Introduction ......................................................................................................................83.2 Overview of the survey.....................................................................................................93.3 Manufacturers of Commercial PV/T Collectors ................................................................10

3.3.1 Water PV/T collectors ..............................................................................................103.3.2 Air PV/T collectors...................................................................................................123.3.3 Conclusion................................................................................................................16

3.4 Existing Building Projects with PV/T Collectors ................................................................163.5 Research and governmental support .................................................................................203.6 Design Considerations .....................................................................................................22

3.6.1 General recommendations .........................................................................................223.7 Conclusions for chapter 3 ................................................................................................233.8 References.....................................................................................................................24

4. Initial calculations ............................................................................................................ 284.1. Spreadsheet model for PV/T collectors...........................................................................28

4.1.1 Performance in terms of energy and exergy ...............................................................285. Simulations ....................................................................................................................... 33

5.1 Collector parameters and characteristics...........................................................................336. Simulation of the PV/T Performance with TRNSYS ........................................................ 35

6.1. PV/T Collector for Water Heating ..................................................................................356.1.1. Theoretical Model of the PV/T - Collector.................................................................366.1.2. Simulation Program..................................................................................................426.1.3. PV/T Collector Efficiency Curves as a Function of Design Parameters.......................456.1.4. Annual Performance of Water-Heating PV/T Collector .............................................516.1.5. Summary................................................................................................................576.1.6. Verification of the Computer Program Accuracy......................................................58

6.2 PV/T - Wall....................................................................................................................606.2.1. Theoretical considerations ........................................................................................616.2.2. Simulation Program..................................................................................................626.2.3. System Annual Yield................................................................................................62Nomenclature ...................................................................................................................72

References ..........................................................................................................................747. Complete system modelling using NSES ........................................................................ 75

7.1 Model description............................................................................................................757.1.1 Reference year data .................................................................................................757.1.2 The solar collector ....................................................................................................777.1.3 Heat Exhange...........................................................................................................837.1.4 Flat-plate collector with heat storage ..........................................................................847.1.5 Solar heat-producing systems.....................................................................................877.1.6 Heat pump system....................................................................................................91

7.2 Results of simulations ......................................................................................................92


7.2.1 Software verification.................................................................................................927.2.2. Small PVT system simulation....................................................................................947.2.3. Medium-size PVT system simulation.........................................................................977.2.4. Large PVT system simulation................................................................................. 1017.2.5. PVT system with heat pump................................................................................... 102

7.3 References................................................................................................................... 1108. Conclusions ................................................................................................................... 111

8.1. Water-Heating PV/T collector ...................................................................................... 1118.2. Air PV/T collector (PV/T – Wall) ................................................................................. 113

9. Recommendations for future work ................................................................................ 114


1. Introduction to solar co-generation

In recent years building integration of photovoltaic modules has become more and more popular inthe industrialised part of the world, where national support programmes has accelerated thedissemination of grid connected PV systems. The installation of a building integrated PV (BIPV)systems has certain advantages compared to a traditional PV system mounted in a separatestructure. Function as a rain screen, sun shading device and a visually attractive cladding of thebuilding are the most popular “added values” the house owner gets from the BIPV system.

Traditional (silicon) photovoltaic modules will produce more electricity the cooler they are. Typicallypower increases with 0,2-0,5% per °C decrease of temperature, but when a PV module isintegrated in a facade or roof, it will normally get warmer than a module mounted in free air. It isthus logical to remove the excessive heat from the module, giving the reason for the growing interestin “solar co-generation” or photovoltaic/thermal (PV/T) collectors. If the surplus heat from the PVmodule can not only be removed, but be used to fulfil thermal energy needs of the building, an extraadded value could be achieved.

The status of commercial PV modules is that only 10-15% of the incident solar energy istransformed to electricity. The potential heat production from a given surface is thus much higherthan the electrical performance, but it is an open question if this heat can be used in a sensible way.There seem to be several obstacles:

- Most buildings need heating in winter when the solar gain is at its lowest- The heat is needed at a higher temperature than the surrounding air, leading to increased module

temperature unless a heat pump is used.- For heating of domestic hot water, a heat exchanger is needed.- The collectors could become very hot and thereby damaged if circulation of cooling media is

blocked.- The construction may be too complex and thus expensive compared to separate PV and thermal

collectors.

The obvious advantages are:

- The total area used to extract a given amount of electricity and heat may be smaller than fortwo separate systems

- The materials used for a PV/T plant, and thus the total energy and economy balance, may bebetter than for separate units.

- The roof or facade will have a more uniform look.

In Denmark there already exist a handful of BIPV demonstration projects, where PV/T solutionshave been integrated on an experimental basis. It is not yet proven that any of these demonstrationprojects will benefit from that, or they have been better off with separate systems. An urgent needtherefore exist to establish guidelines and calculation methods for future BIPV projects. It is thepurpose of this project to evaluate the feasibility of using combined PV/T collectors in typical Danishbuilding installations.


2. Literature Study

Several reports and papers exist within the field of PV/T collectors. The majority of the publishedpapers and reports are concerning theoretical developments of the efficiency of PV/T collectors. Inthe following only a few of these reports and papers are considered.

Norwegian PhD project [1].In this project an experimental model of a PV/T collector was constructed and measured. Themodel consisted of a conventional solar thermal collector with PV cells pasted to the absorber. Oneof the main conclusions from this study is that the thermal bonding between PV cells and absorberplate is very important for the thermal efficiency. The electrical efficiency is reduced when the flowrate of the cooling fluid is low, because the PV cells near the outlet then became relatively warm,and thus having a lower electrical output.

Report from IEA workshop [2].Several authors have contributed to this report, which also describe experimental results. The mainconclusion is that most PV/T collectors have same thermal characteristics than ordinary non-selective thermal solar collectors. In this report the evaluation of PV/T collectors is also discussed,and the exergy method is suggested as one of the ways to compare different constructions.

The authors of this report have adopted the method.

PV-Hyphen projectThis is a EU Joule project that has carried out extensive studies of PV/T systems, in particular PVmodules with ventilated air behind the PV cells for cooling the PV cells. The project involvedpractical measurements as well as theoretical work and simulations with the ESP-r programdeveloped for passive solar heating. Eight different (office) buildings at different locations in Europewere selected for a total simulation of energy usage. The conclusions were that in southern/midEurope, and in buildings with a modest fresh air demand, the savings are negligible. In NorthernEurope, and for buildings with a high and constant fresh air demand, the savings are up to 10% ofthe annual heating demand. The effect on the electricity production caused by cooling the PV cellsis very small.

The above mentioned references and others concerning PV/T collectors are listed in chapter 3.8


2.1 Definition of PV/T collector designs

PV/T collectors being considered in this report are collectors, which can provide both electrical andthermal energy. These hybrid collectors are divided in two groups:

Water PV/T collectors.One example is a conventional flat plate solar heat collector with integrated PV cells on theabsorber, to produce both thermal and electrical energy.

Air PV/T collectors.These can be façade or roof integrated PV cells with ventilation air passed behind or infront of the PV cells.

Furthermore, water PV/T collectors can be divided into groups according to temperature levels ofthe heat transfer fluid. This range from low-temperature applications for e.g. swimming pool andheat pump applications to medium-temperature applications around 55°C for e.g. domestic hot waterapplications. In the present study focus has been on the medium-temperature applications. In thestudy of F. Leenders et al. [21-22] further information can be found regarding the benefits andpotential markets for low-temperature applications.

2.1.1 Water PV/T collectors

For these systems, water is used as heat transfer fluid. The PV cells are pasted either directly onthe absorber or interior on a coverplate with a dielectric material. This means that the only contactbetween the PV cells and the absorber or the coverplate is a high thermal contact. The heattransfer fluid runs inside the ducts on the absorber and collects heat from the absorber. If the PVcells are pasted to the absorber, heat is also extracted from the PV cells resulting in a higherelectrical efficiency of the PV cells.

Useful thermal energy is extracted to one end of the ducts where it can be utilised. The ducts canbe coupled either in series or in parallel, which effects the efficiency of the system. The heattransfer fluid can be circulated by either a pump (a pumped system) or by the difference in specificgravity of the heat transfer fluid (a gravity system).

Figure 2.1: A typical water PV/T collector

PV cells

Absorber

Ducts for heat transfer

Coverplate


2.1.2 Air PV/T collectorsThe other type of PV/T collector is an air-based system. Instead of water, air is used as heattransfer fluid. The PV cells are either pasted to the interior of the coverplate or to an absorber orthe PV cells are acting as an absorber or coverplate itself. The air can be circulated by eithernatural ventilation or forced ventilation.

Figure 2.2: Example of an air PV/T collector

This kind of PV/T collector have been tried out in practice in several different projects and a fewhave been commercialised, e.g. the Canadian “PV SOALRWALL”. The PV SOLARWALLconsists of a perforated metal absorber in front of the exterior wall of a building with an airgab inbetween. In front of the perforated metal absorber a PV module is mounted is such a way that thecooler ambient air is allowed to pass behind the PV module. Heat generated by the PV module andthe metal absorber will be transferred to the air and thereby cooling the PV module causing a higherelectrical efficiency for the PV module. See also chapter 3.

Figure 2.3: Canadian SOLARWALL with PV cover, see also chapter 3.3.2 [2]

Coverplate

Airflow PV Cells

Absorber


3. Market survey of PV/T solar collectors

3.1 Introduction

A questionnaire was circulated to manufacturers, designers and research institutes in the field ofPV/T collectors. Purpose of this questionnaire was to collect the basic information about thesystems and products available or planned for the building market the coming years. Especiallyinformation characterising the energetically characteristics, weight and dimensions relevant whenconsidering building integration and key data to be used for further analysis of the systems wereasked:

- Weight per m2 and per kW for air and water based systems- Price per m2 and per kW for air and water based systems- Area per kW for air and water based systems- Range of temperatures supplied for air and water based systems- Thermal output power range for air and water based systems- Electrical output power range for air and water based systems- Typical efficiencies (thermal and electrical) for air and water based systems

Generally, a positive response has been achieved, but especially on the “one-off” project (ratherthan product based) building integrated systems, limited information has been made available.However several manufacturers have expressed their interest in further monitoring of their systemsboth in realised projects and laboratory experiments.


3.2 Overview of the survey

During the survey 32 examples of PV/T projects or products were found. Of these, 8 appear to becommercially available products. The remainder, appear to be either one-off designs for specificbuilding projects or else products under development. Most PV/T systems concentrate typically onpreheating of ventilation air. This might be due to the fact that a cavity behind the PV panels isalmost always available. Thus, it seems logical to use this cavity against little costs for thermal heatgains.

Table 1 shows a categorical system breakdown. The total given in the table 1 is greater than thetotal number of products found because some products can be mounted in more than one manner.

Heat transfermedium

Mountinglocation-method

Numberidentified

Commercial availableProducts

Air Roof integrated 12 3Facade integrated 10 4Separate module 9 3

Water Roof integrated 5 2Facade integrated 3 2Separate module 9 3

Table 1: Breakdown of PV/T systems identified

As mentioned earlier, there are currently only 8 commercial available products. Some companieshave announced that they soon would have a PV/T collector in production. However, they arehaving a problem maintaining the long time stability of the PV cells as they are integrated on anabsorber.

Heat transfermedium

Manufacturer Nationality

Air Millennium Electric(formerly Chromagen Solar Energy Systems)

Israel

Aidt Miljø A/S DenmarkConserval Engineering CanadaGrammer KG GermanyPhototronics Solar-technik, Putzbrunnpart of ASE in Germany.

Germany

Water Millennium Electric(formerly Chromagen Solar Energy Systems)

Israel

ICEC AG SwitzerlandSekisui Chemical Co., Ltd Japan

Table 2: Breakdown of commercially available PV/T systems identified by manufacturers


The systems located were constructed or first made available to the market in the period 1991 to2001. The year 1998 seemed to be a particularly busy year with a number of activities beingundertaken/initiated in the USA under the PV BONUS II program, in Germany and Switzerland.The manufacturers/designers were located in 12 different countries as listed below.

Canada 2 Japan 2Denmark 3 Netherlands 2Germany 4 Spain 1Israel 1 Switzerland 6Italy 1 USA 8Austria 1 Great Britain 1

3.3 Manufacturers of Commercial PV/T Collectors

3.3.1 Water PV/T collectorsIn Israel PV/T collectors have been developed by the company "Millennium Electric" (formerlyknown as Chromagen) since 1991. The commercially available PV/T collector is a flat plate solarheat collector with PV cells integrated on the absorber. Chromagen first tested their PV/T collectorin several locations in Israel, finding that the PV/T collector could supply apartment's electricity andhot water demand. The system was developed in such a way that additional generated electricitycould be sold to the local Electricity Company. The average cost for the PV/T collector, named"Multi Solar System", are about USD 940/m2 collector. The system can be grid-connected or stand-alone. It is sold with a minimum of 2 modules with a total collector area of 4.64 m2. The dailythermal and electrical output is about 1.5 kWh/m2 heat and 0.4 – 0.8 kW/m2 electricity in Israeliclimatic conditions. The product is marketed commercially and about 20 systems have been sold inIsrael [11] and 20 systems have been sold world-wide. 362 systems are contracted for USA in2002.

Multi Solar System in Singapore

In Switzerland, ICEC AG has developed a PV/T collector “HYSOLAR”, seehttp://www.icec.ch/products.html

In Asia, Sekisui Chemical Co., Ltd in Japan have developed a PV/T collector for domestic hotwater. The hybrid collector converts about 10% of the solar energy into electricity and 30% into hot


water on an annual average. It was commercialised in 1999 and further development is taking placeso the thermal energy from the PV/T collector can be used for space heating.

Sekisui Chemical Co., Ltd

Sekisui Chemical Co., LtdSee also http://www.sekisui.co.jp/general/english/eco/report2001_e.htmTwo German companies, “SolarWerk” and “SolarWatt” have developed a similar kind of PV/TCollector. Both systems are flat plate solar heat collectors with PV cells integrated on the absorber.“SolarWerk” developed their system in co-operation with the Institut fur Solarenergiforschung(ISFH) GmbH in Hameln, Germany. ISFH participated in the development and in the experimentson the PV/T Collector.

The contact person for the various performance tests on the PV/T Collector done at ISFH is RolandSillmann. The product developed is named “Spectrum” with a collector area of 2.2 m2. It can beinstalled either as a stand-alone or as grid-connected system. [Information received from SolarWerkHomepage at http://www.solarwerk.de/spectrum.htm


SolarWerk “Spectrum”

The other German Company “SolarWatt” developed a product named “MultiSolar” which is also aflat plate solar heat collector with PV cells integrated on the absorber, [13].

Both companies have however problems maintaining long time stability of the PV cells as they areintegrated on an absorber. This seems to be a common problem for companies interested incommercialising PV/T systems where the PV cells are integrated on the absorber.

3.3.2 Air PV/T collectorsThe Danish company “Aidt Miljø A/S”, has developed a solar air collector with integrated PVcells and fan. This product preheats ventilation air, but the main purpose of the product is actually toprovide dehumidification of the air in cabins, garages, allotment houses, mobile homes etc. The PVcells supply a fan in the top of the collector with electricity. The fan draws in outdoor air through aperforated aluminium plate on the whole backside of the collector, the air is warmed up in thecollector and is blown into the room.Two sizes are available: 0.35 m2 and 0.7 m2 with capacities of 25 m3/h and 50 m3/h respectively.


Aidt Miljø A/S “SolarVenti”j “SolarVenti” installed on a cabin

More information can be found on the homepage of the company www.aidt.dk.

The Canadian company “Conserval Engineering Inc.”, has developed a PV solar wall productwhere PV panels can be mounted onto a perforated absorber. This building integrated PV/TCollector can be used as a facade or roof element and is named “PV SolarWall”. The PV panelsare mounted in such way that cool ambient air is allowed to pass behind the PV panels in a uniformway. Heat generated from the PV cells will be transferred to the air, which can then be used forheating ventilation air. The PV Solar Wall is a variation of their standard SolarWall, which collectsthermal energy using the same perforated absorber plate without the PV panels. [14 andInformation received from Conserval Engineering at their homepagewww.solarwall.comm.html#12c.]

Conserval Engineering Inc. “PV SolarWall”


Phototronics Solartechnik of Putzbrunn, part of ASE, have developed an a-Si commercialcurtain-wall module about 1m by 0.6 m with 6% efficiency. These modules have among others beenused on the PV facade at the Bavarian Environment Ministry. This product is designed for verticalfacades of commercial and residential buildings and four basic product approaches are offered. Oneof these is a combined PV/Thermal panel incorporating semi-transparent PV panels for viewwindows within an insulated glass sandwich with warm air heat recovery. More at http://www.ase-international.com/english/start_e.html


Grammer Solar & Bau in Germany currently have a commercial available PV/T product wherethe PV panel is cooled by air for preheating ventilation air. The product is made available in 4different module sizes ranging from 50 kWp to 250 kWp per module, http://www.grammer-solar.de/solarthermie/pvkollektor.htm.

Grammer Solar & Bau in Germany




More information can be found on the homepage of the company www.grammer-solar.de, orhttp://www.twinsolar.de/index1.htm.3.3.3 ConclusionAir-cooled PV panels for electrical and thermal gain is currently the most commonly used PV/Tsystem. It is most suitable for façade type applications (second skin, Conserval and PhototronicsSolartechnik). These BIPV systems add a little thermal to the PV. Otherwise, thermal systemsare available that adds a little PV to their thermal gains e.g. PV powered thermal collectorsproduced by Grammer Solar & Bau in Germany.

In between are products being developed, e.g. Solarwerk, that both optimise the use of PV and thethermal system. These systems have PV cells integrated on the absorber in a conventional solarthermal collector. There are a number of companies working in developing commercial liquid PV/Tcollectors but currently only three companies Millennium Electric, Sekisui Chemical Co., Ltd,and ICEC AG have succeeded. There seems to be a problem in maintaining the long time stabilityof the solar cells as they are integrated on the absorber. These systems need to be developedfurther to serve the large consumer market of solar powered energy (heat and electricity).

3.4 Existing Building Projects with PV/T Collectors

A “PV SolarWall”, made by the Canadian Company “Conserval Engineering”, has been installed atthe West Prep School in Toronto, Canada. The 15 m2 SolarWall, with two 60-Watt UNISOLARPV panels, was installed to improve the indoor air quality in the classrooms. The heat generatedfrom the conventional SolarWall and from the PV panels is transferred to cooler ambient air, toprovide fresh air to the classrooms and thereby improve the indoor air quality and reduce the heatingcosts. The electricity generated from the two 60 Watt PV panels is used for running two ventilationfans which provide between 0 and 680 m3 of air per hour [14].

In 1993, the Japanese government invested more than 1200 million yen on PV demonstrations. Anarea of interest was the development of PV/T Collectors for buildings in Japan. One prototype wasdeveloped for residential applications and was tested on a test house in Japan. The PV/T systemconsists of PV cells backed by a thermal absorber and it produces daily about 3.2 kWp of electricityand 25 kW of thermal energy. [Information received fromwww.nrel.gov/ncpv/documents/japan.html]. Also in Japan, Sekisui Chemical Co., Ltd hasinstalled PV/T collectors for domestic hot water in several residential homes. The hybrid collectorconverts about 10% of the solar energy into electricity and 30% into hot water on an annualaverage.


Sekisui Chemical Co. Ltd. Sekisui Chemical Co. Ltd.

Richard Komp and Terry Reeser of SunWatt Corporation [9] have constructed and operated an airPV/T Collector for a passive solar house in Louisville , Kentucky. It is built into the roofstructure of an attached sunspace and uses natural convection to extract excess heat from finnedmodule assemblies, and delivers that heat into the house during the winter. During the summer, theheat is exhausted from clerestory windows, creating a draft of cooler air into the lower part of thebuilding. The electrical output is delivered through a battery to low voltage lights and otherappliances in the house.

Solar Design Associates, Inc was involved in a roof-integrated array of PV/T Collectors at theMontana State University Research Centre that delivers both thermal and electrical energy tothe building. An array of PV cells was also roof integrated at a custom solar residence on the coastof Maine, where the arrays are passively cooled from behind by ambient cool air. The heated air isthen used for ventilation purposes. [Information received fromhttp://www.solardesign.com/prodev.html.

Innovative Design from North Carolina has developed a integrated PV system that use wasteheat from a PV array to heat up water [10]. A roof-integrated example has been installed on anApplebee’s restaurant that uses 32 amorphous PV modules. Eight of the modules are connectedto a fan that circulates air through a series of passages underneath the 32 modules. About 7 % ofthe total area of the system is clear glass between the PV cells, facing a black-painted high-absorbing metal pan. As insolation increases and the temperature goes up, a fan switches on tocirculate the heat away from the PV modules towards a heat exchanger. The heat flows through aclosed loop and is thereby not wasted as it is in conventional PV modules. Since there is a bigdemand for hot water in the building, all the hot water produced from the solar energy system isused. As it is a roof-integrated system, the costs for conventional roof finishing were saved.

In 1994 in Ispra, Italy, the ELSA building had its 25 m high south facade covered by an area of505 m2 of amorphous PV cells with an electrical output rating of 21 kWp. The PV cells weremounted onto an insulated wall with an air gap behind the PV cells. The heated air behind the PVcells is used for ventilation purposes [15].


Similarly PV/ T energy systems are installed at the Library of Mataro in Spain and at the YellowHouse in Aalborg, Denmark. At the Library of Mataro in Spain, PV cells are mounted onto thefacade and on skylights on the roof. In chambers behind the PV cells air is heated and is then usedfor preheating the water in the conventional gas-fired heating system of the building [16]. TheYellow House in Aalborg, Denmark has 5 different groups of PV installations in the facade fordemonstration purposes. [17-21]

Yellow House in Aalborg, Denmark

During the last 10 years, Atlantis Energy Systems, Ltd. from Switzerland, have made severalPV/T systems around the world. One system, installed in 1991 at the factory building “Aerni” inArisdorf, Switzerland, has a ventilated PV facade and ventilated PV skylights with a total electricaloutput of 62 kW and a thermal output of 115 kW. Atlantis Energy Systems Ltd. also developed aPV/T shingle roof, two of them are named “Brig” and “Rigi” where PV panels are installed on theroof with ventilation air passed behind the PV cells. Both systems have been in operation since 1993without interruption and have shown good results [21].

At the City Archives in Rotterdam, Netherlands some 1840 m2 of PV cells are installed on theroof. Beneath the PV cells heat is generated. During the summer this heat is stored in the ground toprovide heating in the winter. During the winter, cool air is stored in the ground to provide coolingduring the summer. [Information received from www.dubo-centrum.nl and 22]

Several systems of the type “Multi Solar System” from "Millennium Electric" (formerly known asChromagen) with a daily output of 2 to 4 kWh electricity and 6,000 kcal of hot water have beeninstalled in residential homes in Klil, a small mountainous community in northern Galilee, Israel,isolated from the national grid. The systems have been operating and monitored since 1991 and havebeen providing the total energy demands of the typical family household. [11]


Multi Solar System in Klil, Israel

In Denmark, several projects using the PV-Vent concept have been realised. Fresh air for lowenergy ventilation systems with heat recovery passes behind PV modules integrated in the buildingfacade or roof, thereby both preheating the ventilation air and cooling the PV modules. Theventilation systems are directly supplied with DC electricity and a so-called “PV-mixer” measuresthe PV electricity and when there is not sufficient PV electricity, supplementary electricity issupplied from the grid. One of the projects was realised in a housing block with 27 flats inLundebjerg near Copenhagen. [53]

Lundebjerg in Skovlunde, Denmark Solar Chimney, Lundebjerg, Denmark


3.5 Research and governmental support

The DOE initiated the Building Opportunities in the United States for PV (PV Bonus) program in1993 to develop cutting-edge solar products for the building industry. Program objectives were todevelop technologies and foster business arrangements for products cost-effectively integrating PVor PV/hybrid technology into buildings. An important factor is that these products must be installedwithout the need for specialised training. The program was conducted through two competitivesolicitations, termed PV Bonus I and PV Bonus II. Twenty-two partnerships were initiated underPV Bonus and from these, five new products were developed. Products included solar roofingshingles and a factory-built modular home integrating PV. PV Bonus II began with 16 partnerships.Seven were selected for additional work, resulting in five commercially available products. Productsrange from an “enabling” PV application to dual-purpose or hybrid products, and products for uniqueapplications.

PowerLight Corporation is completing final tests for a PV/T product named PowerRoll TM, acombined PV/thermal hybrid system for medium-temperature hot water applications. The productcombines the USSC flexible triple-junction module adhered to a heat-transfer backing material.Lessons learned in this work are the challenge of combined testing to meet solar concentratorstandards, PV module standards and UL requirements. Technical lessons learned included materialsselection, such as an adhesive meeting safety codes and surviving outdoor exposure and operating atelevated temperatures. [23].

PowerLight Corporation “PowerRoll”

The team of Solar Design Associates (SDA), United Solar Systems Corp. (USSC), and SunEarthInc. is developing a hybrid PV/thermal product called PhotoTherm. The product is a unitisedcombination of a water thermal collector and the USSC triple-junction a-Si thin-film module.PhotoTherm resembles a traditional solar thermal design, except the PV module replaces the topsurface of the absorber plate. The current Phototherm product is designed for installation on anexisting roof. The partners gained experience in defining a solar product capable of highertemperature operation and selecting materials to lower product cost. Because the hybrid product willoperate as a PV module and a source for hot water, qualification tests had to be defined. Theproduct must also meet requirements for safety (UL), PV modules and solar thermal products andbuilding codes.


Solar Design Associates, United Solar Systems Corp. And SunEarth Inc. “PhotoTherm”

IT Power's Test and Training Centre, located near to the office in Eversley, Great Britain, iscurrently working on Photovoltaic/Thermal system development - concept assessment, systemdesign & testing, see http://www.itpower.co.uk/services/r&d.htm

At North Carolina Solar Center, the Solar Engineering Specialists Rob Stevens and ShawnFitzpatrick are conducting an exciting patentable PV/T research project. Stevens and Fitzpatrickhope to develop a practical way to use heat from the backside of a PV panel to provide water andspace heating and possibly drive an air conditioning system. “One of the key features of this designis that the panels will be roof integrated, or incorporated into the actual structure of the roof,” saidStevens. More information can be found athttp://www.engr.ncsu.edu/news/news_articles/solar.center.html

North Carolina Solar Center


3.6 Design Considerations

The following chapter summarises the “state-of-the-art” regarding recommendations for designingBIPV solutions with PV/Thermal solar collectors. The recommendations are provided by IEAPVPS Task7 and other external companies and scientists working in the field of designing solarthermal systems and PV systems.

3.6.1 General recommendationsIn choosing what type of PV/T system is most suitable for a particular project, the project demandsneed to be considered, e.g.:

1. Temperature and characteristics of thermal load2. Thermal load (kW)3. Electrical load (kW)4. Suitable mounting locations5. Building constraints, e.g. weight bearing capacity, aesthetics

Some general notes:- System is sized for the thermal load, as this is more usually the constraint. Particularly

with grid connected buildings where the electrical load can be considered infinite- The current PV/T building integrated systems are mostly air based. Some water-based

systems are under development.- PV/T water based systems are most suitable for low temperature applications (swimming

pool, heat pump combinations).- Water based systems are normally designed as separate modules, in part because they

have evolved from solar water heating systems and also presumably because there areless concerns regarding the built quality of factory built modules and hence the risk ofwater leakage.

- Domestic water heating loads normally require relative high temperatures (in some casesmin. 60°C to avoid Legionellosis) and usually water based systems in order to provideheat exchange with storage.

- Medium temperature demands normally use water based systems, due to the practicallimitations in capacity-flow of air-based systems.

- Process heating (industrial applications) – no references known so far. Will usuallyrequire a combination with storage to achieve a constant energy flow and usuallyrelatively high temperatures. Some design ideas have been developed for preheatingelements for burners, moisture control of bio-fuel but not realised so far.

- (Systems or projects associated with thermal storage for the long-term storage of heatcould also be discussed).

The current technical level of the commercial PV/T systems still need to be verified tested andmonitored. Still many issues regarding the combination of materials, the dependency on temperaturelevel on the overall yield and the optimum combination of heat and electricity production for variousclimates and applications needs to be determined.


3.7 Conclusions for chapter 3

The status of commercial PV modules is that only 10-15% of the incident solar energy istransformed to electricity. The potential heat production from a given surface is thus much higherthan the electrical performance, but it is an open question whether this heat can be used in asensible way. Combining PV and a solar thermal collector for tap water heating ends up with atemperature compromise. PV needs to have a low temperature to maintain a high efficiency,whereas a solar thermal collector requires a high temperature. With the current technologies, thePV/T combination has a lower efficiency than two separate systems and, due to the initialdevelopment stage, the PV/T combination is also more expensive. However, advantages areforeseen in aesthetics, future (production & installation) cost reductions and market / consumerrequirements. Based on the survey carried out and discussions with various building designers listed,there seem to be several more obstacles:

Based on the survey carried out and discussions with various building designers listed, there seem tobe several obstacles:

- Most buildings (e.g. in Western European climates) need heating in winter when the solar gainis at its’ lowest

- The heat is needed at a higher temperature than the surrounding air, leading to increased moduletemperature unless a heat pump is used.

- For heating of domestic hot water, a heat exchanger is needed for safety and health regulations- The collectors could become very hot and thereby could get damaged if circulation of cooling

media is blocked.- The construction may be too complex and thus expensive compared to separate PV and thermal

collectors.

A number of obvious advantages exist, and if the obstacles are overcome, there seems to be a largepotential for PV/Thermal Solar collectors in the future: However, much R&D work is still required,and knowledge transfer is needed etc.

- The total area used to extract a given amount of electricity and heat may be smaller than fortwo separate systems

- The materials used for a PV/T plant, and thus the total energy and economy balance, may bebetter than for separate units.

- The roof or facade will have a more uniform look, providing the potential for pre-fabricatedsystems developed for various types of roofs.

- When using integrated elements a potential saving in installation costs compared to separatesystems can turn out to be a very important factor for the future development of the market forPV/Thermal solar collectors.

- A careful design with utilisation of low temperature levels for e.g. for swimming pools and incombination with heatpumps.


3.8 References

[1] B. Sandnes. Oslo Universitet.

[2] Photovoltaic/Thermal Solar Systems. IEA workshop report. Amersfoort, 17-18 September1999.

[3] M. Mattei, C. Cristofari and A. Louche, Modelling a Hybrid PV/T Collector, 2nd WorldConference on Solar Energy Conversion, July 1998.

[4] PV Power and Heat Production: An added Value. Bent Sørensen, Roskilde University.16th European Photovoltaic Conference, Glasgow 2000.

[5] Technology Review on PV/Thermal concepts. F.Leenders et al. 16th EuropeanPhotovoltaic Conference, Glasgow 2000.

[6] Solar Photovoltaic/thermal Co-generation Collector. B.J.Huang et al. Dept. of MechanicalEngineering, National Taiwan University, Taipei, Taiwan.

[7] Cooling of Building Integrated Photovoltaics by Ventilation Air. M.Sandberg, Laboratory ofVentilation and Air Quality, KTH Dept. of Built Environment, Gävle.

[8] “Hybrid Collectors, Theoretical Developments and Performance Evaluation of PhotoVoltaicThermal Collectors”, Master Thesis by Bruno Nielsen, Esbensen Consulting EngineeringLtd. Copenhagen.

[9] Richard Komp and Terry Reeser, "Design, Construction and Operation Of A Site BuiltPV/Hot Air Hybrid Energy System", Article obtained from Richard Komp at SunWattCorporation at +1 207-497-2204, 1998.

[10] K. Sheinkopf, "PV System With Thermal Heat Recovery", Article of the work of the IEA,obtained from www.caddet-re.org/re/html/body_298art2.htm

[11] Dr. A. Elazari, "Multi Solar System, Field Experience In Israel", Several articles obtained from Dr. Elazari at Chromagen in Tel Aviv, fax: 972 - 3 - 525 - 6305.

[12] "Building Opportunities In The U.S For Photovoltaics (PV:BONUS), Two", Article obtainedfrom Robert J. Hassett, U.S. Department of Energy, Tel: +1 202 586 8163.

[13] "MULTISOLAR", Multisolar is a registered trademark of KRUSE Gmbh Technik.

[14] "Hybrid Solar Collectors for Portable School Classrooms", Article obtained from PerDrewes, Ontario Hydro Technologies.

[15] ”Thermal and Power Modeling of the Photovoltaic Facade on the ELSA Building, ISPRA",“Thermal Aspects of PV Integration in Buildings”, “Analysis of Fluid Flow and HeatTransfer within the PhotoVoltaic Facade on the ELSA Building, JRC ISPRA”, 3 papersfrom the 13th. European Photovoltaic Solar Energy Conference, October 1995.


[16] "The Library of Mataro description and first results of monitoring", Article by Dr. AntoniLloret, Labotatoire des Interfaces et des Couches Minces Ecole Polytechnique, 91128Palaiseauu cedex, France.

[17] "The Yellow House- An innovative solar renovation of multi storey housing" O. B.Jørgensen, Esbensen Consulting Engineers Ltd. 1997, proceedings from 7th. InternationalConference on Solar Energy at High Latitudes – North Sun `97, Espoo-Otaniemi, Finland

[18] "Integration of Solar Energy in future renovation of multi storey housing - The YellowHouse" O. B. Jørgensen, Esbensen Consulting Engineers Ltd. 1997, proceedings fromEuroSun `96, Freiburg, Germany

[19] "Monitored results from the Yellow House" O. B. Jørgensen and L. T. Nielsen, EsbensenConsulting Engineers Ltd. 2000, paper presented at EuroSun 2000 in Copenhagen, Denmark

[20] "The Yellow House-Final Report" Esbensen Consulting Engineers Ltd and SBSByfornyelse, 2002, Copenhagen, Denmark.

[21] "The Importance of Hybrid PV-Building Integration", Paper by M. Posnansky, AtlantisEnergy Ltd. 3012 Bern, Switzerland, Tel: +41 031 300 3220 Fax: +41 031 300 3230.

[22] "City Archives Rotterdam", Paper received by email from Siard Hovenkamp [email protected]

[23] “Building Integrated PV and PV/Hybrid Products – The PV:BONUS Experience”,Presented at the NCPV Program Review Meeting, Lakewood, Colorado 14-17- October2001.

[24] “Technology reveiw on PV/Thermal concepts”, Paper by F. Leenders, Ecofys, NL, B.G.C.vand der Ree, TNO Bouw, NL: and W.G.J. vand er Helden, TUE University of TechnologyEindhoven. Presented at the EuroSun Conference Copenhagen 19.-22. June 2000.

[25] “Photovoltaic/Thermal Solar Systems”, Report on a joint workshop of the IEA SolarHeating & Cooling Programme, IEA Photovoltaic Power Systems Programme, Amersfoort,the Netherlands, 17-18. September 1999. Edited by ir. F. Leenders and drs B.G.C. van derRee, Ecofys NL.

[26] L. W. Florschuetz, "Extension of The Hottel-Whillier Model to The Analysis of CombinedPhotovoltaic/Thermal Flat Plate Collectors", Solar Energy, Vol. 22 pp. 361 - 366, 1979.

[27] Susan D. Hendrie, "Evaluation of Combined Photovoltaic/Thermal Collectors", Proceedingsof the International Solar Energy Society, Vol. III pp. 1865 - 1869, 1979.

[28] D. J. Mbewe et al., "A Model of Silicon Solar Cells for Concentrator Photovoltaic AndPhotovoltaic/Thermal System Design", Solar Energy, Vol. 35, No. 3, pp. 247 - 258, 1994.

[29] C. H. Cox and P. Raghuraman, "Design Consideration for Flat Plate Photovoltaic/ThermalCollectors", Solar Energy, Vol. 35, No.7, pp. 227 - 241, 1985.


[30] A. Braunstein and A. Kornfeld, "On The Development of The Solar Photovoltaic AndThermal (PVT) Collector", IEEE Transaction on Energy Conversion, Vol. EC-1, No.4, pp.31 - 34, December 1986.

[31] Ram Kumar Agarwel and H. P. Garg, "Study of a Photovoltaic-Thermal System-Thermosyphonic Solar Water Heater Combined with Solar Cells", Energy Conversion AndManagement , Vol. 35, No. 7, pp. 605 - 620, 1994.

[32] Ram Kumar Agarwel, H. P. Garg, and J. C. Joshi "Experimental Study On A HybridPhotovoltaic-Thermal Solar Water Heater And Its Performance Predictions", EnergyConversion And Management , Vol. 35, No. 7, pp. 621 - 633, 1994.

[33] Jai Prakash "Transient Analysis Of A Photovoltaic-Thermal Solar Collector For Co-Generation Of Electricity And Hot Air/Water", Energy Conversion And Management, Vol.35, No. 11, pp. 967 - 972, 1994.

[34] Takumi Takashima, et al., "New Proposal For Photovoltaic/Thermal Solar Energy UtilizationMethod", Solar Energy, Vol. 52, No. 3, pp. 241 - 245, 1994.

[35] Ram Kumar Agarwel and H. P. Garg, "Some Aspects Of A PV/T Collector/ForcedCirculation Flat Plate Solar Water Heater With Solar Cells", Energy Conversion AndManagement , Vol. 36, No. 2, pp. 87 - 89, 1995.

[36] Takumi Takashima, et al., "On The Consideration Of Total Efficiency OfPhotovoltaic/Thermal Panel", T. IEEE Japan, Vol. 115-B, No. 4, pp. 430 - 435, 1995.

[37] Trond Bergene and Ole Martin Loevvik, "Model Calculations On A Flat Plate Solar HeatCollector With Integrated Solar Cells", Solar Energy, Vol. 55, No. 6, pp. 453 - 462, 1995.

[38] K. Sopian et al., "An Investigation Into The Performance Of A Double Pass PhotovoltaicThermal Solar Collector", AES , Vol. 35, pp. 89 - 94, ASME 1996.

[39] K. Sopian et al., "Performance Of A Hybrid Photovoltaic Thermal Solar Collector", AES ,Vol. 36, pp. 341 - 346, ASME 1996.

[40] Gouri Datta and H. P. Garg , "Theoretical And Experimental Studies On A SolarPhotovoltaic Thermal (PV/T) Liquid Heating System With Thermosyphonic Flow", WorldRenewable Energy Congress, Vol. III, pp. 1815 - 1819, Denver, Colorado 1996.

[41] K. Sopian and K. S. Yigit et al., "Performance Analysis Of Photovoltaic Thermal AirHeaters", Energy Conversion And Management, Vol. 37, No. 11, pp. 1657 - 1670, 1996.

[42] H. P. Garg and R. S. Adhikari, "Conventional Hybrid Photovoltaic/Thermal (PV/T) AirHeating Collectors: Steady State Simulation", Renewable Energy, Vol. 11, No. 3, pp. 363 -385, 1997.

[43] K. Sopian et al., "Research and Development Of Hybrid Photovoltaic Thermal Solar AirHeaters", International Journal Of Global Energy Issues, Vol. 9, Nos. 4-6, pp. 382 - 392,1997.


[44] C. Choudhury and H. P. Garg , "Performance Of A Two-Pass Photovoltaic/Thermal AirHeater", World Renewable Energy Congress, Vol. III, pp. 1803 - 1806, Denver, Colorado1996.

[45] S. Gasner and L. Wen, "Evaluation Of Unglazed Flat Plate Photovoltaic-Thermal CollectorsIn Residential Heat-Pump Applications", Solar Engineering, ASME, pp. 302 - 308, 1982.

[46] G. V. Tsiklauri et al, "Combined Photovoltaic And Thermal Modules With HydrogenAccumulation For Solar Power Plants", ECE Energy Series, Vol. 11, pp. 194 - 199, 1993.

[47] Mark M. Koltun, "The First Russian 1 MW Combined Photovoltaic And Solar ThermalPower Plant: Basic Ideas Advantages Of The Project", Renewable Energy, Vol. 5, part I,pp. 179 - 181, 1994.

[48] M. A. K. Lodh, "Hybrid Systems Of Solar Photovoltaic, Thermal And Hydrogen: A FutureTrend", International J. Hydrogen Energy, Vol. 20, No. 6, pp. 471 - 484, 1995.

[49] Douwe de Vries, "Design of a photovoltaic/thermal combi-panel", PhD-thesis, at theUniversity of Technology in Eindhoven.

[50] "Entwicklung eines PV-Hybrid-Kollektors", 8. Symposium Thermische Solarenergie,Ostbayer-isches Technologie Transfer Institut, Regensburg, 1998, pp 77 – 82.

[51] "PV-Hybrid and Thermo-Electric-Collectors", Gunther Rockendorf and Roland Sillmann,Institut für Solarenergieforschung Gmbh, Hamln/Emmerthal, ISFH. Tel. +49 5151/999-521

[52] "Photovoltaic cogeneration in the built environment", M.D. Brazilian, F. Leenders, B.G.C.van der Ree, D/. Prasad, ":, Solar Energy, Vol. 71, No. 1, pp 57-69

[53] "PV-Vent Low Cost Energy Efficient PV-Ventilation in Retrofit Housing", Non-Nuclear-Energy Programme Joule III, Paper by Mr. Peder Vejsig Pedersen and Ms. AnneRasmussen, Cenergia Energy Consultants obtained from www.ecobuilding.dk.


4. Initial calculations

Whether a PV/T collector is better than a separate PV module and a solar collector or not, dependson several factors:

- Is space a problem, or is the total collector area unimportant?- Does the load profile fit to the combined production of heat and electricity?- Does the installation become simpler or more difficult than in separate systems?- Does the production costs increase due to a more complex construction?

In order to get a first estimate of the characteristics of PV/T the traditional and well-documentedset of equations for solar thermal collectors are used. The only difference to a PV/T collector is thata part of the collected energy is extracted as electricity instead of heat. If the formulas arecorrected for this and secondary effects such as radiation heat transfer inside the solar cells areneglected, they are valid for the PV/T collector.

4.1. Spreadsheet model for PV/T collectors

A model in Excel was made for the two mentioned PV/T collector types. Two cases wereinvestigated for the first type, namely:

a) Standard PV module with ideal rear surface cooling, PV cells are acting as the absorber

b) As for a) but with thick (15 mm) acrylic coverplate for improvement of U-value

c) As for b) but with an airgab between coverplate and PV cells for additional improvement ofU-value

The results are calculated at standard conditions used for characterisation of thermal collectors,namely G=800 W/m², V=5 m/s, and Ta=293 K

4.1.1 Performance in terms of energy and exergyWhen the various types of PV/T collectors are to be evaluated, a principal question arises – what isthe value of the electricity versus the heat form the collector? The consumer’s rate of electricityand heat is to a large extent a politically determined value, and if those rates were used the resultswould not be universally valid. We have therefore chosen to present two key figures for eachcollector:

1) The total energy yield per year for the Danish Test Reference Year (TRY). The results arecalculated from the 1st. law of thermodynamics, known as the energy efficiency.

2) The total exergy per year, which is the part of the energy that could theoretically be convertedto work in an ideal Carnot process. The results are calculated from the 2nd. law ofthermodynamics, known as the exergy efficiency.


If economy is calculated on basis of case 1, the cost of electricity and heat is thus the same, whilefor case 2 the electricity is given a much higher value than the heat. The reality will be somewherein between these two extremes.


The energy efficiency is calculated as

GQthermal

thermal =η

G

Q powerpower =η

G

QQ powerthermaltotal

+=η

While the exergy efficiency is calculated as

( )G

TT(K293/(K2931Q athermalthermal

−+−⋅=ε

G

Q powerpowerpower == ηε

( )G

Q

GTT(K293/(K2931Q powerathermal

total +−+−⋅

=ε

Where

G))TT(1(Q arefref,elpower ⋅−⋅−⋅= βη

G)G

)TT(U(Q a

L0thermal ⋅−⋅−= η

The constants and the calculated variables for use in the above expressions are found in table 4.1

Nocoverplate

15 mm acrylic coverplate directly onPV cells

15 mm acrylic coverplate with anairgab between coverplate and PV

cellsG 800 W/m2 800 W/m2 800 W/m2

0η 0,6101 0,6236 0,6100

LU 14,8192 8,3618 7,2165

ref,elη 12,5 % 12,5 % 12,5 %

refβ 0,005 0,005 0,005T 293 K 293 K 293 K

Table 4.1. Constants and variables for use in efficiency expressions


Energy Efficiency of PV/T collector

00,10,20,30,40,50,60,70,80,9

1

0 10 20 30 40 50 60 70 80

T-Ta

Thermal Electrical Total

Exergy Efficiency of PV/T collector

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

0 10 20 30 40 50 60 70 80

T-Ta


Figure 4.1 Efficiency of PV/T collector without coverplate

Table 4.1 and the curves in Figure 4.1 show that the high heat loss coefficient due to the lowthermal resistance of glass results in a low thermal performance at temperature above 20K overambient temperature.


00,10,20,30,40,50,60,70,80,9

1

0 10 20 30 40 50 60 70 80

T-Ta



0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

0 10 20 30 40 50 60 70 80

T-Ta


Figure 4.2: Efficiency of PV/T collector with 15 mm acrylic coverplate directly on PV cells

In case b) the thermal performance is improved for temperatures above 10K over ambienttemperature.

The electrical output is almost the same for the two, because acrylic has very good opticaltransmission properties so the thickness is not a serious drawback.



00,10,20,30,40,50,60,70,80,9

1

0 10 20 30 40 50 60 70 80

T-Ta



0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

0 10 20 30 40 50 60 70 80

T-Ta


Figure 4.3: Efficiency of PV/T collector with 15 mm acrylic coverplate and airgab betweencoverplate and PV cells

If an air gap is introduced between absorber/PV cells and the coverplate, the thermal performanceis slightly improved, but not to the level of modern solar collectors with a selective coating.

All simulations were made with a back insulation of 30 mm mineral wool and an ideal heat transferfrom solar cells to the collector fluid.

It is clear that the practical performance depends on the system the collectors are a part of.Assuming that the PV part is grid connected, and all produced power therefore is useful, it is mainlythe temperature level and thermal storage size that determines the actual yield.


5. Simulations

5.1 Collector parameters and characteristics

After the initial calculations - presented in chapter 4 of this report - showed a great potential ofPV/T collectors the next step is to identify the most promising design of PV/T collectors regardingmain fields of application:

• Water PV/T collectors for hot water.

• Air PV/T collectors for warm air.

The detailed mathematical models have been set for the above PV/T units and the following user-friendly computer programs have been developed:

(i) Subroutines within TRNSYS Program for performance simulation of the PV/water-heatingand the PV/air-heating components (developed at TI)

(ii) Visual Basic program for detailed simulations developed at NOVATOR AdvancedTechnology Consulting

In order to identify optimal design of the PV/T components multi-parametric analyses were carriedout. Following parameters have been taken into consideration:

• Absorptance of absorber (PV cells front surface)• Emittance of the absorber (PV cells front surface)• Quality of the thermal contact between the PV cells and the absorber• Finn efficiency factor• Back insulation thickness• Temperature coefficient of PV cells• Efficiency of PV cells

The results of this analysis are presented in chapter 6. After identification of optimal design, theeffect of these parameters on system performance (i.e. efficiency curve) for two basic types ofcollectors (selective and non-selective absorber) has been analysed. The conclusions on the resultsextracted from annual energy simulations are also outlined in chapter 6 of this report.

With respect to air PV collectors, two different assumptions are made:

• The PV cells are situated on the coverplate acting as an absorber and fully absorb incidentradiation,

or• The PV cells are situated on the coverplate acting as an absorber and are partly transparent to

incoming radiation.


In the analysis of energy yield we consider the three most important energy yields of the PV/Tcollector as follows:

• Electrical energy extracted all over a typical year• Availability of thermal energy for space heating delivered by PV/T collector• Availability of thermal energy for water heating delivered by PV/T collector – using air/water heat exchanger

A multi-parameter analysis has been carried out using two reference constructions of air PV/Tcollectors. The first reference assumes a small light transparency for the solar spectrum and thesecond reference assumes a light transparency of 80%.

The first strategy was to determine the airflow rate, which leads to maximum annual yield for agiven PV/T collector design. After determination of an optimum airflow rate for a particular system,the next step is determination of optimal design parameters, which will lead to a maximum energyyield.

In chapter 6, the focus is on the PV/T collector itself and therefore, the analysis has been based onenergy contribution assuming constant temperature levels for inlet fluid in water PV/T collectors.For air PV/T collectors (once-through-flow units) the profile of inlet air temperature is equal toambient air temperature according to Danish Reference Year.

Finally, computation of typical system yields under real load consumption has been carried out.These results are presented in chapter 7.


6. Simulation of the PV/T Performance with TRNSYS

6.1. PV/T Collector for Water Heating

The basic structure of a hybrid thermal and photovoltaic collector (PV/T) is shown in Figure 6.1.The PV cells are in thermal contact with the absorber. A fraction of the solar energy incident on thecollector surface is transformed to electricity while the remaining part of the solar energy istransformed into thermal energy in the same manner as for a conventional solar thermal collector.

Figure 6.1: Construction of a PV/T Collector for Water heating [1]

Back wall

Insulation

Fluid Circuit

Absorber

PhotoVoltaicCells

Sealfilm

Coverplate

Frame

Fluid connection

Powerpoint


6.1.1. Theoretical Model of the PV/T - CollectorCollector thermal performance is usually described by the useful heat output as a function of inputradiation and collector operating temperature relative to its surroundings. When the equation of theperformance curve is known for a specific solar collector, the system designer has the informationneeded to employ any of several recognized computational techniques to predict the daily, seasonal,or annual energy output of the collector under the anticipated use conditions of the system beinganalyzed. The analytical derivation of this efficiency is briefly reviewed in the next sectionexpression ([2] Hotel, Whillier and Bliss).

The General PV/T Collector Performance EquationIn order to characterize the performance of a PV/T solar collector properly, an energy balance mustbe performed which considers all the energy flows to and from the collector. For a typical glazedflat-plate solar collector with PV cells as absorber, this balance may be expressed as

eltyheatcapacilosssolaruseful q)q(qqq −+−=

where

quseful = the rate at which useful energy is deliveredqsolar = the rate of solar radiation absorbedqloss = the rate of energy loss from the collector to the environment caused by

convection, infrared radiation, and conductionqel = the rate of electrical energy extracted from the collectorqheatcapacity = the rate of energy storage within the collector

The qsolar term is a function of the optical properties of the collector coverplate and absorber surface(PV-cells). The qloss term depends upon how well the collector is thermally isolated from itssurroundings. qheatcapacity is a function of mass and type of collector’s materials.

The General PV/T-Collector Performance Equation with Heat CapacityA modified multi-node collector model [3 Bosanac et al. 1993] has been used for this analysis. Themodel has the following features:

- The collector is modeled with distributed capacities in flow direction.- A linear dependence of the heat loss coefficient on the surrounding air speed as well as on the

temperature difference between collector and ambience is assumed.- Incident angle modifiers for beam (as a function of incident angle) and diffuse irradiance are

used.

In this section the main features of the model are briefly described.

Each node of a flat-plate collector is characterized by:

[ ] n,uelanLennn qq)TT(UG)(FA

dtdTC −−−−′= τα

where G is the equivalent normal irradiance taking into account irradiance components multiplied byrespective incident angle modifiers:


albalbdiffdiffbeambeam GKGKGKG ++=

UL is the overall heat loss coefficient:

)TT(UvUUU anTvoL −++=

qu,n is the rate of energy gained by the collector node:

)TT(Cmq 1nnpcn,u −−=

(τα)e is the transmittance-absorptance-product at normal incidence. An incidence angle modifier forbeam irradiance is defined by the modified Ambrosetti equation ([4] Ambrosetti 1983):

=

2tan-1)(K 1/r

beamθ

θ

The incident angle modifier for diffuse irradiance assuming isotropic distribution is used as derived in([3] Bosanac et al.1993). The incident angle modifiers for diffuse irradiance and for albedo areassumed to be equal. They are both derived in ([3] Bosanac et al.1993) as a function of theparameter, r.

Hence, the following parameters fully characterize the presented model:

• The optical efficiency of collector array, e)(F τα′ .• The overall heat loss coefficient if Tn = Ta and v = 0, Uo.• The coefficient characterizing wind dependence of overall heat loss, Uv.• The coefficient characterizing temperature dependence of overall heat loss, UT.• The total thermal capacity of collector array, C.• The incident angle modifier coefficient, r.

As the temperature of the collector mass increases or decreases during the day, energy is stored orreleased from the collector thermal mass. In the most cases, collector heat capacity effects can beneglected, yielding in quasi-steady-state conditions and simplified form of equation:

ellosssolaruseful qqqq −−=

The collector thermal output, quseful, may be represented graphically by a second order curve asshown in Figure 6.2 The ordinate intercept of the curve equals the maximum output, which isachieved when the collector is delivering energy at the ambient temperature. The unavailable portionof the energy falling on the collector is that which is reflected from the coverplate or absorber, orabsorbed by the coverplate.


Figure 6.2: Efficiency curve of the collector (tm = mean fluid temperature and ta = ambienttemperature)

As useful heat is removed from the collector at higher temperatures the efficiency decreases, sincelosses from the absorber increase in proportion to its temperature above its surroundings. Radiationaccounts for a significantly greater proportion of the losses at elevated operating temperature fromsome collectors than the others. A second or higher order efficiency equation is often used todescribe the performance of collectors in which radiation causes significant heat loss at higheroperating temperatures.

Energy into the CollectorThe net rate of solar radiation absorbed by a collector, qsolar, is a function of the radiation on thecoverplate and the optical and radiative properties of the materials constituting the coverplate andabsorber. Since no real glazing material is perfectly transparent, one part of the radiation coming onthe cover is absorbed and the other part is reflected by the glazing material; only a fraction istransmitted through the cover. The transmitted fraction is partly absorbed by the absorber and partlyreflected back toward the cover; this reflected radiation is again partly transmitted through thecover, partly absorbed by it, and partly reflected back to the absorber. The result of this multipleabsorption, reflection, and transmission is only a fraction of that radiation is ultimately absorbed bythe collector.

The parameter that quantifies the capability of the collector to absorb solar radiation is called theeffective transmittance-absorptance product, (τα)e. The description "effective" is important whilethe energy absorbed is primarily a function of τ, the transmittance of the glazing(s), and α, theabsorptance of the absorber plate surface, the complex interactions discussed above modify theproduct (τα)e, in complex ways, particularly in collectors which have two or more glazing layers.Using (τα)e, the net rate at which incoming solar energy absorbed by a collector may be expressedas

)(GA=q easolar τα


where G is the total incident radiation per unit area, measured in the aperture plane of the collector,and Aa is the collector absorber area.

The dependence of the collector performance on the physical properties of the cover and absorbersis more complex than suggested by the above discussion, since these material properties alsoinfluence the radiative heat losses from the absorber plate. In this respect the wavelengthdependence of the radiative properties of the cover and absorber materials can be usedadvantageously in solar collector design. The ideal glazing material (at least from the thermalperformance point of view) should be transparent to solar radiation in order to maximize thetransmitted fraction of the incident insulation, but opaque in the infrared region where the radiativeheat losses from absorber plate occur.

Glass is one material, which exhibits these characteristics quite well; glass is virtually opaque toradiation of wavelengths higher than 3 µm, but highly transparent to solar radiation, while manyplastic materials, particularly films, are transparent to both solar and infrared radiation. Similarly,absorber plate materials are available which have high absorptance in the solar spectrum tomaximize the useful absorbed radiation, and low infrared emittance, in order to minimize radiativeheat losses. Materials that exhibit such behavior are called selective. A selective absorber is definedhere as a material achieving an absorptance in the solar spectrum of at least 0.85 while having aninfrared emittance no greater than 0.6. Collectors using selective absorber surfaces will generallyhave lower heat loss (qloss) than those using conventional nonselective absorbers, particularly athigher collector operating temperature.

In the case of the PV/T collector, PV cells act as non-selective absorber having coefficient ofemittance of approximately 90%.

Heat Losses from the CollectorThermal losses from solar collectors occur in three ways: conduction, convection and radiation.Heat losses due to conduction are usually negligible, unless poor collector design or constructionresults in the collector case or mounting structure coming into direct thermal contact with theabsorber or inlet and outlet piping. Convective losses are a linear function of the temperaturedifference between the collector glazing and ambient air. These losses can be substantial due to theeffects of wind on the outer glazing. Within the collector, convection also transfers heat to theglazing(s) from the absorber. Radiative losses are relatively small at conventional domestic water orspace heating temperatures. However, since these losses are a function of the difference betweenthe fourth power of the absorber absolute temperature and the sky absolute temperature, which isusually several degrees lower than the ambient air temperature, radiative losses may becomesignificant at higher operating temperatures. Although convective and radiative losses occur from allof the exposed surfaces of the collectors, it is a common practice to express the overall heat lossfrom the collector as a function of the absorber area, Aa. This is because the major radiative andconvective losses from a well-insulated collector occur primarily through the glazing.

For both experimental and analytical purposes, it is common to combine the convection and radiationheat transfer terms to yield a single heat loss coefficient based upon the temperature differencebetween the average collector absorber plate temperature and the ambient temperature. When thisis done, the radiation heat transfer term must be linearized. Thus;

)U+U(´F=U radiationconvectionL


and

)T-T(AU=Q apaLloss

where Uconvection and Uradiation are the overall convection heat transfer coefficient and the linearizedoverall radiation heat transfer coefficient, respectively. Tp is the average absorber temperature.

Useful energy from the collectorBy substituting the expressions developed previously for qsolar and qloss, the equation describing thequasi-steady performance of a solar collector becomes:

[ ] elanLenuseful q)TT(UG)(FAq −−−′= τα

Electrical energy extracted from collector may be approximated by a linear function of irradianceand temperature difference Tn-Ta.

[ ])TT(EG)(EAq ante0nel −−= τα

where Eo is the efficiency of PV cells and Et is their temperature coefficient. So we can write:

[ ])TT)(EUF(G)E)(F(Aq antLoenuseful −−′−−′= τα

Let tLL EUFB −′= and oeeo, E)(F −′= ταη , then the above equation can be written as

[ ])TT(BGAq anLe,onuseful −−= η

The new parameters BL and ηo,e represents the effective heat loss coefficient and the effectiveoptical efficiency, respectively.

These equations are the basic equation used in developing analytical models to describe collectorperformance. The useful energy delivered to storage or load by a collector can also be determinedexperimentally by measuring the inlet and outlet collector temperatures, the properties of the heattransfer fluid, and the mass flow rate of that fluid through the collector; thus

)T-T(C m =Q iopuseful &

where

m& = fluid flow-rate through the absorberCp = specific heat capacity of the fluid


To = outlet fluid temperatureTi = inlet fluid temperature

Collector EfficiencyThermal and electrical efficiency of the PV/T collector should be measured separately. Thermalefficiency is measured according to the ISO 9806-1 Standard and Electrical efficiency bymeasurements of I-U curves by capacitive load.

The thermal efficiency of a flat-plate solar collector is defined as the ratio of the useful heatdelivered by the collector to the total solar radiation intercepted by the collector:

collector theby dintercepteenergy solarcollectedenergy useful actual

=η

By substitution, the collector efficiency becomes:

)/GT-T(U-)(= apLeταη

The difficulty in using this equation is that the average temperature of the absorber plate, Tp, isusually unknown. The system designer does, however, know or can estimate with reasonableaccuracy the temperature of the fluid entering the collector, Ti, since the fluid temperature eitherapproximates that which comes from storage or the supply water main. Therefore, in order toprovide a more useful expression for collector efficiency, Whillier rewrote this equation substitutingTi for Tp, and introducing the collector heat removal factor, FR, to compensate for the reduced heatlosses. Hence

)/G]T-T(U-)[(F= aieR ταη

FR represents the ratio of the actual useful energy gain to the maximum possible useful energy gain.The maximum possible useful energy gain in a solar collector occurs when the whole collector is atthe inlet fluid temperature Ti. In such an ideal case, heat losses to the surroundings are at aminimum. Obviously, FR cannot exceed unity; water collectors commonly have FR values between0.7 and 0.9.

Analogously, if we apply mean collector fluid temperature the efficiency becomes:

)/G]T-T(U-)[(F= amLeταη ′

where

2T-T+T=T io

im

If F′ , FR, U, and (τα)e were constant, the collector efficiency would be a linear function of thereduced temperature difference T*.


*L0 TUF′−=ηη

where

GTT

T am* −=

In practice this it not exactly true, because UL varies with fluid and air temperature, (τα)e isinfluenced by the relative proportion of beam, diffuse, and ground-reflected radiation, and FR isweekly dependent on UL. Hence, a higher order correlation between η and T* is more realistic.

2*2

*10 GTaTa −−=ηη

where a1 and a2 are fitting constants.

The ordinate intercept, ηo, represents a measure of the ability of the collector to absorb solar energyand to transfer it to the collector fluid. If electrical energy is extracted from collector, this ability isreduced to ηo,e in accordance with the first law of thermodynamics. The slope of the equation,

LUF ′ , represents a measure of the ability of the collector to prevent heat losses to the surroundings.If electrical energy is extracted from the collector, this ability is improved to BL.

6.1.2. Simulation ProgramThe performance of a PV/T collector depends on design parameters and on various weather andoperating conditions, e.g. irradiance, ambient temperature, wind velocity, inlet fluid temperature, etc.Therefore it is necessary to develop simplified theoretical model in order to carry out multi-parameter analysis.

A module for the TRNSYS simulation program was developed in order to describe the behavior ofthis collector and compute the electrical and thermal yields of the system. The developed program isaccompanied with user friendly interface to help those without any experience to perform therequired computations. The user can simply vary the design parameters to see the response of thesystem under study, accordingly.

The basic principle of the simulation program built as a subroutine in the TRNSYS program isshown in Fig 6.3. The iterative procedure for computation of plate and cover temperature isrepeated for case without and with electrical load. The user friendly screen output is shown in figure6.4.


Figure 6.3: The flow diagram of the simulation program for PV/T collector

Computat ion of Collec tor Efficiencyand Plate Temperature w/o PV Load

with estimated plate and c over temperatures

Tp-Tp'<0.1K

Computat ion of the Heat -Loss C oef ficient

Tc-Tc'<0.1K

C ollector Yields:Heat Yield

Electr ical Energy Yield

Tc=Tc'

yes

no

no

yes

Tp=Tp'

Tp = es timated plate temperatureTp' = com puted plate tem perature

Tc = est im ated cover temperatureTc' = computed cover temperature


Figure 6.4: A sample of screen user interface for the simulation of PV/T collectors based onwater heating


6.1.3. PV/T Collector Efficiency Curves as a Function of Design ParametersIn order to show possible variation in the PV/T collector performance, the influence of the following,most influencing parameters have been taken into consideration:

• Absorptance of absorber (PV cells front surface).• Emittance of the absorber (PV cells front surface).• Quality of the thermal contact between PV and water piping (fin-efficiency factor).• Back insulation thickness.• Temperature coefficient of PV cells.• Efficiency of PV cells.

The effect of these parameters on system performance (i.e. efficiency curve) for two basic typesof collectors has been analyzed. Table 6.1 shows parameter values for these collectors (selectiveand non-selective collector). The selective absorber (PV cell might have a thin layer on the frontside with high transmittance in solar spectrum and low emittance coefficient in the infraredspectrum. The results of the analysis are shown in figures 6.5 and 6.6.

Collector Parameters Non-SelectiveAbsorber

SelectiveAbsorber

Area (m2) 2 2Emittance of Absorber Plate (-) 0.90 0.1Emittance of Cover (-) 0.88 0.88Back Insulation Conductivity (W/K/m) 0.023 0.023Back Insulation Thickness (mm) 50 50PV Reference Temperature (oC) 25 25Index of Refraction of Cover Material Glass GlassAbsorptance of Absorber Plate (-) 0.92 0.92Temperature Coefficient of PV Cells (%/oC) 0.40 0.40Collector Fin Efficiency Factor (-) 0.90 0.90Cell Efficiency (-) 0.10 0.10Thickness of Cover (mm) 4 4Extinction Coefficient of Coverplate (1/m) Glass GlassWind Velocity (m/s) 5 5Slope Angle (deg.) 50 50Flow Rate (kg/h/ m2) 3600 3600

Table 6.1: Parameters for a non-selective collector


Parameter=Fin Efficiency Factor

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,80

0,85

0,90

0,95

PV off


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,80

0,85

0,90

0,95

PV on

Parameter=Emittance of Absorber Plate

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm

2/W)

ther

mal

effi

cien

cy

0,1

0,3

0,5

0,7

0,9

PV off


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,1

0,3

0,5

0,7

0,9

PV on

Parameter=Back Insulation Thickness

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

10 mm

20 mm

30 mm

40 mm

50 mm

PV off


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

10 mm

20 mm

30 mm

40 mm

50 mm

PV on

Figure 6.5: Thermal Efficiency of the PV/T collector as a function of reduced temperature T*

for several parameters of a non-selective collector


Parameter=Absorptance of Absorber Plate

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,80

0,90

0,92

0,95

PV off


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,80

0,90

0,92

0,95

PV on

Parameter=Thickness of Back Cover

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

2 mm

3 mm

4 mm

5 mm

PV off


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

2 mm

3 mm

4 mm

5 mm

PV on

Parameter=Temperature Coefficient of PV Cells

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,1

0,2

0,3

0,4

0,5

PV on

Parameter=PV Module Efficiency

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,05

0,10

0,15

0,20

PV on

Figure 6.5: (continued … )



0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,80

0,85

0,90

0,95

PV off


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,80

0,85

0,90

0,95

PV on


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,1

0,3

0,5

0,7

0,9

PV off


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,1

0,3

0,5

0,7

0,9

PV on


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

10 mm

20 mm

30 mm

40 mm

50 mm

PV off


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

10 mm

20 mm

30 mm

40 mm

50 mm

PV on

Figure 6.6: Thermal Efficiency of the PV/T collector as a function of reduced temperature T*

for several parameters of a selective collector



0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,8

0,9

0,92

0,95

PV off


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,8

0,9

0,92

0,95

PV on

Parameter=Temperature Coefficient of PV Cells

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,10%

0,20%

0,30%

0,40%

0,50%

PV on

Parameter=PV Module Efficiency

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

0,05

0,10

0,15

0,20

PV on


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm

2/W)

ther

mal

effi

cien

cy

2 mm

3 mm

4 mm

5 mm

PV off


0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08

T*(oCm2/W)

ther

mal

effi

cien

cy

2 mm

3 mm

4 mm

5 mm

PV on



The conclusions on this analysis are summarized as:

Absorber (PV cells) coefficient of absorptance:The absorber (PV cells) coefficient of absorptance influence optical efficiency of the collector. Byhigher coefficient of absorptance, collector is being more efficient for all values of reducedtemperatures T*. Coefficients of absorptance of crystalline Si solar cells is satisfactory, itapproximately equals 0.9.

Absorber (PV cells) emittance:The absorber (PV) coefficient of emittance affects the heat loss coefficient of collector. In thissense, by lower coefficient of emittance, collector is being more efficient at higher reducedtemperatures T*. Coefficients of emittance of crystalline Si solar cells is approximately equal to 0.9(as for black paint). Reaching of low coefficient of emittance would require special, transparentthin-film layers to be applied on PV cells' front surface.

Collector fin-efficiency factor:Collector fin-efficiency factor is a key value in determining collector efficiency. Qualitatively itrepresents the degree of thermal contact between absorber and the water circulating in the collectorpiping. In the PV/T collector, where the PV cells are being pasted to the classical metal absorber,the fin-efficiency factor is reduced, as the thermal contact (absorber-water) is weaker. The thermalcontact between the PV cells and the absorber is essential for PV/T collectors to improve thethermal yield as it can be noticed from the respective figures.

PV-cell efficiency:Respective figures show the influence of cells efficiency on the thermal efficiency: lower PV cells'efficiency - higher thermal efficiency - according to the energy balance equation.

Thickness of the coverplate:The thickness of cover is an influencing parameter effecting mainly the optical efficiency of thePV/T collector.

Back insulation thickness:Back insulation thickness influences heat losses at operating temperatures higher than ambienttemperature.


6.1.4. Annual Performance of Water-Heating PV/T Collector

Influence of collector thermal heat capacity on the Yearly Energy YieldIn order to determine necessity of using thermal capacity in the simulations, an analysis of theinfluence the thermal capacity has on the accuracy of energy yield computation has been carriedout. The analysis is done by consecutive annual simulations using thermal capacities of thermalcollectors (10-20 kJ/K). Two inlet temperatures has been applied; 10 oC and 50 oC. Analysis showsthat the influence is higher for temperature about 50 oC as the automatic control unit more ofteninterrupts the pump operation. However, even for unusually high thermal capacities of 20 kJ/K, thisinfluence does not reach 1%.

The negligible influence which the thermal capacity has on the annual performance of PV/Tcollectors, as shown in the Tables 6.2 and 6.3, arguments for that further analysis can be carried outneglecting the collector thermal capacity. The main benefit is the reduced time necessary forsimulation.

Thermal Capacity of the PV/T Collector [kJ/K] Energy Yield under TRY-Denmark [kWh]0 184810 184920 1851

Table 6.2: Influence of thermal capacity on the energy yield (fluid inlet temperature: 10 oC)

Thermal Capacity of the PV/T Collector [kJ/K] Energy Yield under TRY-Denmark [kWh]0 63410 63620 639

Table 6.3: Influence of thermal capacity on the energy yield (fluid inlet temperature: 50 oC)

Annual Energy YieldsThe analysis carried out in the previous section refers to thermal efficiency only. It is thereforeimportant to carry out analysis about energy yield under real operating conditions. The energy yieldof the PV/T collector is computed under the Danish Test Reference Year (TRY). In order toenable comparison between various constructive parameters, the computation of energy yield hasbeen performed for constant inlet fluid temperature.

The influence of the following parameters have been taken into account:

• Absorber (PV) absorptance• PV-cell efficiency• Wind velocity• Inlet fluid temperature

In order to carry out parameter analysis, a reference PV/T collector has been selected as shown inTable 6.4. The simulation results are shown in Figure 6.7. Table 6.5 shows sum of electrical andthermal annual energy yield as a function of two variables; the coefficient of absorptance (α) andthe coefficient of emittance (ε).


Geographical dataSite CopenhagenLatitude 55.4oNLongitude 12.4oEGround reflectance 0.2

PV parametersCell type Mono SiCell efficiency 10 %Cell reference temperature 25 CPower temperature coefficient -0.4 % / K

Thermal consumer dataInlet fluid temperature 30o CFlow rate 100 kg/h/m2

Thermal parametersCover material Glass AR nCollector area 2 m2

Absorber emittance 0.16Cover emittance 0.88Back insulation conductivity 0.1 W/KBack insulation thickness 50 mmPlate absorptance 0.90Collector fin efficiency factor 0.90Cover thickness 4 mmWind velocity 2 m/s

Table 6.4: Reference values of collector design parameters


System output vs. absorber emittance

0200400600800

1000120014001600

0,10 0,30 0,50 0,70 0,95

absorber emittance

ener

gy (

kWh/

year

)

PV+T

PV/T

PVT

PVPURE

System output vs. plate absorptance

0200400600800

1000120014001600

0,60 0,70 0,80 0,90 0,95

plate absorptance

ener

gy (

kWh/

year

) PV+T

PV/T

PVT

PVPURE

System output vs. collector fin efficiency factor

0200400600800

1000120014001600

0,825 0,85 0,875 0,9 0,925 0,95

collector fin efficiency factor

ener

gy (

kWh/

year

) PV+T

PV/T

PVT

PVPURE

Figure 6.7: Effect of system parameters on the energy yield under reference conditions statedin

Table 6.4PV+T: electrical and thermal energy generated by separated PV and Thermalcollectors, PV/T: electrical and thermal energy generated by a PV/T system


PVT: electrical energy generated by PV cells in a PV/T systemPVPURE: electrical energy generated by PV cells placed outside of collector


System output for different cell-types

0200400600800

1000120014001600

mono Si poly Si a-Si

cell-type

ener

gy (

kWh/

year

) PV+T

PV/T

PVT

PVPURE

System output vs. cover thickness

0200400600800

1000120014001600

2,0 3,0 4,0

cover thickness (mm)

ener

gy (

kWh/

year

)

PV+T

PV/T

PVT

PVPURE

System output vs. wind velocity

0200400600800

1000120014001600

1,0 2,0 5,0

wind velocity (m/s)

ener

gy (

kWh/

year

) PV+T

PV/T

PVT

PVPURE



System output vs. inlet fluid temperature

0

500

1000

1500

2000

2500

10,0 30,0 50,0 70,0

inlet fluid temperature (oC)

ener

gy (

kWh/

year

) PV+T

PV/T

PVT

PVPURE

System output vs. collector tilt angle

0200400600800

1000120014001600

30,0 40,0 50,0 70,0 90,0

collector tilt angle (deg.)

ener

gy (

kWh/

year

) PV+T

PV/T

PVT

PVPURE

System output for high and low absorber emmitance vs. inlet fluid temperature

0

500

1000

1500

2000

10 30 50 70

inlet fluid temperature (oC)

ener

gy (

kWh/

year

)

Absorber emmitance=0.10

Absorber emmitance=0.95



Ti = 10 oC Ti = 30 oC Ti = 50 oC Ti = 70 oCα ε PV/T PVT PV/T PVT PV/T PVT PV/T PVT

kWh kWh KWh kWh kWh kWh kWh kWh0.60 0.10 1222.0 320.5 773.6 298.0 501.3 275.2 335.9 252.30.60 0.30 1233.0 320.8 747.5 298.3 467.0 275.5 308.2 252.70.60 0.50 1246.0 321.2 717.8 298.6 431.2 275.9 283.7 253.20.60 0.70 1263.0 321.5 683.6 299.0 393.8 276.3 265.4 253.70.60 0.95 1292.0 321.9 633.6 299.4 347.0 276.9 255.2 254.40.70 0.10 1390.0 314.6 922.3 292.6 615.4 270.1 411.1 247.60.70 0.30 1400.0 315.0 893.1 292.9 573.5 270.5 370.8 248.10.70 0.50 1413.0 315.4 859.5 293.3 528.1 271.0 331.7 248.60.70 0.70 1429.0 315.8 820.4 293.7 479.0 271.4 296.0 249.10.70 0.95 1455.0 316.3 761.4 294.2 413.6 272.0 262.6 249.90.80 0.10 1553.0 309.0 1071.0 287.3 736.6 265.2 500.3 243.10.80 0.30 1562.0 309.4 1039.0 287.7 688.2 265.7 448.9 243.60.80 0.50 1574.0 309.9 1002.0 288.2 635.1 266.2 397.3 244.20.80 0.70 1589.0 310.3 959.3 288.6 576.4 266.7 346.2 244.80.80 0.95 1615.0 310.9 893.7 289.2 494.9 267.4 288.9 245.60.90 0.10 1711.0 303.5 1217.0 282.3 861.8 260.5 598.8 238.70.90 0.30 1721.0 304.0 1184.0 282.7 807.8 261.0 538.0 239.30.90 0.50 1732.0 304.5 1144.0 283.2 747.9 261.6 474.4 239.90.90 0.70 1746.0 305.0 1098.0 283.7 681.0 262.1 410.2 240.60.90 0.95 1771.0 305.6 1027.0 284.3 585.7 262.8 331.4 241.40.95 0.10 1789.0 300.9 1290.0 279.8 925.0 258.2 650.2 236.60.95 0.30 1798.0 301.4 1255.0 280.3 868.6 258.8 585.3 237.20.95 0.50 1809.0 301.9 1215.0 280.8 805.7 259.3 516.5 237.80.95 0.70 1823.0 302.4 1167.0 281.3 735.1 259.9 446.0 238.50.95 0.95 1847.0 303.1 1093.0 281.9 633.8 260.6 357.3 239.4

Table 6.5: System annual yield for different combinations of absorptance coefficient (α,emittance coefficient (ε) and Inlet fluid temperature (Ti) at 10, 30, 50 or 70 oC.

PV/T is the electrical and thermal energy generated by the system andPVT is the electrical energy generated by PV cells

6.1.5. SummaryThe coefficient of absorptance for the absorber (PV cells) influences significantly the opticalefficiency of the collector in such way that the higher absorptance, the higher efficiency for thecollector. Coefficient of absorptance for crystalline Si PV cells is satisfactory high (mean valueabout 90 - 94% [1].

The coefficient of emittance for the absorber (PV cells) affects the heat loss coefficient withincreasing operating temperature and the lower emittance the higher the efficiency of the collector.Coefficients of emittance of crystalline Si PV cells is approximately 0.9 (as for selective metalabsorber). Reaching of low coefficient of emittance would require special, transparent thin-film


layers to be applied on PV cells' front surface. If PV/T collector is used as a pre-heater to anothercollector coupled in series, the high coefficient of emittance is sufficient.The thermal contact between PV-and the absorber is essential for PV/T collector to improve thethermal yield as it can be noticed in Figs. 6.5 and 6.6. Special attention is to be paid that this thermalcontact will be as good as possible.

The analysis of energy yields shows that maximum energy yield can be obtained under twoconditions:

• The coefficient of absorptance for the absorber/PV cells should be as high as possible.

• The coefficient of emittance for the absorber/PV cells at 0.1 produces the best performance forfluid inlet temperatures 30, 50 and 70 oC. However, for inlet fluid temperature of 10 oC, the bestresults are achieved with coefficient of emittance of 0.95. This result was expected, as theefficiency of solar thermal collectors with selective absorber is lower than that for solar thermalcollectors with non-selective absorber.

Taking into account that the coefficient of emittance of Si PV cells is approximately 0.9, PV/Tcollectors with Si PV cells will perform better when operating at lower inlet fluid temperatures. Onepossibility to force lower fluid temperatures is to apply a PV/T collector as a pre-heater collector toanother collector coupled in series. The second, more efficient collector at higher reducedtemperatures, may be a collector with selective coating absorber, vacuum collector, etc.

Ratio absorptance/emittance of 0.95/0.1 gives app. 18% more energy than the ratio 0.90/0.95 fortemperature level of 30 oC, app. 50% more energy for 50 oC and approximately 80% more energyfor temperature level of 70 oC. In opposite, for temperature level of 10 oC, ratio of 0.95/0.1 wouldgive approximately 5% less energy than ratio 0.95/0.95.

6.1.6. Verification of the Computer Program AccuracyIn order to verify the developed simulation program, a prototype of a PV/T collector manufacturedby RACELL was tested. The RACELL PV/T collector has the specifications presented in Table6.6. The obtained results were compared with the simulated results.

Area 2.2 m2

Emittance of PV cells 0.95Emittance of coverplate 0.88Back Insulation Conductivity 0.023 K/WmBack Insulation Thickness 50 mmIndex of Refraction of coverplate 1.526PV Reference Temperature 25 °CAbsorptance of PV Cells 0.93Power temperature coefficient of PV cells -0.40 % / KCollector Fin-Efficiency Factor 0.89Module Efficiency at Reference Conditions 0.08Thickness of coverplate 4 mmExtinction Coefficient of coverplate 1.53Wind velocity 5 m/s

Table 6.6: Reference values applied for simulation of efficiency curve of RACELL PV/Tcollector


The results for system thermal performance are shown in Figures 6.8 and 6.9, which reflect a goodagreement between measured and calculated results. This means that the TRNSYS program canaccurately predict the performance of the water-heating collector. The heat loss coefficient is lowerin comparison with measured values for the case without electrical load. The reason for this may bein higher wind velocities during tests at higher reduced temperatures, T*. Theoretically, the slopeshould be lower as by higher reduced temperatures the efficiency of PV cells decreases resulting ina relatively higher thermal efficiency.

Figure 6.8: Comparison of the efficiency curves without electrical load for RACELL collector

Figure 6.9: Comparison of the measured and simulated efficiency curves (with electricalload) for RACELL collector

measuredη = -2.1508T*2 - 6.5714T* + 0.692

R2 = 0.9847

simulatedη = -21,804T*2 - 5,5003T + 0,6702

R2 = 1

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

-0,01 0,005 0,015 0,025 0,035 0,045 0,055 0,065 0,075

T*[Km2/W]

Eff

icie

ncy

measuredcalculated

measuredη = -18.338T*2 - 6.4965T* + 0.6193

R2 = 0.988

simulatedη = -23,482T*2 - 5,2139T + 0,609

R2 = 1

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

-0,01 0,005 0,015 0,025 0,035 0,045 0,055 0,065 0,075

T*[Km2/W]

Eff

icie

ncy

measuredcalculated


6.2 PV/T - WallIn this chapter the performance of a so-called PV/T wall is described. The basic construction isshown in Figure 6.10. Two different modes of operation are considered:

• PV panels are situated on the cover and fully absorb incident radiation• PV panels are situated on the cover and are partly transparent to incoming radiation

Figure 6.10: Construction of the PV-Wall with a cooling duct behind

Hot Air into Building

Forced Circulation

PV Cells

Wall


6.2.1. Theoretical considerationsPV/T wall performance has been simulated by dividing the PV/T wall in a serial connected thermalnodes. Number of nodes depends on the height of the PV/T wall and the temperature gradient,which may be expected. It is assumed that PV strings are situated in a way that same operatingtemperature is valid for a single string (e.g. the same PV/T-wall-node). In Figure 6.11, theequivalent thermal network of a single node of PV/T-wall is presented.

Figure 6.11: Equivalent thermal network of single node of the PV/T Wall

As the thermal capacity for PV panels and cooling duct may be neglected, the governing equationsfor computation of PV/T-wall performance may be written for each node of the multi-node PV/Twall as follow:

pvc1f1crprctatt EThThThThTUTU)-k1S( =−+−+−+

0ThThThThTUTUSk prcrp2f2pbabt =−+−+−+

usefulc1f1p2f2 qThThThTh =−+−

where

S = incident hemispherical irradiance taking into account incident angle modifie rkt = light transparency coefficient of the PV moduleUt = top heat loss coefficient of the PV modulehr = radiation coefficient between two parallel plates of the cooling ducth1 = conductive-conduction coefficient between PV cells and circulating airh2 = conductive-conduction coefficient between back surface of cooling duct and the

circulating airTc = temperature of the PV moduleTp = temperature of the cooling ductTf = air temperature of the circulating airEpv = the electrical energy extracted from each node of the PV/T wallUb = back heat-loss coefficient of the cooling duct

T a

u t

S(1-k t)

E pv

h 1

T f q useful

h 2

T p

h r

ub

Tc


Radiation coefficient, hr is approximately computed using mean temperature of the both cooling-ductplanes:

)()3 11/e1/273.15(T4h dpvfr −++= εσ

where

σ = Stefan-Boltzman constantεpv = coefficient of emittance of PV back surface where air is circulatingεd = coefficient of emittance of the plane of cooling duct situated opposite to the PV

panel

The convection coefficients h1 and h2 are computed using the following relation:

cdair8.0

2,1 A/kRe0158.0h =

where

Re = Reynolds number computed on the basis of the applied air flow-rateAcd = distance between the plateskair = thermal conductivity of air

6.2.2. Simulation ProgramOn the basis of the theoretical model given in the previous section, a subroutine for the TRNSYSprogram has been built. Figure 6.12 shows the basic principles of the program. The program isiterative as temperature of PV module must be estimated in the first call and afterwards comparedwith computed values using realistic heat-loss coefficient values. A sample of user friendly outputscreen is shown in Figure 6.13.

6.2.3. System Annual YieldThe analysis is performed for three separate energy yields of the PV/T wall:

• Eel, electrical energy extracted all over the Danish Test Reference Year (TRY).• Q1, thermal energy for space heating delivered by PV/T wall in the period from 1st October to

30th April; such that the ambient temperature (Ta) is lower than 12 oC and the outlet airtemperature, To, is greater than 20 oC.

• Q2, thermal energy for water heating delivered by PV/T wall in the period from 1st May to 30thSeptember; such that the outlet air temperature, To, is greater than 10 oC.

In this analysis Es is the system total yield and ηel, η1, η2, and ηs are the system efficiency due toEel, Q1, Q2, and Es, respectively.


Figure 6.12: The flow diagram of the simulation program for PV/T wall

Computation of IAM and Optical Eff ic iency

Estimat ion of Plate Tem perature Tp andC omputation of Plate Temperature Tp'

Tp-Tp'<0.1K Tp=Tp'no

Computat ion of Heat -Los s Coeficients

Y es

Estimat ion of Cover Temperature Tc andCom putation of Cover Tem perature Tc'

Tc-Tc '<0.1KTc=Tc'

PV/T Wall Yields:Heat Yield

Electr ical Y ield

no

Yes

Estimation of Heat Los s Coeff ic ients due t o C onvection, Conduct ion and Radition


Figure 6.13: User interface for the simulation of PV/T-Wall collectors based on spaceheating


The multi-parameter analysis has been carried out using two reference constructions of the PV/T-Wall. The first reference (REF-1) assumes a small light transparency for the solar spectrum and thesecond reference (REF-2) assumes a light transparency of 80%. In this analysis, the systems areassumed to be located in Copenhagen, where the annual solar incident onto the collector plane isfound to 889 kWh/m2. The number of nodes for simulation is 10 with variable air inlet temperature.The design parameters for the two systems are shown in Table 6.7 followed by results of simulation(see Figures 6.14 - 6.19)

Design Parameters REF-1 REF-2Area (m2) 4 4Air Channel Depth (m) 0.01 0.01Light Transparency of PV-Wall 0.20 0.80Incident Angle Modifier 3 3Absorptance of PV Cells Surface 0.90 0.90Absorptance of Inner Wall 0.95 0.95Emittance of PV Cells Back Surface 0.90 0.10Emittance of Inner Wall 0.90 0.10Top Heat Loss Coefficient (from PV) (W/K/m2) 6 6Collector Width (m) 1 1Back/Edge Heat Loss Coefficient (W/K/m2) 1 1PV Cells Type Mono Si Mono SiPV Cells Efficiency 0.14 0.14PV Cell Reference Temperature (oC) 25 25

Table 6.7: Design parameters for REF-1 and REF-2


REF-1The first strategy was to determine the airflow rate, which leads to maximum annual yield. This ratewas considered for further simulations.

Figure 6.14: REF-1, system annual yield vs. air flow rate

Figure 6.15: REF-1, system efficiency vs. air flow rate

From the two figures 6.14 and 6.15 the airflow rate, which leads to maximum annual yield wasfound to be 50 m3/h/m2 for REF-1. By this analysis, the power driving the fan is not computed.

0

200

400

600

800

1000

1200

1400

1600

1800

0 10 20 30 40 50 60 70

air flow rate (m3/h/m2)

ener

gy (

kWh/

year

)Eel

Q1

Q2

Es

05

101520253035404550

0 10 20 30 40 50 60 70

air flow rate m3/h/m2

effic

incy

(%

)

η

η

η

η

s

el

1

2


Distribution of thermal energy based on the airflow rate at 50 m3/h/m2 is then found and can be seenin figure 6.16. This distribution forms a significant starting point for system designers. It helps thedesigner to choose a suitable value of the airflow rate for water heating systems.

Figure 6.16: REF-1, Distribution of useful thermal energy (Q2) for outlet temperatureTo>10oC during the period [1st May – 30th Sep.]

Air flow rate = 10 m3/h/m

2

Q2(30<To<40)15%

Q2(To>40)67%

Q2(20<To<30)12%

Q2(10<To<20)6%

air flow rate = 20 m3/h/m

2

Q2(10<To<20)8%

Q2(20<To<30)17%

Q2(30<To<40)22%

Q2(To>40)53%

air flow rate = 30 m3/h/m2

Q2(10<To<20)11%

Q2(20<To<30)23%

Q2(30<To<40)31%

Q2(To>40)35%


2

Q2(10<To<20)13%

Q2(20<To<30)29%

Q2(30<To<40)38%

Q2(To>40)20%


2

Q2(20<To<30)35%

Q2(30<To<40)40%

Q2(To>40)10% Q2(10<To<20)

15%


2

Q2(20<To<30)41%

Q2(30<To<40)35%

Q2(To>40)6%

Q2(10<To<20)18%


REF-2Determination of the airflow rate, which leads to maximum annual yield is here found for REF-2.

Figure 6.17: REF-2, system annual yield vs. air flow rate

Figure 6.18: REF-2, system efficiency vs. air flow rate

From the two figures 6.17 and 6.18 the airflow rate, which leads to maximum annual yield wasfound to be 40 m3/h/m2 for REF-2. Again the power driving the fan is not computed.

0

200

400

600

800

1000

1200

1400

1600

1800

0 10 20 30 40 50 60 70

air flow rate (m3/h/m2)

ener

gy (

kWh/

year

)

Eel

Q1

Q2

Es

05

101520253035404550

0 10 20 30 40 50 60 70

air flow rate m3/h/m2

effic

incy

(%

)

η

η

η

η

s

el

1

2


Distribution of thermal energy based on the airflow rate at 40 m3/h/m2 is then found and can be seenin figure 6.19.

Figure 6.19: REF-2, Distribution of useful thermal energy (Q2) for outlet temperatureTo>10oC during the period [1st May – 30th Sep.]

Air flow rate = 10 m3/h/m

2

Q2(30<To<40)11%

Q2(To>40)74%

Q2(20<To<30)10%

Q2(10<To<20)5%


2

Q2(10<To<20)7%

Q2(30<To<40)17%Q2(To>40)

62%

Q2(20<To<30)14%


2

Q2(10<To<20)9%

Q2(20<To<30)19%

Q2(30<To<40)24%

Q2(To>40)48%


2

Q2(20<To<30)24%

Q2(30<To<40)33%

Q2(To>40)32%

Q2(10<To<20)11%


2

Q2(20<To<30)29%

Q2(30<To<40)39%

Q2(To>40)19%

Q2(10<To<20)13%


2

Q2(20<To<30)35%

Q2(30<To<40)40%

Q2(To>40)10% Q2(10<To<20)

15%


After determination of optimum airflow rates for both systems, another study was performed inorder to determine the combination design of parameters, which will lead to highest energy yields.Table 6.8 shows the system performance under different combination of design parameters.

Table 6.8: Annual energy yields of REF-1 and REF-2 for different combination of designparameters. Shaded areas represent optimum combinations

Investigation of Table 6.8 shows that the light transparency of PV wall and the incident anglemodifier are most critical parameters of the system. In other words these parameters has a greateffect on system performance. The effect can in addition be seen in Figures 6.20 and 6.21.

Figure 6.20: Effect of PV-Wall light transparency on system yield

Figure 6.20 and Table 6.8 show that the energy yield of both systems increases as the lighttransparency of the PV wall increases. At a transparency factor of 80% this yield is 1640 kWh for

1580

1600

1620

1640

1660

1680

0 0.5 1

transparency of the pv wall

ener

gy

(kW

h/y

r.)

REF-1

REF-2

REF-1m[m 3/h/m 2] kt IAM εpv εd Ut Eel[kWh] Q1[kWh] Q2[kWh] Es[kWh]

50 0.2 3 0.9 0.9 6 344 345 908 159750 0.5 3 0.9 0.9 6 215 411 985 161250 0.8 3 0.9 0.9 6 86 490 1063 164050 0.2 4 0.9 0.9 6 344 358 947 164850 0.2 5 0.9 0.9 6 343 365 967 167550 0.2 3 0.1 0.9 6 344 342 906 159250 0.2 3 0.9 0.1 6 344 342 906 159250 0.2 3 0.9 0.9 10 346 250 795 1390

REF-2m[m 3/h/m 2] kt IAM εpv εd Ut Eel[kWh] Q1[kWh] Q2[kWh] Es[kWh]

40 0.8 3 0.1 0.1 6 86 558 1016 166040 0.2 3 0.1 0.1 6 341 394 856 159140 0.5 3 0.1 0.1 6 214 483 937 163440 0.8 4 0.1 0.1 6 86 586 1084 175640 0.8 5 0.1 0.1 6 86 603 1125 181440 0.8 3 0.9 0.1 6 86 557 1016 165940 0.8 3 0.1 0.9 6 86 557 1016 165940 0.8 3 0.1 0.1 10 86 464 913 1464


REF-1 and 1660 for REF-2. However, the electrical yield is only 86 kWh for both systems. Thismeans that a system with high light transparency factor is indeed classical air heating collector.

Figure 6.21 shows that the presence of PV cells on the front wall reduces the total energy yield byapproximately 8 % (from 1814 kWh to 1675 kWh) for a very good IAM coefficient (5). However,for an ordinary IAM coefficient (3) the difference is in the ranges of 4%.

Figure 6.21: Effect incident angle modifier on system yield

Experimental results for a PV/T Wall could unfortunately not be carried out and therefore thetheoretical results presented in ([5] Duffie & Beckmann) for an air-heating collector has been usedto validate the model described in the chapter 6.2. The difference in the computed values of heatloss coefficients was found to be less than 0.1% what might be expected as relevant physicalprocesses have been taken into account by both (theoretical) methods.

1550160016501700175018001850

0 1 2 3 4 5 6

incident angle modifier

ener

gy

(kW

h/y

r.)

REF-1

REF-2


Nomenclature

Aa collector absorber areaAcd distance between the platesAn area of a collector nodeBL effective heat loss coefficientC collector thermal capacityCn heat capacity of a collector nodeCp specific heat capacity of the fluidEel electrical energy extracted from a PV/T wallEo efficiency of PV cellsEpv the electrical energy extracted from each node of the PV/T wallEt temperature coefficient of PV cellsEs overall energy delivered by the PV/t wallFR collector heat removal factorF´ collector efficiency factor (regarding mean fluid temperature)G incident total radiation on a collector surfaceGalb incident ground reflected radiationGbeam incident beam radiationGdiff incident diffuse radiationGn heat capacity of a collector nodeh1 conductive-conduction coefficient between PV cells and circulating airh2 conductive-conduction coefficient between back surface of cooling duct and the

circulating airhr radiation coefficient between two parallel plates of the cooling ductKalb incident angle modifier for ground reflected radiationKbeam incident angle modifier for beam radiationKdiff incident angle modifier for diffuse radiationkair heat transfer coefficient of airkt light transparency coefficient of the PV modulemc mass of collector nodem& fluid or air flow rateQloss heat losses from the collectorQuseful useful energy from the collectorQ1 thermal energy delivered by the PV/T wall for space heatingQ2 thermal energy delivered by the PV/T wall for water heatingqel rate of electrical energy extracted from the collectorqheatcapacity rate of energy storage within the collectorqloss rate of energy loss from the collectorqsolar rate of solar radiation absorbed by the collectorquseful rate at which useful energy is delivered by the collectorqu,n rate of energy gain by a collector nodeRe Reynolds number computed on the basis of the applied air flow-rater parameter for incident angle modifierS incident hemispherical irradiance taking into account incident angle modifierTa air temperature in the vicinity of collectorTc temperature of the collector cover or temperature of the PV moduleTf air temperature of the circulating airTi inlet fluid or air temperature


To outlet fluid or air temperatureTm collector mean fluid temperatureTn temperature of a collector nodeTp average absorber plate temperature or temperature of the cooling ductT* reduced temperatureUb back heat-loss coefficient of the cooling ductUL overall heat loss coefficientUo overall heat loss coefficient when T = Ta and v = 0UT coefficient characterizing temperature dependence of heat coefficientUt top heat loss coefficient of the PV moduleUv coefficient characterizing wind dependence of heat coefficientv wind speed in the vicinity of collectorα absorptance of the absorber plateε emittance of the absorber plateεd coefficient of emittance of the plane of cooling duct situated opposite to the PV

panelεpv coefficient of emittance of PV back surface where air is circulatingη collector thermal efficiencyηo collector optical efficiencyηo,e collector effective optical efficiencyθ incident angleσ Stefan-Boltzman constantτ transmittance of the collector cover(τα)e effective transmittance-absorptance product


References

[1] IEA, workshop on PV/Thermal systems, Amersfoort, the Netherlands, September 17-18 1999.

[2] Hotel, Whillier, Trnsactions of the Conference on the Use of Solar Energy,2, 74, University ofArizona, 1958.

[3] M. Bosanac, A. Brunotte, W. Spirkl and R. Sizmann, Use of Parameters Identification forFlat-Plate Collector Testing under Non-Stationary Conditions, Renewable Energy Sources, 4,217-222, 1994.

[4] J.P. Ambrosetti, Das neue Bruttowaermeertragsmodel fuer Sonnenkollektoren, EIRWuerenlingen, ISBN-3-85677-012-7, 1990.

[5] Duffie & Beckman, J. Wiley &Sons, Inc, New York 1980.

[6] ISO 9060 Solar Energy - Specification and Classification of Instruments for MeasuringHemispherical Solar and Direct Solar Radiation

[7] ISO 9459 part 3 Solar Heating - Domestic Water Heating Systems; Performance Testing forSolar Plus Supplementary Systems ISO/DIS 9846 Solar Energy - Calibration of a PyranometerUsing a Pyrheliometer

[8] IEC 60904-1:1987, Photovoltaic devices - Part 1: Measurements of photovoltaic current-voltage characteristics.

[9] IEC 60904-2:1989, Photovoltaic devices - Part 2: Requirements for reference solar cells.

[10] IEC 60904-3:1987, Photovoltaic devices - Part 3: Measurements principles for terrestrialphotovoltaic (PV) of photovoltaic current-voltage characteristics.

[11] IEC 60904-6:1994, Photovoltaic devices - Part 6: Requirements for reference solar modules.

[12] QC 001002: 1986, Rules of Procedure of the IEC Quality Assessment System for ElectronicComponents (IECQ), Amendment 1 (1992).


7. Complete system modelling using NSES

7.1 Model description

7.1.1 Reference year data


7.1.2 The solar collector


7.1.3 Heat Exhange


7.1.4 Flat-plate collector with heat storage


7.1.5 Solar heat-producing systems


7.1.6 Heat pump system


7.2 Results of simulations

7.2.1 Software verification


7.2.2. Small PVT system simulation


7.2.3. Medium-size PVT system simulation


7.2.4. Large PVT system simulation


7.2.5. PVT system with heat pump


7.3 References


8. Conclusions

The main goals fulfilled within the project are:• The most promising designs of PV/T collectors have been recommended and

• The software tools aimed to manufacturers and system designers have been developed

The most promising designs of the PV/T collectors have been identified as:• A PV/T water-heating collector with the PV cells acting as absorber in a direct thermal

contact with the metal absorber containing the water piping

• A PV/T Wall where warm air produced by cooling of PV panels being utilised for spaceand/or water heating via air/water heat exchanger.

The second important output of the project is development of the software tools enabling potentialmanufacturer to optimise design of a PV/T collector. The PV/T collectors’ energy yield is functionof the position (e.g. different building orientations, different shadow conditions over the day) andoperating conditions (e.g. temperature level, consumer energy requirement and daily profile ofenergy consumption). The developed user-friendly tools may be used for simulation of PV/Tcollectors’ energy contribution under variety of construction and operating conditions.

8.1. Water-Heating PV/T collector

The effect of the most influencing parameters on system performance (i.e. efficiency curve) fortwo basic types of units (selective and non-selective absorber) has been analysed. The conclusionson the results extracted from annual energy simulations follows:

• The coefficient of absorptance for the absorber/PV cells influences significantly the thermal andelectrical efficiency of the PV/T collector. Coefficient of absorptance of crystalline Si solarcells is satisfactory high (mean value about 90 - 94%). However, manufacturer must payattention to choose the type of module with maximum coefficient of absorptance.

• The coefficient of emittance of the absorber/PV cells affects strongly the heat loss coefficient.Coefficient of emittance of crystalline-Si solar cells is approximately equal to 0.9 (similar toblack-painted absorber). Thus, reaching of low coefficient of emittance would require special,transparent thin-film layers to be applied on PV cells' front surface. That can significantlyimprove the thermal efficiency of energy conversion at operating temperatures higher than 30°C. A potential manufacturer should address two main topics: (i) technological difficulties toapply additional thin-film layer and (ii) reduction in electrical efficiency having PV-cellsoperating at high temperatures.


The analysis of energy yields shows that maximum energy yield is obtainable under the followingconditions:

• The coefficient of absorptance for the absorber/PV cells should be as high as possible.

• If the coefficient of emittance for the absorber/PV cells is about 0.1 the best performance forfluid inlet temperatures 30, 50 and 70 oC is reached. However for inlet fluid temperature of app.10 oC the best results are achieved with coefficient of emittance at 0.95. This result wasexpected, as the efficiency of solar thermal collectors with selective absorber is lower than forunits with non-selective absorber.

• The thermal contact between PV cells and the absorber is essential for a PV/T collector toimprove the thermal yield. Special attention is to be paid that this thermal contact will be as goodas possible.

Taking into account that natural coefficient of emittance of Si-PV cells is approximately 0.9, PV/Tcollectors with Si PV cells will perform better when operating at lower inlet fluid temperatures. Onepossibility to force lower fluid temperatures is to apply a PV/T collector as a pre-heater unit toanother collector coupled in series. The second, more efficient collector at higher temperaturelevels, may be a collector with selective-coating-absorber, vacuum collector, etc.

Ratio absorptance/emittance of 0.95/0.1 gives app. 18% more energy than the ratio 0.90/0.95 fortemperature level of 30 oC, app. 50% more energy for 50 oC and approximately 80% more energyfor temperature level of 70 °C. For temperature level at 10 oC, ratio of 0.95/0.1 would giveapproximately 5% less energy than ratio 0.95/0.95.

Since the efficiency of PV cells decreases with increasing temperature, the temperature of the PVcells should be kept as low as possible, giving a low operating temperature. However, as long as theoperating temperature of the PV/T collector is lower than the temperature of the PV cells if theyare stand alone units, the PV cells have a higher efficiency for a PV/T collector.


8.2. Air PV/T collector (PV/T – Wall)

A multi-parameter analysis has been carried out using two reference constructions of the PV/T-Wall indeed Danish climatic conditions. The first reference assumes PV/T unit with a small lighttransparency (app. 20%) within the solar spectrum and the second reference assumes a lighttransparency of 80%.

Our first strategy was to determine the airflow rate, which leads to maximum annual yield. For theboth units this rate was found to be between 40 and 50 m3/h/m2. Overall system efficiency wasfound to be between 45 and 47%. Although these figures are obtained under favourable operatingconditions – they show large potential of air-based PV/T units.

After determination of an optimum airflow rate for both systems, another study was performed inorder to determine the set of design parameters, which will lead to the best annual energy yield.

Analysis of annual energy yield of different PV/T-wall designs resulted in conclusion that the bestperformance is gained with PV/T unit sacking the fresh air directly from outside without closed-loopcirculation. These operating conditions are the most favourable as PV cells operate most efficient atlow ambient temperatures and efficiency of thermal conversion is maximal in that case.

Investigation showed that the light transparency of PV wall and the incident angle modifier are mostcritical parameters of the system. Further investigation is necessary to identify optimal design ofPV/T wall in dependence on consumer’s profile-demands and a thermal mass coupled behind thePV/T unit.


9. Recommendations for future work

The analysis carried out in this work clearly shows the advantages of PV/T collectors for specificoperating conditions.

In further work, it is necessary to analyse the energy contribution of the water-based PV/Tcollectors. Typical applications to be analysed and optimised are:

- Once-through-flow collectors exclusively for large systems where consumption of hotwater exists all over the daytime (hospitals, factories etc.)

- Pre-heater elements in closed loop configurations – here an additional collector (withselective absorber or evacuated tube) is necessary coupled in series to increase thetemperature level.

- The above mentioned solutions shall be analysed for the typical consumers in order tooptimise typical systems design with respect to particular load profile and/or weatherconditions. Two typical prototypes should be designed and realised in co-operation withan industry partner (e.g. Batec/Racell, Arcon etc.).

- Focus on developing PV cells having a surface with a low coefficient of emittance, e.g.non-selective absorber, resulting in a higher thermal and electrical efficiency at higheroperating temperatures. If this “problem” is solved there exist a large potential for PV/Tcollectors.

This project showed great potential of air PV/T collectors, especially the PV/T Wall. Thesecollectors may be applied as a building integrated elements. In the further research it is necessary toidentify typical designs for the building-integrated elements for different architecture requirements.The simulation program developed within the project enables wide analysis of application possibilitiesof the PV/T Wall to be carried out with involvement of architects. The simulation program has to becoupled with already developed shadow computation program to enable simulation for systemswhere shadow analysis is required.

Further investigation is required on optimal strategy of operation of the PV/T wall in scope of thewhole system (buildings, hot-water storage, water/air heat exchanger). Criteria for optimisation ofboth design and control strategy will be based on energy/exergy contribution of the PV/T wall withregard to consumer load profile – both water-heating and air-heating.

Further analysis is necessary to identify the conditions where the PV/T Wall with natural air-circulation may be applied.

Analysis of the introduction of thermal mass within the PV/T Wall to improve applicationspossibilities is one of the next investigation topics. Prototypes of air PV/T collectors should bedesigned and realised in co-operation with an industry partner (e.g. VELUX etc.).

Final Report EFP project 1713/00-0014 Photovoltaic/Thermal Solar ...

Documents

Transcript of Final Report EFP project 1713/00-0014 Photovoltaic/Thermal Solar ...