Analysis of the Fored
Ventilation in the Air Gap
Themainpurposeoftheprevioushapterswastoidentifythebestttingmodel.
In this hapter theaim is toinvestigatehowvarious onditionsin theairgap
behindthePVmoduleinuenetheheattransferandingeneraltheestimations
ofthemodels. Thisisalsoanopportunityoftestingthereliabilityofthemodel.
Changes in the onditions an reveal strength and weaknesses of the model.
Theappliedmodelinthishapteristheextendednon-linearsinglestatemodel
preditingthe temperaturesat the topofthemodule. This model hasproven
tobetheoverallbestdesribingmodelthoughithastroublespreditingduring
thedayhours.
10.1 The set-up and the data
Themodelapplied forestimationistheextended non-linearsinglestatemodel
from in Chapter8. Whileestimatingthemodelsit hasbeomelearthat,due
toutuationinthewind,theausallteredwind,
W f
,needstobexed. IfW f
isnotxedthestandarddeviationofthestateparametersbeomestoohigh. It
isthereforehosentox
W f
totheaverageofthemeasuredwindofthespeiTheparametersofthemodelsareverysensitivetovariationsinthewindspeed.
Thismeansthatthemodelshavetobeestimatedfrom datawheretheaverage
windis quitesimilar.
Twofators arehangeablein relationto the airgap: the foredveloity and
theairresistane. Thefatsofthehangeablefatorsarelistedbelow.
•
The fored veloityJust asin theprior analyses thefored ventilation is held onstantin a
24 hour period. In the analysis three veloities are examined. For the
purposesofomparisontheambientwindspeedisingeneralbelow1m/s.
1.38m/s (Level 6)
2.49m/s (Level10)
3.43m/s (Level13)
•
The air resistaneThere aretwoset-upsinordertohangetheairresistanein thegap:
Nonsresultsinafreelaminarow
Finsreatingaturbulentow
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Figure10.1: Left: Aphotoofthensinsidetheairgap intheBIPVmodule[Christ
2001℄. Right: Ashemeshowingtheforedairowinthegap.
Thetransversalns,see Figure 10.1to theleft, are plaedin thegap in order
tomakethelaminarairowturbulent. OntherighthandsideinFigure10.1is
theory onerningthebehaviourof theairow inthe gap. Priorresearh has
fousedonairowin gapswherethensarepresentatbothsidesofthegap.
Inthe artileby[Chin-Hsiang &Wen-Hsiung1990℄ theowpatternsin agap
where thensarein staggeredpositionsat eahsideof thegap aredisussed.
Othershavemaderesearhhavingthensplaedexatlyoppositeoneanother.
These studiesreveal that the inominglaminar owbeomes moreturbulent.
TheowinthegapbehindthePVmoduleisexpetedtobelessturbulentsine
aertainamountoftheairwillsueedinmovingalongthePVmodule. Inthe
aseoftheBIPVmoduleitisnotpossibletoplaensatbothsidesofthegap,
sine the nsshould then be mounted at the rear side of the module. If ns
weremountedattherearsidethismightleadto anunwantedinuene onthe
module temperature. Inthis situation alsotheheat transferoeient,whih
hastobeestimated,willbeaetedinanunfavorableway. Themountingould
aet thefuntionalityofthefragilemodule.
The set-upontainingthens inreasestheeetive heat transferareain the
airgap. Thisan leadtoaninreaseoftheheat transferfrom thePV module
to theairin thegapbehindthePVmodule[Christ 2001℄and[Bazdidi-Tehrani
&Naderi-Abadi2004℄. Comparedtotheset-upwithoutnstheset-upwithns
willleadtoahigherairtemperatureinthegapandmoreoveradereasein the
module temperature. Thiseet is favourable,sinea dereasein the module
temperatureraisestheeletrialperformaneofthemodule. Thetransferofthe
heat fromthemoduletotheventilationairisalsodesirable,iftheairismeant
to heatthebuilding.
The applied data are all 24-hour data. It is examined that the fored
velo-ity is held onstant in 24-hour periods. In the original data there was also a
foredventilationLevel7,but no24-hourperiods weremeasured. Underthese
irumstanes itwasdeidednot to applythese data. In thelightof the
pre-vious analysesit is found that 24-hour data are aeptablein order to obtain
stable and usefulmodels. Thedata aremeasuredduring aone-month period.
InAppendixA.1 atable ontaininginformation asto when theinformationis
olleted. Thismayleadtodeviationintheexternalonditionssuhas
temper-atureand windspeed. Aordingto thelogbook ofthetesting,[Gandinin.d.℄,
allthedaysweresunnydays. Theirradiationisoneoftheimportantonditions
withaviewtohavingequaltestingonditions,whendealingwithPVmodules.
Table10.1: Averagesandstandarddeviationsoftheresidualsforthemodelshaving
dierentveloitiesintheairgapwithandwithoutns
Fins Veloity Average(std.dev.)
6
− 6.301 · 10 − 4 (3.091 · 10 − 2 )
Yes 10
− 2.292 · 10 − 4 (3.144 · 10 − 2 )
13
2.206 · 10 − 4 (3.423 · 10 − 2 )
6
3.477 · 10 − 5 (3.091 · 10 − 2 )
No 10
− 1.162 · 10 − 4 (3.332 · 10 − 2 )
13