A study of the potential benefits of mi-transparent photovoltaics
in commercial buildings
Konstantinos Kapsis ⇑,Andreas K.Athienitis
Department of Building,Civil and Environmental Engineering,Concordia University,1455de Maisonneuve W.,EV.6-139,Montreal,QC H3G 1M8,Canada
Received 8September 2014;received in revid form 7February 2015;accepted 17February 2015
Available online 11March 2015
Communicated by:Associate Editor Jean-Louis Scartezzini
Abstract
This study investigates the potential benefits of mi-transparent photovoltaic windows on the energy,daylighting and thermal performance of commercial buildings.A general simulation methodology is propod and utilized,integrating thermal,electrical and daylighting analysis.The impact of various bu
ilding design parameters on the lection of ideal optical properties of mi-transparent photovoltaics is examined.The potential performance of Poly-Si,a-Si/l c-Si and organic cell technologies is also studied.The lection of the module optical properties is shown to be nsitive on the daylight and lighting controls applied and photovoltaic cell technology utilized.The lection of a mi-transparent photovoltaic module with 10%visible effective transmittance resulted in the lowest annual end-u electricity consumption (as low as 5kW h/m 2/yr).Finally,simulation results suggest that high cell operating temperatures of up to 64°C could occur that might cau accelerated degradation when organic thin film technologies are ud.Ó2015Elvier Ltd.All rights rerved.
Keywords:Semi-transparent photovoltaic;Building-integrated;Energy simulation;Organic PV
1.Introduction
Effective building fac ¸ade design should contribute to the creation of a pleasant,glare-free,thermally comfortable environment that will reduce building energy expenditures and optimize daylight utilization (Boyce et al.,2003).In most commercial and high-ri residential buildings,where reducing the costs of cooling energy is important,an inte-grated strategy to control the transmission of solar ra
diation needs to be adopted.Rather than having reflective,tinted or fritted windows to reduce solar transmission,mi-transpar-ent photovoltaic windows may be ud to reduce solar heat gains and generate solar electricity (Bahaj et al.,2008;James et al.,2009;Qiu et al.,2009),while still provide adequate
daylighting and view to the outdoors (Vartiainen,2001).The term Semi-Transparent Photovoltaic (STPV)is ud here to cover a broad range of PV technologies,from Si-bad cells (arranged in such a way as to allow light to pass through the resulting space between the opaque cells)(Baum,2011)to “e-through ”thin films (Lynn et al.,2012),such as a-Si/l c-Si (Klein et al.,2012;Sai et al.,2014),organic PV (Krebs,2009;Li et al.,2012)and perovskites (Eperon et al.,2014;Snaith,2013).As STPV technologies are penetrating the building industry,they are expected to play a key role in on-site electricity generation of new and retrofitted high-performance commercial and institutional buildings;on-site electricity generation can partly offt daily electricity consumption,eliminate grid transmission loss and potentially contribute to grid “peak demand shaving ”,resulting in reduced need for peak-capacity power plants (Athienitis and O’Brien,2015).
dx.doi/10.1016/j.solener.2015.02.016
0038-092X/Ó2015Elvier Ltd.All rights rerved.
⇑Corresponding author.Tel.:+15148482424x7080.
E-mail address:c_dia.ca (K.Kapsis)./locate/solener蓝莓之夜百度影音
Available online at
ScienceDirect
Solar Energy 115(2015)
120–132
STPV windows can be utilized to cover large skylight and fac ¸ade surfaces and regulate solar heat gains and day-light (Bizzarri et al.,2011;Roberts and Guariento,2009).In order for this to be done effectively,and to ensure mar-ket acceptance of this technology within the building indus-try,the solar,optical and thermal properties of STPV windows and their effect on a building energy performance need to be studied and quantified.Issues such as heat management (STPV cells overheating and non-uniform temperature distributions between cells located near cen-ter-of-the-window and cells located near-the-edge results in electrical efficiency reduction and STPV window thermal stress),visual and thermal performance as well as cost and durability (at least twenty years of performance)have been shown to be as important as STPV electricity production (Chow et al.,200
7;Fung and Yang,2008;Vats et al.,2012);e.g.,the lection of STPV optical properties has a direct impact on STPV electrical performance,solar heat gains and daylight availability within the building (Fig.1).Improper quantification of such interactions will not only affect the predicted energy u but also misguide the designer toward a sub-optimal design,which will impact ea of functionality,ea of installation and result in an unfavourable adoption of this technology by the end-ur.
The objective of this study is to investigate the potential benefits of STPV windows on the building energy,daylight-ing and thermal performance through the lection of the STPV optical properties.The study focus on cooling dominated commercial building perimeter zones in a conti-nental climate region (Southeastern Canada and Northeastern United States).The end goal of this work is to provide input to the design of cost effective,high perfor-mance STPV windows with optical,electrical,and thermal properties suited to commercial building fac ¸ade applica-tions through a general design methodology that could
be easily followed by architects and engineers during the
preliminary design stage,when there is an opportunity to have the greatest impact on the final design.This differs from standard practice which typically involves the u of energy modeling at the
end of the design process when there is little opportunity to make design changes.2.Brief overview of existing STPV performance studies Fac ¸ade orientation and Window-to-Wall ratio (WWR)play major roles on the annual STPV electricity yield,solar heat gains and daylight availability.In general,near equatorial facing fac ¸ades have the highest annual solar potentials for electricity generation.Equatorial facing sky-lights with a tilt angle near to the altitude of the building site,tend to maximize the electricity yield.Whenever opti-mal orientation is not possible due to site constrains,the STPV windows should face preferably anywhere between ESE and WSW for the north hemisphere (between ENE and WNW for the south hemisphere).When considering the impact STPV windows have on the building energy per-formance,it was shown that the lection of the ideal STPV optical properties was independent of the building orienta-tion (Chow et al.,2007;Miyazaki et al.,2005;Robinson and Athienitis,2009),within this orientation range.Though,Miyazaki et al.(2005)and Ng et al.(2013)showed,through simulations,that the WWR has an impact on the lection of ideal STPV optical properties.Independently of the fac ¸ade configuration and orienta-tion,it is imperative that the STPV module should be the outermost glass layer of a window asmbly.Delisle (2008)demonstrated that by moving the STPV module from being the outermost glass to the middle glass layer of a triple glazed window,electrical yield reduction of up to 22%was predicted,caud mainly by the reduction of transmitted solar radiation to the PV cells.In addition,
酒店面试问题及回答HVAC
Dimmable lighting
Thermal mass
Occupants
Opaque spandrel (w or w/o BIPV)
Solar shading Solar gains (daylight)
Heat gains
Workplanemarch是什么意思
Framing
View ction (w or w/o STPV)
Daylight ction (w STPV)feel like的用法
Electricity generation A schematic of major interactions between the STPV windows,the office space and the occupants.
K.Kapsis,A.K.Athienitis /Solar Energy 115(2015)120–132121
operating cell temperatures of up to16°C higher were anticipated,despite the fact that the cavity between outer-most glass and STPV layer was naturally vented to out-doors to avoid high temperatures.
Park et al.(2010)showed,through experimental work, that highly absorptive STPV module ,col-ored or tinted glass,ud for aesthetic purpos and reduc-tion of solar heat gains)should be avoided,as they can result in PV cell overheating.
The thermal performance of STPV is an area that needs attention becau it exerts a significant influence on the durability of the STPV and other window components, such as spacers,alants and framing.High temperatures need to be predicted,either through testing and/or simula-tion.The allowable temperature ri depends on the STPV technology ,organic,Si-bad,etc.).De Boer and van Helden(2001)predicted PV cell operating temperatures of up to65°C on an office STPV window while Wong et al.(2008)predicted temperatures up to 75°C on STPV skylight
s;Temperatures up to60°C were measured,with no significant temperature gradient,on a commercial building STPV window(Yoon et al.,2011).
Finally,the daylighting and electric lighting controls implemented strongly affected the lection of ideal STPV optical properties(Robinson,2011;Wong et al.,2008), whereas the lection of the HVAC system had no impact (Chow et al.,2007;Miyazaki et al.,2005).
3.Methodology
For this study,an integrated simulation approach(ther-mal,electrical and daylighting)is followed,propod by Garde et al.(2011)and Reinhart and Wienold(2011)and adapted to the needs of the study(Fig.2).Major procedure steps are described below,with“Step2”to“Step5”auto-mated through MATLAB(MathWorks,2014).
Step1:An office model is built using SketchUp3D soft-ware(Trimble,2014).The geometric model is ud to generate(i)a Radiance-bad model through the su2ds plugin(Kjenner,2014)that is imported to Daysim soft-ware(Reinhart,2014)for the office annual daylight-ing/lighting performance and(ii)an energy model through the Legacy OpenStudio plugin(National Renewable Energy Laboratory,2014)that is imported to EnergyPlus software(Lawrence Berkeley National Laboratory,20
15)for the office annual thermal and electrical performance.
Step2:An EnergyPlus Weather(EPW)file is imported to DAYSIM and converted from1-h time-step to5-min time-step(Reinhart and Walkenhorst,2001).The new weatherfile is ud as input weatherfile for Perez “all-weather”sky model(Perez et al.,1990).The simula-tion time-step was lected bad on daylight avail-ability.Walkenhorst et al.(2002)suggested a1-min time-step instead of a60-min time-step in order to reduce errors due to the short-term variability of
daylight.However,the systematic underestimation was found to be in the range of6–18%,while Janak(1999) found a difference of less than3%between the two simulation time-steps.Considering the fact that occu-pants tend to be infrequent shade urs(O’Brien et al., 2013;Van Den Wymelenberg,2012)and that daylight and occupancy nsors tend to respond in the range of 5-min to15-min time-steps in order to avoid occupant disturbance,a5-min time step was considered satisfactory.
Step3:A custom-made MATLAB routine is ud to generate manually-controlled roller shade u schedule, bad on occupant behavioral statistical models devel-oped by Reinhart and Voss(2003)(when assumed“ac-tive urs”)and Kapsis et al.(2013)(when assumed “inactive urs”).The schedul
e is fed to DAYSIM that performs the annual daylight analysis(Reinhart and Walkenhorst,2001).
Step4:Lightswitch-2002(Reinhart,2004)routine embedded in DAYSIM is ud to generate the electric lighting u schedule(either for“active”or“inactive”
urs,)bad on the workplane illuminance levels due to daylight,calculated at“Step3”.
Step5:The various Schedules are then fed to EnergyPlus that performs the annual energy analysis.
An outputfile is generated comprising the office energy consumption for cooling,heating,electric lighting and equipment,the PV electricity generation,the PV cell temperature profile,the solar heat gains and loss through the STPV window,and others.
The aforementioned methodology can be applied using alternative building performance simulation tools and extended to different advanced fac¸ade , electrochromic windows,windows integrating advanced coatings and/or shading devices).
4.Simulation study of a cooling-dominated office utilizing STPV windows
A study was carried out for a cooling-dominated perimeter office zone utilizing STPV windows,adopti
ng the above methodology.The office was located in Toronto,ON,Canada(latitude43.7°N).Major office mod-eling assumptions are summarized in this ction,while detailed subctions on STPV window daylighting,thermal and electrical modeling will follow.
The zone dimensions were4m(width)Â5m(depth)Â3.2m(height).Two WWR were studied:WWR=40% and WWR=60%,respectively(Fig.3).An exterior wall of U=0.301W/m2K was considered.The spandrel and mullion thickness of0.15m was taken into account during daylighting and energy performance simulations.The interi-or walls were assumed to connect with similarly conditioned zones to the office of interest.A medium-weight concrete floor was ud,while furnishings was reprented as internal zone surface area expod to the zone air.
122K.Kapsis,A.K.Athienitis/Solar Energy115(2015)120–132
An occupancy density of10m2/occupant was consid-ered,assuming plug loads of100W/occupant.Fig.4pre-nts the occupancy and plug load schedules(CEC, 2008),from Monday to Friday.An ideal heat recovery ven-tilation system with constant thermal efficiency of80%was taken into account,with ventilation rates of2.5L/s-occu-pant and0.3L/s-m2offloor area(ASHRAE,2004).The infiltration rate was kept constant at0.15L/s-m2of exteri-or surface area.
mtn
An ideal air system,that meets cooling and heating loads at all times,was ud.The annual office end-u elec-tricity consumption was estimated by converting the ther-mal energy for heating and cooling to equivalent electric energy,assuming constant Coefficients Of Performance
Fig.3.(Left-to-right):Office zone schematics with WWR=40%and WWR=60%,utilizing STPV windows.
K.Kapsis,A.K.Athienitis/Solar Energy115(2015)120–132123
(COP)for cooling (COP cooling =3)and heating (COP heating =4),throughout the year as follows:
where E yr is the annual end-u electricity consumption
(kW h/m 2yr),A floor is the office floor area (m 2)and,E cooling (t ),E heating (t ),E lighting (t ),E plug (t )and E STPV (t )is the office cooling load,heating load,electric lighting,plug load and STPV electricity production (kW h)at a time step t ,respectively.环评师
4.1.STPV window daylight modeling
The annual daylight performance of various STPV win-dow configurations were studied using DAYSIM,a Radiance-bad software.The Radiance simulation para-meters ud for the analysis are
48个国际音标发音
summarized in Table 1.The STPV windows were treated as if they had uniform optical properties.The effective transmittance (under visi-ble and solar spectrum)are summarized in Table 2,for both the STPV module (outermost glass of the STPV window)and the STPV window (Insulated double-Glazed window Unit).Five effective visible transmittance values of the STPV module were simulated:10%,20%,30%,40%and 50%.The minimum value of 10%was lected in order to ensure a certain minimum view to the outdoors.Floor,interior walls and ceiling were treated as perfectly diffu (Lambertian)surfaces with visible reflectance of
台词英文
20%,60%and 80%,respectively.The corresponding ther-mal and electrical properties of the STPV window can be found at the related subctions below.
A roller shade was considered and treated as a translu-cent object with direct hemispherical transmittance of 5%and diffu reflectance of 80%.It should be noted that both DAYSIM and EnergyPlus simulate a shade that is either ful-ly open or fully clod,but not partly-clod.Obrvational studies have shown that occupants will less likely move their roller shades from fully open to fully clod positions or vice versa (Inoue et al.,1988;Rubin et al.,1987).In most cas,the shade movements are incremental.Thus,the office mod-el considered intermediate shade positions by “parating ”the roller shade into four smaller shades.This way a single shade was repli
cated as if it was able to be controlled in 5-po-sitions [0(fully open),0.25,0.50,0.75and 1(fully clod)].In all cas,an abnce nsor was ud;the nsor switches the lights off,with a 5-min delay,when occupants exit the room and does not turn the lights on when occu-pants enter the room.The abnce nsor was coupled with a continuous dimming control nsor,utilizing an ideally commissioned photocell,to maintain minimum workplane (0.8m above floor)illuminance level at 500lx during occu-pied hours;the nsor dims the lights to complement day-light and maintain minimum workplane illuminance levels,and switches the lights offwhen the minimum illuminance levels are met by daylight alone.4.2.STPV window thermal modeling
Energy balance equations were employed to estimate the thermal performance of a STPV window,using EnergyPlus.A heat transfer model adopted for a double-glazed STPV
Table 1泰语学习网站
Radiance simulation model parameters.Ambient bounces Ambient division Ambient sampling Ambient resolution Ambient Accuracy Direct threshold 7
1500
20
300
yingyuyinbiao0.1
E yr ¼X t
COP cooling E cooling ðt ÞþCOP heating E heating ðt ÞþE lighting ðt ÞþE plug ðt ÞÀE STPV ðt ÞÂÃ
A floor kW h
m 2yr
ð1Þ
124
K.Kapsis,A.K.Athienitis /Solar Energy 115(2015)120–132
