Summation

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. If}

W f

isnotxedthestandarddeviationofthestateparametersbeomestoohigh. It

isthereforehosentox

W f

^{totheaverageofthemeasuredwindofthespei}

Theparametersofthemodelsareverysensitivetovariationsinthewindspeed.

Thismeansthatthemodelshavetobeestimatedfrom datawheretheaverage

windis quitesimilar.

Twofators arehangeablein relationto the airgap: the foredveloity and

theairresistane. Thefatsofthehangeablefatorsarelistedbelow.

•

^{The fored veloity}

Just 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 resistane}

There 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

− 4.397 · 10 ^{− 5} (3.673 · 10 ^{− 2} )

In document S hai de (Sider 111-116)