stateoftheart

State-of-the-artof

windenergyincoldclimates

,nen,n,

ug,y,-Gould,

x,a,in

April,2003

ABSTRACT

Windturbinesincoldclimatesrefertositesthathaveeithericingeventsorlow

ational

EnergyAgency,IEAR&DWindhasstartedanewannex,WindEnergyinCold

aninternationalcollaborationongatheringandprovidinginformation

listomonitor

reliabilityofstandardandadaptedtechnologyandestablishguidelinesforapplying

report,thestate-of-the-artofarcticwindenergyis

prented:knowledgeonclimaticconditionsandresources,technicalsolutionsinu

andoperationalexperienceofwindturbinesincoldclimates.

5

CONTENTS

1Introduction..................................................................................................................8

2Windturbinesincoldclimates..................................................................................10

2.1NorthernEurope...............................................................................................10

2.2CentralEurope..................................................................................................12

2.3NorthernAmerica.............................................................................................12

2.4Asia...................................................................................................................13

3Knowledgeonclimaticconditions.............................................................................14

3.1Measurements...................................................................................................15

3.1.1Windconditions..........................................................................................15

3.1.2Icingconditions..........................................................................................18

3.1.3Othermeteorologicalparameters................................................................19

3.2Modelling.........................................................................................................20

3.2.1PhysicalModels..........................................................................................20

3.2.2Empirical/StatisticalModels.......................................................................21

3.2.3IcingTypesandDescriptionofCalculationMethods................................21

3.2.4Icingrate.....................................................................................................22

3.2.5TURBICEandLEWICE............................................................................22

3.3Maps.................................................................................................................24

4Technicalsolutionsinu..........................................................................................28

4.1Technicalsolutionsforicing............................................................................28

4.1.1Sensors/Instruments....................................................................................28

4.1.2Blades.........................................................................................................28

4.1.3Othercomponents.......................................................................................30

4.2Technicalsolutionsforcoldclimates...............................................................30

4.2.1Materialsandlubricants..............................................................................30

4.2.2Heatingofcomponents...............................................................................31

4.3Operationalsolutionsforcoldclimates............................................................31

4.4O&Mconstraints..............................................................................................32

5Operationalexperience..............................................................................................33

5.1Operationalexperienceinicingconditions......................................................33

5.2Operationalexperienceinlowtemperatures....................................................34

5.3Finland..............................................................................................................34

5.4Sweden.............................................................................................................36

6

5.5Norway.............................................................................................................43

5.6Switzerland.......................................................................................................44

5.7USA..................................................................................................................44

5.8Canada..............................................................................................................45

6Existingstandards,requirementsandrecommendations...........................................47

6.1Windturbines...................................................................................................47

6.2Resourceestimationandpowerperformancemeasurements...........................47

7Summary....................................................................................................................48

7

1INTRODUCTION

In2001,theInternationalEnergyAgency(IEA)R&DWindProgrammestartedanew

annex,numberXIX,ternational

collaborationbetweentheparticipatingcountrieshasasobjectivestogatheroperational

experienceofwindturbinesandmeasurementcampaignsinicingorcoldclimatesto

elisto

formulatesitecategoriesbadonclimatologicalconditionsandsiteinfrastructureand

thenlinkthewindturbinetechnologiesandoperationalstrategiestothecategories.

Thiswillgiveguidelinestooperatorsandmanufacturersoperatingwindturbinesincold

climates.

Informationisgatheredanddisminatedontheprojectwebsite/.

TheoperatingagentoftheannexisTechnicalRearchCentreofFinlandVTTand

participatinginstitutesareFOI/FFAfromSweden,KjellerVindteknikkfromNorway,

RisøNationalLaboratoriesfromDenmark,theNationalRenewableEnergyLaboratory

(NREL)fromtheUSA,ENCOfromSwitzerlandandNaturalResourcesCanada[1].

Atthemomenttherearearelativelysmallnumberofwindpowerprojectsinthecold

climate,howeverthisglobalmarketgmentistimatedtobesubstantial,althoughno

oesemtobelackof

informationregardingtheoperationalexperienceandexactclimaticconditionsrelevant

tositesincoldclimates,especiallywhenriskoficingisconcerned.

Inaddition,whensitingwindturbinesincoldclimates,theasssmentoftheclimatic

conditions,theirimpactonturbineproductionandeconomy(reliability,O&Mcosts)

havetobemade(Fig.1).Generallyinformationabouttheaverageandminimum

temperaturesonperspectivesitesisusuallyavailablehowevericingfrequencyismore

tIEAandotherinternationalstandardssimplystatethat

standardmethodologiesdonotholdforsitesoutsideofnormaloperatingconditionsand

thusprojectsintheareasareoftencarriedoutwithinadequateknowledgeonicing

andotherextremeweatherconditions.

8

Estimatedlossdueto

safetyriskoficing,Xdays

of0production

Estimatedlossdueto

operationwithicedblades,

XdayswithX%production

loss

Estimatedlossdueto

icedupturbine,Xdays

d

technicalavailability

Estimatedlossdueto

standardoperationallimit,

Xdaysof0production.

bility

Productionlossvscostof

bladeanti-icingsystem

Productionlossandpossible

increainO&Mcostsvscost

ofbladeanti-icingorde-icing

system

Productionlossandpossible

increainO&Mcostsvscost

oflowtemperaturemodifications

Estimatedenergyyield

assumingicefreeconditions

Estimatedinvestmentcosts

assumingstandardtechnology

SITEWITHSAFETYRESTRICTIONS

SITEWITHSLIGHTICING

SITEWITHHEAVYICING

LOWTEMPSITE

mentofsitesincoldclimates:therearesitesinverylowtemperatures

smostfrequentjustbelow0°tes

havetoconsiderallthecasabove.

9

2WINDTURBINESINCOLDCLIMATES

Therearealreadyveralsiteswitheitherexistingorprojectedwindparksincold

climates:NorthernandCentralEurope,NorthernAmericaandAsia(ChinaandRussia).

Alltogetherapproximately500MW(Fig.2).

Thefollowingdescribesoperationalexperiencereportedbythepermanentmembersof

theIEAAnnexonwindturbineoperationincoldclimates.

Sitewithcoldclimate

turbine(s)

Projectedcoldclimate

turbine(s)

onsofoperatingwindturbinesincoldclimatesites[2,3,4,5,6,7].

2.1NORTHERNEUROPE

InScandinaviathereareexistingsitesinFinlandandthemountainsofSwedenand

dinaviaicingisoccasionalonthecoastline,andvereicingconditions

aturesmayfallbelow–20°Cofteninhigher

hecombinationof

verygoodwindconditions,vereicingandlowtemperaturesonarcticfelltops,arctic

modificationsforturbineshavebeendeveloped[12].Thoughdesignedforarctic

conditionsbladeheatingsystemshavebeeninstalledalsotomildericingconditions[9].

IntheLaplandregionofFinlandicingisvere,withrimeicingconditionsupto200

hour

ismostharshbetweenNovemberandFebruaryandoccursmostofteninatemperature

rangeof0°C--7°C.[8]ThereistwofindfarmsinFinnishLaplandOlostunturi-fjelland

Lammasoaivi,nturi-fjellwindfarmconsistsoffive

BonusMkIV600kWwindturbinesandLammasoaivicontainsoneBonusMkIV600

10

binesareequippedwithanice

ctureroftheJE-Bladeheatingsystems

romLapland,windturbinesinFinland

locateincoastalareasandinthesouthernarchipelagowheretheyalsoexperience

imateversionsofthestandardturbinesareud

inth

exceptionisthewindfarmatPori,locatedalongthewesterncoastofFinland(N61.3°,

E21.2°)wherethefourturbinesthatarelocatednexttoabusyroadwereequippedwith

irePoriwindfarmconsistsof9

d

turbinesupplierwasBonusEnergyA/SandtheturbinesutheJE-Systemblade

ccursoccasionallyatPori,mostoftenattemperaturesjust

belowfreezing(0°C--3°C).[9]

AccordingtotheFinnishwindturbinestatisticsnearlyeverysiteinFinlandreports

downtimeduetoicingorlowtemperaturesduringthewinter.[21]

Norwayhasalongshorelinefacingthewarmwatersoftheeasternpartofthenorth

ssuresystemsforminginthepolarjetstreamareasover

thewarmAtlanticwatersmoveeastwardandensurehighwindspeedsandamild

podIslandsandridgesalongthecoastare

edtootherareasintheworldatthesamelatitude,

hCape(71º),-4ºCisthe

lowestmonthlyaveragetemperatureatalevel.

Inclo

experiencewithwindmonitoringprograms,itemsasthefrequencyoficingismore

dturbineslocatedatHavøygavlen(Latitude

71ºand275mabovealevel)donothaveheatedblades.

uentlythenorthern

partofSwedengenerallytakesshelterfromvereicingconditionsbehindthe

fwindturbinesstillexperienced,particularlyinthearea

ofJäonallySweden

experiencecoldeasterlywindsflowingoffoftheunfrozenBalticSeacreatingvere

eventinDecemberof2002caudturbineswithout

bladeheatingsystemstobeputoutofoperationforveralweeks.

Thecombinationoflowtemperatures,below-50inthenorthernSweden,andlowsolar

radiationduringthewintermonthslimittheabilitytode-icestructuresthatareiced

uresthathavebeenicedupmaystayicedforlong

periods.

Occurrencesofsuper-cooledrainarecommoninSwedenandaverecawas

recordedattemperaturesbelow-10asfarsouthasStockholmonFeb6,nt

incloudicingwillalsoimpactalloperatorsoflargeoffshorewindturbinesintheBaltic.

Inaddition,theannualreturnoftheaicealongthecoastline,lakesandnorthernas

willcreateasignificantchallengeforthedesignerofoffshorewindturbinefoundations.

11

Aspartofavoluntarynationalprogramtheincidentsofproblemsrelatedtooperating

windturbinesincoldclimatehavebeenreportedandenteredintoadataba.92impact

reportswerereceivedfor2000,2001,and2002reportingapproximately8000hoursof

misdproduction.

2.2CENTRALEUROPE

.,ref.19,ananalysshowedthaticingconditions

existsinmuchlargerregionsofEuropethanearlierhasbeenexpectedbythewind

lysisalsofoundthatincentralEuropeicingandlow

temperaturesdeterioratewindenergyproductionatelevatedsites,whichunfortunately

herelowoperatingtemperaturesandicing

hasbeenrecordedcentreontheAlps,Apenninesandothermountainousareas.[19]

HeavyicingandahighnumberoficingdaysisobrvedatsitesliketheApenninesin

ampletheAcquaSruzzatestsiteinItaly,atsamelatitudesasNapoli,

untoficingexperiencedatthatsite

wouldleadtoasignificantproductionloswithstandardwindturbines.[19]

Icingandlowtemperaturesarealsoexperiencedfrequentlyinmountainousregionsof

SouthernFrance,intheAlps,andintheAlbsinsouthernGermany.

Icinghasalsobeenrecordedtodeterioratewindenergyproductionofveralwind

farmsatthehighaltitudesinScotlandmountains.

InSwitzerlandveralprojectshavebeencarriedoutinicingandinlowtemperature

rbinesthatexperienceicingandlowtemperaturesarelocatedathigh

altitudes,llysitesbelow

2000metreabovealevelexperienceonlylighticingandsiteswithhigheraltitudeare

antexperienceontheuofwind

energyunderclimaticallyextremeconditionswillbegainedfromthe800kWplanton

theGuetschnearAndermatt(2300mabovea-level)whichwascommissionedin

thefirstwindturbineinSwitzerlandthatusadaptedtechnology

(2200mabovea-level)aswellasCrêtMeuron(1300mabovea-level),will

increatheknowledgeaboutwindenergyproductioninalpineregionandharsh

climaticconditions.

InEuropethereisagrowingandidentifiedinteresttoerectmorewindturbinesatsites

inwhichthewindturbineswillbepronetoicing.[19]

2.3NORTHERNAMERICA

Windfarmsarebeinginstalledinthreegeneralclimaticregimeffectedbycold

orthcentralregion,suchasthe200MWwindplantsintheLake

Benton,Minnesotaarea,snowfallandcoldtemperaturesarecommonbutturbineicing

12

onallyalongtheeasterncoastoftheUS

andCanada,andspecificallythenortheastsuchasthe6MWplantinSearsburg,

Vermont,turbinesarelocatedonlowaltitudemountainridgesorincoastalregimes

ingonthelocationandaltitude,cloudorrimeicingis

tclarificationofsitesarealongthearcticcoast,suchasthe0.8MW

ites

experienceicing,coldtemperaturesandhighdensityairflows.

InCanada,onecansaywithconfidencethatallturbinesinstalled,exceptmaybethe

onesimmediatelylocatedontheEastandWestcoaststhatbenefitfromtheoceaneffect,

willbeexpodtotemperaturesbelow–20°Catonetimeoranotherduringtheyear.

Atmosphericin-cloudicingincounteredontheelevationsofBritishColumbia,

Yukonand,toalesrextent,ng

precipitationontheotherhandismorelikelytooccurinCentralandEasternCanada.

2.4ASIA

InChinaturbinesaregenerallylocatedonsiteswherewinterhumidityistypicallylow,

buttemperaturemaydropbelow–20°C,anddiurnaltemperaturechangesmaybeas

highas40°C.

13

3KNOWLEDGEONCLIMATICCONDITIONS

Generallyinformationabouttheaverageandminimumtemperaturesatasiteisusually

availablehowevericingfrequencyismoredifficulttoobtain,andprojectsareoften

carriedoutwithinadequateknowledgeonicingconditions.

Toassstheconquencesoficingandtherequiredmodificationstostandardwind

turbines,informationonthefrequencyoficingeventsandthedurationoficeon

differentpartsofthewindturbines,suchastheblades,anemometers,nacelle,andtower

analsoeffectwindresourceestimationduetotheoccasionalicingof

anemometersduringameasuringcampaign,whichcanbedifficulttodetect.

AccordingtostatisticsonFinnishcoast,icingcanbe5timesasfrequentat100meters

measurementsoficingareveryrareand

,developmentofmodelstobeudin

especially

trueformountainousareaswherethelocalterraineffectscanbedifficulttoasssin

ementsoftheconditionsfurtherthan1kmawaymayalsonotgive

enoughinformationaboutaspecificsiteinquestion.

Icingoccursattemperaturesbelow0°e,

amountanddensityoficedependonbothmeteorologicalconditionsandonthe

dimensionsandtypeofstructure(moving/static).Therearealsodifferenticingclimates,

suchascloudicing,whensmallwaterdropletsinthecloudimpactandfreezeonthe

surfaceofstructures,demonstrated

inFig.3,whichshowxamplesoftwodifferentsitesinFinland:Poriisacoastalsitein

SouthernFinlandandOlostunturiisasitewithheavyicinginthearcticfellsof

NorthernFinland[8,9].

14

0%

10%

20%

30%

40%

50%

60%

-24-22-20-18-16-14-12-10-8-6-4-20

Temperatureat38mlevel[Cdeg]

F

r

e

q

u

e

n

c

y

[

%]

Olostunturi:Icingeventsathubheightlevel(total1871h)

Pori:Icingeventsat84mlevel(total223h)

ferentsitesinFinlandwithannual

meantemperaturesof0.3°C(Olos)and7.1°C(Pori).

3.1MEASUREMENTS

Propermeasurementofclimaticandicingconditionsinextremeclimatesisnotas

anynsorcanbeconnectedtoadata

acquisitionsystemonehastofindasuitabletofcables,

extremecassimplyusingweatherandUV-resistantequipmentisnotsufficient,

tly,modernnsors,

suchasultra-sonicanemometersanddataacquisitionnetworkscanbeconnectedvia

r,fibrecablesforcoldclimateoperationneedtobeadopted

forsuchuby,forexample,usingnon-freezinggelthatispumpedintoconduits

surroundingtheinteriorcablestopreventwateringressandsubquenticeformation.

Onesuchexampleisdescribedinref[17].Thegelwillalsoprotectacableagainst

breakingifexpodtounforeenexternalloadsbyamaintenancecrews,orreindeers;

ttachmentswill

occasionallybreakandstandardweatherresistantcabletiesarenotsufficientincold

llyweatherresistantNylon12cabletiesshouldbeudincoldclimate

and/orhighmoistureconditions..Asimilarreasoningcanbeappliedtoconnectors.

Instrumentsforcoldclimatemeasurements,includinghumidity,temperature,wind

speed,winddirection,precipitationandradiation,havetobeproperlyheatedunder

icingconditionstomaintaintheiraccuracy(Fig.3).Instruments,moreorlesssuitable

forcoldclimatemeasurements,arecontinuouslybeingdevelopedandevaluatedby

manufacturersandurs[11].Dependingontherequiredaccuracyandinstandard

conditions,theexactlocationofaninstrumentmightberequiredtoadheretoIEA

15

recommendedpracticesorstandards,whichensurepropermountingincludingsufficient

ommendedpracticesarenotavailableforicing

conditions,oneistypicallyrecommendedtostayawayfromsuchevents,likeice

storms,whichisonereasonforthecreationofthisIEAAnnex.

eanemometerinvereicingconditions.(Photo,VTT)

3.1.1Windconditions

Inthefieldofwindenergyaproperlyacquiredwindspeedisofoutmostimportance.

Therearethreebasiccategoriesofneededaccuracywhenmeasuringwindspeedinflat

urceestimationaccuracyrequirementsarearound±5%,theaccuracy

levelofEuropeanWindAtlas[42].StandardIEC61400-1givesanaccuracy

requirementforanemometerudinpowerperformancetestingof±2%[34].Forwind

turbinecontrolaloweranemometeraccuracyisneeded,commonly±3%howeveran

evenloweraccuracylevelmaybeacceptable.[51-53].Manydifferenttypesof

anemometersareavailablewithwidedegreesofaccuracy,downtobelow2%for

icrequirementsaredefinedfordifferentconditionsby

veralnationalandinternationalstandard.

Inaddition,accuratewindspeedmeasurementsincomplexterrainhaveotherdifficult

anddonotneedtheuncertaintyintroducedbyimproperlyheatedormountedwind

velocitynsors.

Upuntilthispointnotmuchattentionhasbeenpaidtoicingofthewindgaugesinthe

winde

issurprisinggiventheimportanceofwindspeedmeasurementinsitingandsystem

averepeatedlyshownthatasmallamountoficereducesmeasuredwind

spee

example,asmallamountofrimeiceonthecupsandshaftofananemometermaylead

tounderestimationinwindspeedofabout30%atwindspeedof10m/elof

underestimationdependsonverityoficingconditions.[Refs.23-26].Thisdecreais

moreinsidiousas,withoutothermonitoringequipment,thereisnowaytodetermineifa

16

yleadtoan

underestimationofthewindspeedorthefailureofaturbinetoshutdowninahigh

windevent.

Althoughmanydifferenticemitigationmethodshavebeeninvestigated,themost

commonsolutionforaccuratemeasurementinicingclimateistheuofaheated

mentssuitableforcoldandicingclimateare

availableandnewdevicesareactivelybeingdevelopedandevaluatedbymanufacturers

andurs[11].

Accesstoanelectricitygridisgenerallyrequiredinordertoheatthensors,which

needtobeproperlyheatedtomaintaintheiraccuracy;solarpanelswillnotsufficeas

oelectricitygridisavailable,

halternativeistheuof

ssFederalInstituteforSnowandAvalanche

RearchhasbeenemployingpropellertypeanemometersintheJuramountainswith

reicingconditionsandtemperaturesbelow0°Cwithhigh

humiditytheirpropellertypeanemometershaveprovidedreasonabledatamorethan

98%ofthetime.

theregionswindmeasurementsareoftenperformedatlowerlevels,30minsteadof

ult,itmaybedifficulttoextrapolatemeasurementresultsto

tricitygridaccessisavailableSODARs

encewithSODARunitsin

Switzerlandhavedemonstratedthatthetechnologymaybeudinharshclimates,but

carefuloversightoftheequipmentisnecessaryanditshoulditislikelynotapplicable

forlongtermmeasurementprograms.

Ifcuporpropellertypeanemometersareud,theanemometer'scupshaftandpost

shouldbeheatedinordertopreventicefromaccumulatingandimpactingmeasurement

quality.

Attentionmustbepaidalsotothepositioningoftheanemometerandwindvaneinicing

reicingconditionstheaccuracygainedthroughheatingisquickly

lostifneighbouringobjectssuchasboomsandmastsareallowedtocollectice.

edforwindturbine

control,considerationsmayalsobeneededtothechangeinwakeandturbulenceontop

ofnacelleunderdifferenticingconditionvenifaheatedanemometerisud.

References19and31providereviewsofanemometerssuitablefortheuinicing

ofthe“WindEnergyProductioninColdClimate”(WECO)project,

fundedinpartbytheEuropeanUnion,veralrearchinstitutionsarecurrently

conductingoperationaltestsonanumberofanemometersandwindmeasurement

OREASIV

conference[22]theannualmarketforice-freensorsonlyinEuropeistimatedtobe

some11millionEuro.

17

3.1.2Icingconditions

ionalice-detectorsudtobeextremely

unreliablehoweverthistechnologyhasimprovedconsiderably,ashasknowledgeabout

tiontoimprovedtechnologyrearch,

commercialorganisationsarealsoconductingandsharingextensiverearchonice

accretionduetotemperature,humidity,radiation,winddirection,windspeedand

precipitation.

Currentlythereareveraltypesoficedetectorsonthemarketandaremainly

manufacturedforaviationandmeteorologicalpurpos.[19]Principlesofoperationof

icedetectors,somespecificicingnsorsandotheradhockapproachesthathavebeen

udinrearchprojectsareprentedinthistext.

FinnishcompanyLabkoOyhastwoversionsoficedetectorstooffer,LID(LabkoIce

Detector)kingprincipleofthebothmodelsisthata

longitudinalwirewavesistransmittedintoannonmagneticwirewithpiezoelectric

ceattenuatesthesignalmorethanwaterorothernonsolid

tenuationcanbemonitoredandudinicingindication[27].

TheInstrumarLimitedicensorIM101functionthroughmeasurementofthesurface

tais

lticingwindowis

programmedintoeachdeviceandwhentheparametersfallwithinthiswindow,an

“icing”101hasaninternalsolidstateswitchthatisclod

osuremaybe

udtoturnon/offlowpowerdevicesdirectly,eud

forcontrollingdevicessuchasalarms,eventrecorders,orheaters.[28]

sorworkson

tionprobevibratesultrasonicallyataresonant

softheicecollectingontheprobecaustheresonant

encydecreaequivalentto0.508mm(0.020inch)ofice

ametime,thedetector

ricingevent

detectedwithinthat60condsretsthetimertozeroandtheicesignalremains

activatedforanadditional60conds[39].

Oneinterestingpossibilityistheuofdewpointdetector,whichisdesignedtooperate

inbelowzerotemperatures,todecidealimitforthe

relativehumiditye.g.97%andassumethatwhentemperatureislowerthanzeroicing

suchadewpointmeasurementasanicedetectorwasstudiedatthePori

siteinFinland[9]andisdiscusdingreaterdetailinthepaperbyMakkonenetal.

[29].Oneofthemainissuesidentifiedinthereportismoregeneral,thereisnoabsolute

referenceforcalibratingicedetectorsbecaueventhemostup-to-dateicedetectorsare

not100%accurate.

18

Forwindturbineapplicationsitispossibletoidentifyicingusingoneheatedandone

standardanemometerslookingatthedifferenceinwindspeed,howeverafewquestions

hslowershouldtheunheatedanemometerreadcomparedtotheheated

onetobeinterpretedasanicingsignal?Isitpossibletoavoidfalalarmscaudby

wakesontopofwindturbinenacellebycarefulplacingoftheanemometer?Whena

standardunheatedanemometerfreezesitmaytakealongtimebeforetheanemometeris

ice-freeagain,ref.[30],whichcanmakedeterminingtheactualicingtimedifficult.

Thismethodhoweverenablestheestimationofenergylossthatawindturbinewould

experienceatthatspecificsiteinicingconditions,especiallyifunheatedanemometers

aretobeudforturbinecontrol.

InanexperimentonmeasuringicinginNorthernCanadatwoheatedandoneunheated

anemometerswereudtomeasuretheactualwindspeed,icingtimeandsublimation

.[33]Oneoftheheatedanemometerswaskeptice-freeandtheother

washeatedaftereveryoccasionwhenthewindspeedofthatanemometershoweda

15%eatedanemometer

sshowedthatitispossibletoestimatetheicingtime

lsodemonstratedthat

onecouldestimateofactualicingtimewiththismethod.

Productionpowerofthewindturbinecomparedtothepresumedproductionpower

accordingtothenacelleanemometermayalsoprovideaguideoficingsinceaturbine

ritisstill

unclearwhatconclusionscanbemadeduetoongoingissuessuchashowsmallshould

thepowerdegradationsbeandhowquicklyafterthebeginningoftheicingcansucha

andiftheturbineislocatedinaremotesiteandnovisual

obrvationsarepossible,reducedpowerproductionisofteninterpretedasan

“anemometererror”whenthecauofsucherrorisicing.

Automaticvisibilitynsorsmayalsobeudasanicedetectorhowevertheentire

instrumentespeciallythelens,mustbeheatedinlowtemperaturesandicingclimate

toensureappropriateoperationofthedevice.

3.1.3Measurementofothermeteorologicalparameters

Itisknownthattemperaturemeasurementsareimpactedbytheirsurroundings,

formanceofthermometersinicing

softhostudieshaveshownthat

errorsofveraldegreesarepossiblewhenthermometersnotdesignedforicing

ayeronathermometeroronaradiation

shieldinsulatestheprobefromthesurroundingairandcausdelaysanddampening

tcatheclodmeasurement

conditionsoftheairinsidetheradiationshieldmaycontinueuntiltheicehasmelted.

[31]

19

Inicingclimatestheradiationshieldsforthermocouplesshouldbeheatedorthe

meteritlfshouldbedesigned

foricingandlowtemperatureoperation.[31]

Measuringhumidityreliablyinicingandlowtemperaturesclimateisalsoanontrivial

tynsorsanddewpointdetectorsshouldbeplacedwiththesame

ribedabovetheimproperuofthe

radiationshieldcouldimpactthetemperaturemeasurement,inwhichthecalculationof

rdhygrometersdesignedfortemperaturesover0°Cwillgive

unreliableresultsatlowtemperatures.[31]

Instrumentsforcoldclimatemeasurements,includinghumidity,temperature,wind

speed,winddirection,precipitationandradiation,havetobeproperlydesignedand

heatedundericingconditionstomaintaintheiraccuracy(Fig.4).Instrumentsthatare

suitableforcoldclimatemeasurementsarecontinuouslybeingdevelopedandevaluated

bymanufacturersandurs[11].

3.2MODELLING

Modelsforpredictinglocalweathereventsincludingwindandicingestimatesarebeing

orfactorlimitingtheprogressof

modellingisthecalculationcapacityofcomputers,whichistoolowtoenableaccurate

cialcomputerprogramsandmodels

forcalculatingshapesand

massoficebuildupandbladeheatingdemandincertainicingconditionshavealso

windturbineicingrearchtookholdthe

aerospaceindustryhaddevelopedcomputerprogramsthatmodelleadingedgeicingof

ate1970’spowercompaniesalsodevelopedmodelstocalculate

els,TURBICEand

LEWICE,thatareudincalculatingicemassandbladeheatingdemandsindifferent

tion,thebasisofmethodsthatare

udtocalculatedifferenttypesoficingfromstandardmeteorologicalobrvationsis

prented.

Ofthevariousmodelsthathavebeendeveloped,twobasiccategories,physicaland

empirical,havebeendistinguishedbadonthedifferentstandpoints,backgroundsand

thedifferentphysicalpropertiesofdifferenticingphenomenon.

3.2.1PhysicalModels

Physicalicingandmeteorologicalmodelsarequitedetailedandrequirespecific

definitionofmeteorologicalparametersincludingthewatercontentoftheair,droplet

size,windspeeds,dellingtheiceaccretiononwindturbine

bladeorpowerline,onehastoalsoknowaccuratelytheshapeandsizeoftheobject

edmodelsarecomputationallydemandingandhave

thereforebeenimprovedtogetherwiththetechnologicalimprovementsofcomputers.

20

Asaparatecategory,fullphysicalmeteorologicalmodelscanalsobeudtopredict

tancemeso-scalemodels(MM5,MC2andothers)havethephysical

odels,generallyudinregional

weatherprediction,canbeudtopredictupcomingicingeventsortoprovideageneral

predictionofthelikelihoodofsucheventsforspecificprojectsunderconsideration.

3.2.2Empirical/StatisticalModels

atecaudbyin-

cloudicingatacertainsitemaybequantifiedfirstbydatafromthenearest

oudheight,cloudcoverandtemperaturedatatogether

withsiteelevationitispossibletoestimatethefrequencyoficingthatsiteislikelyto

experience.

Knowledgeoficingeventshasincreadandmoremeteorologicalandtopographical

terssuchastemperature

(air,object,wet-bulbanddewpoint),winddirection,windspeed,cloudheight,cloud

cover,thehumidityprofile,precipitation,regionaltopography,localtopography,object

size,shapeandmaterialcompositeandsolarradiationhavebeenaddedtomore

comehasbeenthatthemodelscannowalsoprovide

informationabouttheamountandrateoficinginsteadofjustthefrequencyoficing

events.

3.2.3IcingTypesandDescriptionofCalculationMethods

In-cloudicingisconsideredtooccuriftheheightofcloudbaislessthanthesite

elevationandthetemperatureatthesiteisbelowzero.

Empiricalandstatisticalmodelshavebeenmodifiedbecauaccuratecloudba

ingresultscanbeimproved,byusing

statisticalrelationbetweenweathersituationsandcloudposition(cloudbaheightand

horizontallocation/extent).Byusingstatisticalvaluesofdropletsize,windspeed,

direction,andobjectsizeandshape,sof

aaccumulatediceaccretionmayalsobeestimatedwiththismethod.

Calculationsusingfullphysicalmodelswithmeteorologicalandtopographical

parameters,particlesize,concentration,momentum,heatbalanceandobjectshape

changemayprovidemoreaccurateresultsdependingontheaccuracyoftheinitial

alephysicalmodelsrequirelargecalculationcapacities.

Freezingprecipitationoccurswhenitisrainingandwet-bulbtemperatureliesbelow

zero.

Empiricalandstatisticalmodelscalculateicingfrequencyandamountfrom

precipitationintensity,duration,windspeed,meanairtemperature,objectsize,shape

andanempiricalcorrectionfactor.

21

Aswithin-couldicingcalculationsusingfullphysicalmodelsitispossibletomodel

thodalsohas

thesamedrawbacks,theyarecomputationallyintensive.

Frostoccurswhenthesurfacetemperatureofanobjectdropsbelowthefrostordew

untandtypeoffrostaregiven

asanequationoftemperatureratios,empiriccorrectionfactorandhumidity.

Wetsnowandsleetisformedfromdrysnowwhenatlowerelevationsthereisastrong

enoughpositiveheatfluxfromtheenvironmenttomeltthesurfaceofdrysnowflakes.

[43]

3.2.4Icingrate

Therateoficingisdependentonthefluxofparticles(concentrationtimesvelocity)in

hedifferent

sizeandthereforedifferentinertiaofparticles,someofthemwillcollidewithanobject

whileothersmallerones,whichhavelessinertia,followtheairstreamandpassthe

rticlesalsobouncewhencollidingwithanobjectandthuswillnot

pendingontheheatfluxformthesurfacetothe

surroundings,collidingparticlesfreezeattheirimpactspot,rimeice,orformathin

waterfilmonthesurfaceofanobject,enticingprocessalsoleadsto

ral,duetoitscomplexityandthemany

processparametersaphysicalicingmodelthatwouldapplytoallicingprocessstill

aldescriptions,includingheattransfer,ofdifferenticing

processareprentedindetailinref.[44-49].

3.2.5TURBICEandLEWICE

Smallamountsoficeonwindturbinebladesdeterioratetheiraerodynamicperformance

rmore,largeice

cesarealso

hazardouswhentheyshedofftheturbinebladeswithhighvelocity.

Asintroducedpreviously,twosoftwaremodelshavebeendevelopedthatcanbeudto

analyiceformation,TURBICEandLEWICE.

TURBICE

Thisnumericalmodelsimulatesiceaccretion,amountandiceshapesonwindturbine

eenunderdevelopmentattheTechnicalRearchCentreofFinland

(VTT)since1991.

Themodelaccretesiceonatwo-dimensionalairfoilctioninapotentialflowfield

ericalsolutionforthepotentialflow

followsthecommonlyud“panel”ttrajectoriesareintegratedfromthe

steady-stateequationofmotion,usingdropletdragcoefficientsofLangmuirand

22

Blodgett(1946)andBeardandPruppacher(1969).Theintegrationbeginstenchord

lengthsupstreamoftheairfoilction,andiscarriedoutusingafifth-orderRunge-

actpointisdeterminedby

linearinterpolationbetweenthe600coordinatepoints,whichdefinetheairfoilction.

lesofattackexperiencedbya

elcanalsosimulateicingwhenthe

bladeisheated.

TURBICEsimulationshavebeencomparedandverifiedwithdatafromicingwind

tunnelexperimentsforaircraftwindctionsandfromafieldstudyofnaturalwind

tionshaveshowngoodagreementwithactualdata.[54]

InthedevelopmentofthebladeheatingtechnologyTURBICEsimulationshavebeen

utilidinthedeterminationoftheimpingementareaofwaterdropletsonbladesurface

andindeterminationofbladeheatingpowerneededindifferenticingconditions.

Resultshaveenabledtheoptimisationofthenecessaryheatingpowerandhasbeen

utilidinthepositioningofablade-heatingelement.

LEWICE

AnothersoftwarethatcanbeudforiceaccretionandheatingdemandisLewice2.0.

Lewice[40]wasdevelopedbytheicingbranchattheNASAGlennRearchCenterin

Cleveland,iceaccretionpredictioncodethatappliesatimestepping

doesnotpredictthe

degradationinaerodynamicperformancesduetoicingratheritevaluatesthe

thermodynamicsofthefreezingprocessthatoccurswhensupercooleddropletsimpinge

maryuisforevaluatingicingonaircraftbuthasbeenadaptedto

workonotherapplications.

Theparticletrajectoriesandimpingementpointsonthebodyarecalculatedfroma

potentialflowsolutionthatisproducedbytheDouglasHess-Smith2-Dpanelcode

atelyandifspecified,theflowsolutioncanbeobtainedfrom

agridgeneratorandgrid-badflowsolverorreadinasasolutionfilefromthisflow

hstandingthemethodud,theflowsolutiondeterminesthedistribution

ofliquidwaterimpingingonthebody,whichthenrvesasinputtotheicing

growthrateonthesurfacebodyiscalculatedfromthe

icingmodelthatwasfirstdevelopedbyMessinger[40,41].Thisisaniterativeprocess

ocedure

isrepeatedforaspecifictimeduration.

Lewicecanmodelbothdryandwet(glaze)tiontosimulatingtheice

accretion,sinconjunction

withtheiceaccretionroutineandcalculatesthepowerdensityrequiredtopreventthe

i-icingmodesarepossible:runningwetand

tsourcefortheanti-

icingcapabilitycanbespecifiedasbeingelectrothermalorhotair.

23

Inthecurrentapplication,

alsogeneratedataaboutdroplettrajectories,collectionefficiencies,impingementlimits,

energyandmassbalances,otentialflow

cannotmodelstallorpost-stallbehaviour,thecalculationsarevalidforunstalledrotor

regionsonly.

3.3MAPS

AnicingmapofEuropehasbeendevelopedinordertoestimatetheareasinwhich

ersionsoftheEuropeanIcingMap

andFrostmapwereproducedinWECOEUproject[19,20,50],prentedinFigure5.

Noicing

Occasionalicing-lessthan1dayperyear

Lighticing-2-7daysperyear

Moderateicing-8-14daysperyear

Strongicing-15-30daysperyear

Heavyicing-morethan30daysperyear

Weatherstation

apofEurope.[19]

24

AnupdatedversionsoftheEuropeanIcingMaparecurrentlyunderdevelopmentinthe

r,atoolforestimatingthe

numberoficingdaysandicingintensityatagivensiteisstillmissing.

Duetothelocaltopography,variationsinicingverityandintensitymayvarygreatly

withinshortdistancesandthereforeicingmaps,suchasinFigure5,cannotbe

interpretedaxactandmustbeudinconnectionwithlocaltopographical

informationand,ifpossible,withmeasurementstatistics.

AmoreexacticingmapfortheBritishIsleswheretheeffectofterrainhasbeentaken

intoaccountisprentedinFigure6..Theicingmapwasproducedbyfirstexamining

thenumberoficingdaysat0m,250mand500melevationsabovealevelatnine

hreelevelswere

nddetailedestimationofthenumber

oficingdayswastheninterpolatedandextrapolatedbyusingthepreviousthreelevels

ultisaclearpictureofareaswereicingcouldbe

faced.[50]Duetothelocalclimaticconditionsandlownumberofweatherstations

udintheproductionofthemap,theactualnumberoficingdayxperiencedatsome

sitemaydifferfromtheamountprentedinthemap.

numberofin-cloudicingdaysintheUKandIrelandatgroundlevel

andtheweatherstationsudincalculation[50].

allsuchmapstheverity

andintensityoficingmayvarygreatlywithshortdistanceandthemapinFigure7

shouldbeinterpretedasindicativeonly.

25

MapfortheaveragenumberdayswithfreezingprecipitationduringayearinCanadais

prentedinFigure8.

Similargeneralmapsofthisnaturearegenerallyavailablefromtheweatherrvice

agenciesofmostcountriesinnorthernandsoutherncountries.

apofSwitzerlandfor1000m.a.s.l

26

mberofdayswithfreezingrainduringoneyearinCanadabetween

mNationalArchives&DataManagementBranchofthe

MeteorologicalServiceofCanada.

27

4TECHNICALSOLUTIONSINUSE

Thereareawidearrayofsolutionsthathavebeenudtoreducetheimpactofcold

lowingctionof

thisdocumentreviewscurrentexperience.

4.1TECHNICALSOLUTIONSFORICING

4.1.1Sensors/Instruments

Avarietyofheatedwindnsors,asdiscusdingreaterdepthearlierinthispaper,are

available,testedandudatsiteswhereicingisfrequent[11].

Currentlysomemanufacturersalsouanemometerstoindicateweatherturbinesare

heanticipatedproductionpower,calculatedfromthewind

speed,ifferenceislargeenoughanalarm

cas

turbineareshutdown.

4.1.2Blades

Bladeheatingmaybenecessaryorprofitableatsitesthatexperiencefrequenticingor

ak-evencostofsucha

heatingsystemdependsonlostenergyproductionduetoicingandthepriceof

orewhenthefinancialbenefitsofabladeheatingsystemare

evaluated,icingtime,

heatingsystemmayalsoberequiredasasafetyprecautioninconnectiontotheplanning

helimitationsofbladeheatingsystemsistheir

energyconsumption,eapproachtoestimatethebreak-

.[12].

Anumberofdifferentapproachesforthebladeheatinghavebeenprented,developed

andtestedbutcurrentpracticeindicatesthatinheavyicingconditionstheoutersurfaces

ofthebladesneedtobeheatedinordertoachievesatisfactoryresults.

Atpre

Finnishbladeheatingsystem,wherecarbonfibreelementsaremountedtotheblades

nearthesurface,hasthewidestoperatingexperience,from18turbinesatvarioussites,

withatotalofnearly100operatingwinters[12].

Onelowpowerconsumptionmethodforheavyicingenvironmentsistheuof

pneumaticdeicingsystemthatworkswiththerapidexpansionofinflatablemembranes

arsystemhasbeeninuonsomesmallandregionalaircrafts

encefromwindturbineshoweverislacking.

28

Insiteswhereicingisslight,infrequentandtheicingperiodsarefollowedby

temperaturerisingabove0°Corareasofhighwintersolarintensity,bladescoatedwith

ngtheturbineandcirculatingheatedairinsidethe

dthatusblowerandheater

tocirculatehotairinsidetheturbinebladeisunderthedevelopmentinSwitzerland.

weveris

ngthewindturbinewhenicing

r,thismethod

doesrequireicedetectors.

Therehavebeenanumberofotherpropodsolutions,likeblade-heatingsystemsbad

onmicrowavetechnologybuttodatetheyhavenotbeensuccessfullyimplemented.

TurbineSafety

Turbines,withorwithoutbladeheatingsystems,poariskintheformofthrownice.

Irrespectiveofwhethertheturbinesareequippedwithbladeheatingsystems,warning

houldbelocatedatleast150mfromturbineinall

nce19providesamethodtoestimatetheriskthatresultsfromice

pleofawarningsignisshownin

Figure9.

from.

29

4.1.3Othercomponents

Turbinesthataremodifiedforvereicingclimatemustalsocopewithsnowandthe

freezingofmoistureinthegearbox,tproperly

alingthenacelle,itmayfillwithdriftingsnowashasbeenexperiencedinLapland

xesandyawsystemsneedtobeheatedandkeptfreeofice,asdo

anydiskbreaksorparators.

4.2TECHNICALSOLUTIONSFORCOLDCLIMATES

Littlespecificinformationisavailableaboutmaterialpropertiesandlubricantsforcold

ailable

informationcomesintheformofreportscitingfieldexperiencesfromprojectsincold

rehoweversomecommonareasofconcernthatareexpresd

repeatedlyintheareaofturbinematerialsandlubricants.

Mostturbinemanufacturesofferproductsorupgradestoproductsforcold

ormationindicatesthattheuoftheupgradesisrequiredfor

successfulunitoperationintheclimates.

4.2.1Materialsandlubricants

Theuofcoldresistantsteelinallstructuralmemberswithweldsdoesnotincreathe

rdhot-dipgalvanizedboltshaveprovenadequateinlow

temperatures[15].

RecenttestingattheNationalWindTechnologyCentre,USA,haslookedatthecyclic

loadingofwindturbinebladerootstudsatambientandextremecoldtemperatures,-45°

to-51°C(-50°to-60°F).Testingconsidered4140steelrootstuds,aVinylEster/E-

glasslam

thelimitedtests“allofthecoldtemperaturesamplestestedexceededthelifeofthe

roomtemperaturecontrolgroup,thoughnoneofthecoldtemperaturesamplexhibited

anyevidenceofsuperiorconstructionovertheroomtemperaturesamples”[16].The

tests,oneofthefewbeingconductedspecificallytolookatissuesrelatedtowind

turbineconstruction,showthatoperationincoldtemperaturesdonotalwaysresultin

damage,butmayactuallyimprovetheperformanceofthesystem.

30

fatiguepulltestsonbladestudsconductedatNRELcomparingstuds

atstandard(20°C)andarctic(-48°C)temperatures.

Lubrication:Intheareaoflubricationandhydraulicoils,similarpracticalworkhas

as

manufacture

mostcasthelubricantshavebeentestedbuttheoperatorincouragedtoobtain

specificcertificationspriortotheiru.

4.2.2Heatingofcomponents

Attheprentmomentsurfaceheatedgearboxesandgearboxeswithimmerdheaters

withconstantoilcirculation,generatorheatersandalsoheatersforthecabinscontaining

controlelectronicsareudtoavoidcoldrelatedproblems.[15,32].Especially

importantistheprotectionofcontrolelectronicsagainstmoistureandcondensationat

siteswherelowtemperaturesduringthewinterisfrequent.

4.3OPERATIONALSOLUTIONSFORCOLDCLIMATES

Windturbinemanufacturersrecommendthat,eventheturbinesthatareequippedwith

coldweatherpackageshouldbestoppedattemperaturesbelow-30°sites

lesolutionmightbe

31

toallowtheturbinetooperateatpartialloadwherethestresswouldstaybelowthe

designlimit.

Anadditionalconcernwiththeoperationofwindturbinesincoldclimatesisthatlow

noverpowerwind

turbinegeneratorsandhasbeenknowntobrakegearboxesorturbinemainshafts.

Measuringtemperatureanddisablingthewindturbinesduringextremelow

temperatureshavebeenudtoreducethechanceofsuchfailures.

4.4O&MCONSTRAINTS

Turbinesmaylocateatremotesitesandtheaccesstothesitesmaybedifficultoreven

ssiblethattheaccesstositemaybelimitedto

ereforeoutmostimportantthatbasic

toolsthatenablelightrepairssuchaswrenches,hammers,tat

rkingconditionsduetohumidity,highwindspeed,snowingoricingmay

perationandmaintenanceshouldalso

includethemaintenanceofcoldclimatemodifications.

32

5OPERATIONALEXPERIENCE

5.1OPERATIONALEXPERIENCEINICINGCONDITIONS

thecaevenwith

slighticingastheaerodynamicpropertiesofthebladearensitivetominorchangesin

durationoficeonthebladescanbeconsiderablylongerthanthetimeoficing

mesofveralweekswithasingleicingincidenthavebeenreported

inSouthernGermany.

Ontheotherhand,glazeiceaccretionhasbeenshowntocauoverproductiondueto

delayedstallonpassivepitchcontrolledwindturbines[57].Inmostcasthiswillbe

ration

overratedpowercausadditionaldamagetothecomponentsandwillresultina

shorterlifeofthegenerators,bearingandgearboxes.

Thestructuralloadsofaturbinemayincreasignificantlyduetoicingoftheblades,

tion,iceusuallyshedsfrom

thebladesunevenlyresultinginfurtherloadingontheturbine[10]duetothemass

imbalance,orcesresultintwobasicload

types;extremeloadsandfatigueloads,dependingontheturbinesstructuraldesignand

rlydesignedcontrolsystemshouldaddressissuesofextreme

loads,irrespectiveoftheiroriginandsinceotherextremeloadcaus,suchasasingle

failingbladepitchmechanism,typicallyresultinhigherloads,theextremeloadcas

eloadingissimilarly

sicalinfluenceofthelatter

isrelativelyeasytoestimatebuttheknowledgeregardingthefrequencyofsuch

occurrencesisscarce,eloadingcaudby

aerodynamicforces,suchasthocaudbymererimeiceaccretion,arelikelytobe

underestimatedbytoday’sinternationalrecommendations.[58]

Icethrownofftheblademayalsopoasafetyriskeveninareaswhereicingis

infrequent,specificallywhentheturbinesaresituatedclotothepublic,suchasroad

andskiingresorts.

Icesheddingoffthetowerorthenacellecanalsopoasimilarthoughamorelimited

higherespeciallyforthervicepersonnel.

Caswhereicingoftheyawgearhasresultedinthedamageofyawingmotorhave

beenrecordedinFinland.

Icingalsoaffectswindnsors,bothinresourceestimationandcontrollingtheturbine.

Awindturbinewithanicedcontrolanemometermaynotstarteveninstrongwinds,

dloadsarecaudifapitchcontrolsystem

33

ndvanemayleadtooperationin

misalignedyaworaproductionstopduetothemisalignment.

5.2OPERATIONALEXPERIENCEINLOWTEMPERATURES

Lowtemperatureffectonmaterialsandinwindturbinesprimarilyonglassfibre

structures,plastics,ubricationoilsandgreashavebeen

re

temperatureandcondensationhavealsodamagedcontrolelectronics.

cationof

standardhydraulicsystemmayalsonotbelimitedtothespecificoil,modificationofthe

tubes,valvesandequipmentassociatedwiththehydraulicsystemmayalsoberequired.

Duetohighviscosityofstandardoilsinlowtemperaturesordifferentpropertiesofcold

temperatureoils,turbinestart-upmaybedelayedtohigherwindspeedswhichwill

impactoverallturbineperformance.

Whengoingtoverylowtemperatures,theneedforcoldweatherorweatherresistant

materialxtendsforboththesteelandplasticsudinthesystemfabricationbutalso

wire

forwhichtheinsulationbecomesbrittlemayfracture,leadingtoshorting,hascaud

ieceof

equipment,eventhemosttrivial,mustbeassdforflexibilityandusabilityat

extremetemperatures.

Alsorv

mayresultinincreadO&Mcostsorextendeddowntimeoftheturbine.

Anotherfactorthathasbeenidentifiedistheincreadsystemloadingduetothehigh

tuncommontohave(stallcontrolled)turbines

produceover20%lcasofgenerator

overheatinghavebeenreportedinCanadaandFinlandcaudbyoverproductiondueto

highairdensity[13].Thisleadstoproductionlossandprobablyhasleadtogenerator

failures[14].Impactsonthegearboxandbreakingsystemswilllikewineedtobe

r,duetothe

complexityofthesystems,specifictestsandtheimpactofcoldtemperaturesonthe

subsystemshavenotgenerallybeencarriedout.

5.3COUNTRYOPERATIONALEXPEREANCE

5.3.1Finland

AtOlostunturi,thesitedescribedindetailedinction2.1),innorthernFinlandstandard

powerperformancemeasurementswerecarriedoutforasingleturbineandthe

performancewasfoundtobeaccordingtothemanufacturerspowercurvewhenicing

theharshesticingperiodstheperformancedidnot

34

extremeicingcasthebladeheating

tiontothatinsomecastherunbackwateron

thebladeduringicingandbladeheatingwasfoundtofreezeafterithadpasdthe

thefallsandspringsthebladeheatingsystemwasable

terof2000to

2001thebladeheatingsystemud43674kWh,whichcorrespondsto3.6%ofthe

turbine’dheatingenergygrewduetotheproblems

withicedetectors,whichledtothebladeheatingequipmentreceivingmorepowerthen

helesswindenergyconversionatsiteslikeOlosturturiwithouta

bladeheatingsystemswouldbeimpossibleandunprofitableduetoturbinedowntime.

Similarpowerperformancemeasurements,asatOlostunturi,werecarriedoutatPori,

-cloudicingwasobrvedtobe

ronglysuggests

thaticingbecomesamoreimportantissuetocoastalwindparksatsiteslikePoriwhen

r,thewintersduringthe

measurementperiodweremilderthanaverageandicingwasonlyobrved

deheatingsystemwasudfor10minuteverynighttoavoidice

asontoinstallabladeheatingsystemtoawindturbineat

siteslikePoriisforsafetyofpublic.

InPori,lightingfrequencyishigherthaninNorthernFinlandandlightingstrikestothe

bladeheatingelementshavebeenregisteredalthoughdamagetotheiceprevention

systemcouldnotbedetected.

PowerconsumptionofthePoriicepreventionsystemwasmeasuredtobe1%ofthe

turbine’imumheatingpoweroftheturbinesis6%ofthe

nominalpoweroftheturbines.

ReporteddowntimesduetoicingandlowtemperaturesinFinlandbetween1996-2001

areprentedinTable1andTable2respectively.

eddowntimesduetoicinginFinland1996-2001.[21]

FINLAND

HoursTurbinesHoursTurbinesHoursTurbinesHoursTurbinesHoursTurbinesHoursTurbines

Lapland1192159851

Åland973443

BayofBothnia8584372598253275737414315

SeaofBothnia21

GulfofFinland

Total18423020

Totalnumberofturbines1

Shareofthetotaldowntimeofthe

turbinesthatreportedicingduringthe

year

45%21%9%12%9%26%

299719981999

¨

35

eddowntimeduetolowtemperatureinFinland1997-2001.[21]

FINLAND

HoursTurbinesHoursTurbinesHoursTurbinesHoursTurbinesHoursTurbines

Lapland45033211006

Åland11

BayofBothnia2818964

SeaofBothnia60217337

GulfofFinland

Total8852044253917

Totalnumberofturbines2129386161

Shareoftheoperationalhoursofthe

turbinesthatreportedlowtemperatures

duringtheyear

0%2%3%1%2%

920002001

IcingretardsmorewindenergyproductionthanlowtemperaturesinFinland(Table1

andTable2).Thereareveralreasonsforthisdifference,suchasthatlowtemperatures

operationwastakenintoaccountinthedesignprocessofmostturbinesoperatingin

tion,themajorityofFinland’swindturbinesarelocatedincoastalareas

oweverisrecorded

regularlythroughouttheentirecountry.

5.3.2Sweden

Aspartofanationalprogram,monthlyoperationalstatisticsfrommostwindturbine

ownershavingobtainedinvestmentsubsidieshavebeencollectedformorethan10

om622windturbines,totalling345MWdisperdnationally,were

availablebytheendof2002[55].Morerecently,theaveragecapacityofinstalledunits

alenergyproduction

alsoincreadfromthe2001amountof609GWAby23%.Figure11showsthe

locationandimpactofreportedcoldclimateincidentreportsduring2000-2002.

36

ationof565outofatleast622windturbinesisshown(red).Cold

climatereportshavebeensubmittedfrom47turbines(blue).Left:Distributionof92

:Distributionof8022

thecircle

reprentsthenumberofreportsandincidenthoursrespectively.

TheSwedishstatisticalincidentdatabacontainsatotalof1337recordsreportedto

haveoccurredinbetween1998-01-31to2002-12-31resultinginatotaldowntimeof

161,523hours.92incidents(7%)arerelatedtocoldclimateresultingin8022(5%)lost

otalforcoldclimates,thereportedlowtemperaturedowntime

totalled669hours(8%)whiletheequivalentforicingeventswas7353h(92%).The

numberofunrecordedicecascansafelybeassumedtobeoverwhelmingdueto

manualreportingincombinationwithaninherentlackoftechnologyandmethodsto

ticenergyreportinghasbeenappliedin

recentyears[56]andcovers,byendof2002,65%tclear

whetherautomaticreportingwillincreaordecreathewillingnessofwindturbine

operatorstosubmitcoldclimateincidentreportsoncemanualreportingisnolonger

mandatory.

37

Datacollectedthroughthemandatoryreportsshowthatnumberofincidentsofcold

climatereportingareincreasing,figure12.

Reportedcoldclimatehours&numberofincidents

0

500

1000

1500

2000

2500

3000

3500

J

a

n

-

0

0

A

p

r

-

0

0

J

u

l

-

0

0

O

c

t

-

0

0

J

a

n

-

0

1

A

p

r

-

0

1

J

u

l

-

0

1

O

c

t

-

0

1

J

a

n

-

0

2

A

p

r

-

0

2

J

u

l

-

0

2

O

c

t

-

0

2

Quarter

H

o

u

r

s

0

5

10

15

20

25

30

N

u

m

b

e

r

o

f

i

n

c

i

d

e

n

t

sHoursreported

Numberofincidents

Sweden2000-2002

8022hrs,92incidents

ynumberofreportedcoldclimatedowntimeandnumberof

incidents.

Anassumedreasonfortheincreasingnumberofreportedcold

climaterelatedincidentsisthefactthatwindturbineswerebuilt

recentlyinareaswithmoreextremecoldclimateconditions,

Figure13.

Thevariouscausofincidentsareshownforiceandlow

temperatureinFigure14andFigure15respectively.

coldclimate

region.

38

39

Incidentslabelledascaudby"Ice"

0

10

20

30

40

50

60

70

RotorbladeBlade(hull)WindindicatorTemperatureOther*

%

%oftime

%ofcounts

*=Pitch(hydraulic),Elctric(fu),Controlsyst.,Controlcomputer,

Controlsyst.(cable,contacts),Vibration,Wholeturbine,Unknown

Sweden2000-2002

7354hrs,72incidents

fincidentsduetoice.

Incidentslabelledas"lowtemperature"

0

5

10

15

20

25

30

35

40

45

50

P

i

t

c

h

P

i

t

c

h

(

h

y

d

r

a

u

l

i

c

)

P

i

t

c

h

(

b

e

a

r

i

n

g

)

C

o

n

t

r

o

l

c

o

m

p

u

t

e

r

G

e

a

r

b

o

x

H

y

d

r

a

u

l

i

c

Y

a

w

e

n

g

i

n

e

%

%oftime

%ofcounts

Sweden2000-2002

668hrs,20incidents

fincidentsduetolowtemperature.

Thereare16(3withunknownco-ordinates)windturbineslocatedwithinthemaincold

reshownintable3:

rbinetypeswithincoldclimateregion

#ManufacturerModelRatedpower[kW]Control

10NEGMiconNM900/52900/200stall,2speed

1NEGMiconNM750/48750/200stall,2speed

1NEGMicon750kW750/175stall,2speed

1*NEGMiconNM72C/15001500/400activestall,2speed

2BonusMkIV600/120stall,2speed

1(2*)VestasV52-850850pitch,variablespeed

1VestasV66-1051750pitch,variablespeed

0(1)Nordex600kW600/125stall,2speed

19(Total#ofwindturbines)

*notincludedinnationalstatistics

Thegeographicallisting,fromsouthtonorth,aswellas1stdateofoperationareshown

inTable4:

phicallisting,fromsouthtonorth,and1stdateofoperation,ofprent

windturbineswithinthemaincoldclimateregion.

#LocationDateManufacturerModel

1Äppelbo00-12-17NEGMiconNM900/52

0*(1)Rodovålen198-10-21Nordex600kW

1Rodovålen298-10-07BonusMkIV,600kW

1Rodovålen398-10-23NEGMicon750kW

1Rodovålen03-01-01VestasV52-850

1Bydalen02-09-04NEGMiconNM750/48

1Gråsjön,Kall00-11-08VestasV66-105

3Klimpfjäll01-01-16NEGMiconNM900/52

1Suorva98-10-13BonusMkIV

6Viscaria,Kiruna01-09-18NEGMiconNM900/52

1**Digerberget02-01-01NEGMiconNM72C/1500

1**Almåsa,Krokom?VestasV52-850

1**Vallrun,Krokom?VestasV52-850

19(Total#ofwindturbines)

*replacedbyaV52-850notyetincludedinnationaloperationalstatistics

**turbinesnotyetincludedinnationaloperationalstatistics

Prolongedreduced,aswellasabnt,windenergyproductionfromwindturbinesdueto

hexamplecanbedetectedinthe

operationalstatisticsfromthesouthernmostwindturbinelistedabove;Ä

16showsthemonthlyenergyproductionfromtheÄppelboturbinedividedbythe

averagedittofromtwoidenticalwindturbineslocatedinthesouthernpartofthe

40

rgyoutputinDecember2002canbeentohitarecordlowduetoice.

Theturbinedoesnothaveheatedblades.

RelativeperformanceofÄppelboNM900/52

-50

0

50

100

150

200

250

Jan-01Apr-01Jul-01Oct-01Jan-02Apr-02Jul-02Oct-02Jan-03

%

yenergyproductionfromtheÄppelboturbinedividedbythe

averagedittofromtwoidenticalwindturbineslocatedinthesouthernpartofthe

country.

AccesstooffshoresitesinsouthernSwedenhasbeenlimitedduringthepastwinterdue

toice,17showssuchan

dfarmisclobutyetsofaraway.

41

tooffshorewindfarmsinsouthernSwedenhasbeenlimiteddueto

ice.

Thepastwinter,2002/2003,icebuild-upsonwindturbinebladeshavebeenreported

pleofsuchanoccurrenceis

dwascalmandproductionlossweresmallinthis

kofbeinghitbyapieceoficebeingshedfromastartingwind

odynamicdragofaflyingpieceofice

needtobeincludedinriskzonecalculations,asthisrightfullywillreducethesizeofthe

affectedriskzone.

42

ld-uponawindturbinebladeduringwinterof2002/2003.

5.3.3Norway

Norwayhasalongshorelinefacingthewarmwatersoftheeasternpartofthenorth

ssuresystemsforminginthepolarjetstreamareasover

thewarmAtlanticwatersmoveeastwardandensurehighwindspeedsandamild

podIslandsandridgesalongthecoastare

edtootherareasintheworldatthesamelatitude,

hCape(71º),-4ºCisthe

lowestmonthlyaveragetemperatureatalevel.

Norwaydoesnothaveacentralizedsystemforcollectionofoperationalexperience

sfordowntimeduetoicingorlowtemperaturearetherefore

notavailable.

Sofar2003,97MWofwindpower,estwind

farmHavøygavlenislocatedatLatitude71ºoughit

isthenorthernmostwindfarmintheworld,

averagewinter(January)temperatureisabout-6ºdfarmwasinstalled

autumn2002,ingtotheoperator,very

43

meminorproblemswithbatteriesandoil

possiblyduetolowtemperaturehavebeenreported.

Nord-TrøndelagElektrisitetsverkhaxperiencewithoperationofwindturbinessince

indturbinesarelocatedinthemiddleofNorway,aroundlatitude65º.One

VestasV66hasbeeninstalledat230mabovealevel,andasmallwindfarmis

heinstallationofthewindturbinesthere

omeyears

ofusagetheturbineswereequippedwithheatingofthehydraulicoiltoprevent

problemsrelatedtorestartinlowtemperatures.

elocatedat

Latitude62ºyarcticadaptationmadeisthe

rtotheturbinesatHavøygavlen,theyhaveno

ousproblemswithlowtemperaturesoricinghavebeen

asbeenreportedoccasionallyatthetimeofstandstillofthe

eenpossibletostartturbineswithbladescoveredwithicebyforced

heforcedstarticehasshedfromtheblades.

5.3.4Switzerland

Switzerlandhaslongexperienceofwithwindenergysiteasssmentsinalpineareas.

Suchsitexperienceharshclimaticconditionssuchaslowtemperatures,high

turbulenceandextremegusts.

InSwitzerlandveralwindenergyprojectshavebeencarriedoutinicingandinlow

rbinesthatexperienceicingandlowtemperatureslocateat

desrangefrom1300metreto3000metreabovethealevel.

Typicallysitesbelow2000metreabovealevelexperiencelighticingandsiteswith

antexperienceon

theuofwindenergyunderclimaticallyextremeconditionswillbegained,withthe

800kWplantontheGuetschnearAndermatt(2300mabovea-level)whichwas

thefirstwindturbineinSwitzerlandthatus

rprojectssuchas

(2200mabovea-level)aswellasCrêtMeuron(1300mabovea-level),

willincreatheknowledgeaboutwindenergyproductioninalpineregioninharsh

climaticconditions.

5.3.5USA

Operationalexperienceofwindturbinesincoldandicingclimatesislimitedandthe

private,unsubsididnatureofmostinstallationsmakecollectingdataonsystem

downtimedifficult.

Asstatedpreviouslywindturbineshavebeinginstalledinthreegeneralclimatic

orthcentralregion,suchasthe200MWwind

plantsintheLakeBenton,Minnesotaarea,snowfallandcoldtemperaturesarecommon

44

orsintheregionshave

orth-east

andnorthwestpartsoftheUS,suchasthe6MWplantinSearsburg,Vermont,turbines

arelocatedonlowaltitudemountainridgesorincoastalregimeswhereicingis

common,b

mostcasprecipitationisintheformofsnow,whichdoesnotimpactturbine

mercompanyUSWindpowerconductedextensivetestsofwind

ghaltitudemountainridge

experiencedverrimeiceandcold,herearchfromthe

sites,whichwereactiveinthemidtolate1980’ersites

areatmuchlowerelevationsandthusdonotexperiencethesamerimeiceconditions.

Thelastclarificationofsitesarealongthearcticcoast,suchasthe0.5MWplantlocated

inKotzebue,itesdoexperiencecoldtemperatures

andhighdensityairflows,es

installedintheareasareoutfittedwithcoldweatherpackages,includingoilheaters

theturbinesinstalledhaveincludedbladeheating

options,otherthentheuofblackpaintedblades.

Ofthesitesoutfittedwithgovernmentalsupportedmonitoringsystems,reportsof

downtimeresultmoreduetoturbinemaintenanceincoldclimatesascomparedtoactual

operationalissues.

5.3.6Canada

OneoperatorinCanadahasidentifiedoverproductionincoldtemperatureshasbeingits

600kWTackemachinelocatedinTiverton,

Ontario,cond-averagedpowerpeaksof950kWwererecordedin-20°Cweatherand

thegeneratoroverheatedandtrippedout[13].Alsoona65kWBonusmachinelocated

inKuujjuaq(58°N),a5-minuteaveragepoweroutputof89kWwasrecorded[13].

YukonEnergyCorporationhasasignificantamountofexperienceinoperatingwind

panyowns

twoturbines:one150kWMarkIIIBonusandone660kWV47VestasinHaeckelHill,

Yukon(altitude1430meters).Theywereinstalledin1993and2000respectively[32].

Maissan[32]reportsthatlowtemperaturesteels,syntheticlubricantsandheating

systemsforitemslikegearbox,generatorandelectricalcabinetshaveworkedwell.

However,anemometersandaerialpowerlinesprovedtobeadverlyaffectedbyin-

tion,problemswereencounteredwiththeicedetectorthatcontrols

detectorwasremovedandthe

ricedetectorwasinstalledbutoutsidethe

rdedapproximately800hoursofrimeicingat

thesite[32].

BadontheexperiencesofYukonEnergy,Maissanidentifiesicingasprobablythe

axperimentedwithaprotectivecoatingontheirfirst

tiontothemoreobvioussolutionsfor

45

coldweatherclimates,herecommendsthatturbinesbefittedwithfullbladesurfaceice

protectionandwishedthatsuchasystemhadbeenavailableforthecondturbine

wouldliketoetheoperatingtemperaturerange

reachdownto-40°C[32].

46

6EXISTINGSTANDARDS,REQUIREMENTS

ANDRECOMMENDATIONS

6.1WINDTURBINES

Certifyingwindturbinesforcoldandmountainousregionsrequiresreliableprocedures

forthepredictionoftheamountoficeaccretionduringstandstillandoperation.

Internationaldesignstandardstakeicingloadcasintoconsiderationindifferentways.

TheIEC-61400-1WindTurbineGeneratorSystems-Part1SafetyRequirements

recommendstotakeiceloadsintoaccountbutaspecialloadcaisnotgivenandno

minimumicerequirementsaregivenforstandardwindturbines[34].Germanischer

Lloydrequiresthattwoicingcasforrotatingpartsandonefornon-rotatingpartsmust

atingpartsthetwocasare“all

bladescoveredwithice”and“allbutonebladeicedover”.Fornon-rotatingpartsicing

formulafor

calculatingthedesigniceloadsisgiven.[35,36]TheDanishEnergyAgencygivesit’s

laicecharacteristicsandformulafor

dsfromdynamicaicebehaviourare

advidtobenoticed,butnoclearrecommendationhowtoestimatethodynamic

ftherotatingpartsfollowstheguidelinesofGermanischerLloyd.

HoweveratNorthSeathedesignicethicknessisrecommendedtobeincreadfrom

30mmto150mmduetothewatersprayforpartslessthan20mfromthewaterlevel.

[37]

6.2RESOURCEESTIMATIONANDPOWERPERFORMANCE

MEASUREMENTS

TheIEC-61400-12WindTurbineGeneratorSystems-Part12Windturbinespower

performancetestingtsnorequirementsforequipmentsordatatreatmentudin

powerperformancetestinginlowtemperaturesorinicingclimate[38].

47

7SUMMARY

Windturbineshavebeenandarelocatedtosuchsiteswhereturbinesareexpodto

suchlowtemperaturesoutsidethestandardoperationallimitandtositeswhereturbines

faceicing,whichretardenergyproduction,tlycapacityof

about500MWlocatesonsites,whichcanbedefined,ascoldclimatewindturbinesites.

rbineshavebeen

recordedtooperateincoldclimateinScandinavia,NorthAmerica,EuropeandAsia.

peratureandicing

ed

measurementequipmentshouldbedesignedforlowtemperatureandicingclimateu.

yasmall

amountoficemayreducemeasuredwindspeedsignificantlyandlargeiceaccretions

mpleasmallamountofrimeonthecupsand

shaftofananemometermayleadtounderestimationofwindspeedabout30%atwind

speedof10m/earchhasbeendoneinthisfieldanddevicessuitablefor

tiontothe

measurementinstrumentsitlfotherpartsofthemeasurementsystemshouldalsobe

,connectorsandcabletiesspecified

atingfortheboomofwind

nsorsinvereicingclimatesshouldbeprovidedtoavoiddistortedresults.

Extensiveandreliabletemperaturedataiscommonlyproducedforweatherforecasts.

Suchtemperaturerecordingnabletheestimationofextremetemperaturesand

easurementsarehoweverrareandarenot

ssibletocalculateestimation

ofincloudicingfromvisibilityobrvations,whichincludecloudbaheight

hat

gisconsideredto

deterioratepowerproduction,itisadvisabletoaddicingmeasurementstoresource

estimationmeasurementsifsucharecarriedout.

ectorsformeteorological

sopossibletomeasureicingindirectlywithdew

hereisanalogybetweenanemometerandawindturbine,

persistencyoficingmaybeevaluatedeconomicallywithat-upoftwoanemometers,

ingtothecurrent

knowledgeicedetectorsaremostsuitableforicingtimemeasurementsandtherefore

alsoforcontrollingapossibleheatingdevis.

Meteorologistshavedevelopedmodelsforestimationofdifferenttypeofatmospheric

icingandtheeffectsoficingandfromotherstandpointaviationindustryhasdeveloped

modelstocalculateweightandshapeoficeaccumulationsontheleadingedgeofa

omplexity

oficingphenomenonandaerodynamicsaswellascurrentperformanceofmodern

48

personalcomputers,thedevelopmentofmoreaccuratemodelshasbeenmoderate.

Mapstodescribeannualicingtimehavebeendevelopedbutstandardidmethodto

calculatelocalicingtimefrommeteorologicalmeasurementsstilllacks.

Technicalsolutionsforwindturbinesoperatingatlowtemperatureoraticingclimate

peraturespecifiedmaterialsandoilsshouldbeudif

rbinemanufacturershave

tiontocold

specifiedmaterialsud,

tionto

ectors,

coatingsthatpreventicetosticktothebladesanddifferentbladeheatingsystemsare

available.

Windturbineshavebeensitedtocoldclimatesitesforsomeyearsandtodayoperational

dinaviadowntimeforolderturbineshavebeenrecordeddue

tolowtemperature,modernturbinesinsteadarealreadyadaptedtothelowtemperatures

peratureshavealso

recordedtoextendthedurationofmaintenanceandreparationbreaksduringwinter.

Severityoficingvariesalotdependingonlocalparameterspeciallyaltitude

asbeen

recordedtoretardenergyproductionatelevatedsitesinScandinavia,Alpineregionsof

orway

forexampleicinghavenothadthatkindofeffecttowindpowerproductionthatit

wouldhavebeenrecorded,eventhoughturbineslocateupto200mlevelabovealevel

usunderlinesthefactthat

rtolowtemperatureicingandsnowhas

y

reicingclimateofCanadaand

Finlandsystemsthatkeepbladesfreeoficehavebeenfoundcompulsory.

49

REFERENCES

[1]InformationonWindEnergyinColdClimates/

[2]StatisticsandinfoforSwedishwindturbines:.

[3]InfoforNorwegianwindturbines:.

[4]InfoforSwisswindturbines:.

[5]InfoforUSAwindturbines:.

[6]InformationforCanadianwindturbines:.

[7]InfoforFinnishwindturbines:/ene/tuloksia/.

[8]Laakso,T.,Follow-upofwindparkinOlostunturi(inFinnish)2001,VTTEnergy,

Espoo.62p.+rgyreports24/2001.

[9]Marjaniemi,M.,Laakso,T.,Makkonen,L.,Wright,J.,ResultsofPoriwindfarm

measurements,2001,VTTEnergy,Espoo.83p.+app.5p.

[10]Antikainen,P.,Peuranen,S.,Iceloads,V,Windpower

productionincoldclimates,ProceedingsofanInternationalconference,Levi,Finland

hMeteorologicalInstitute.

[11]Tammelinetal,Meteorologicalmeasurementsundericingconditions–Eumetnet

SWSIIproject,FMIreports2001:6

[12]Peltola,E,MarjaniemiM,StiesdalHandJärvelä,JAnicepreventionsystemfor

thewindturbineblades,1999EuropeanWindEnergyConference,1-5

March1999,Nice,France,pp.1034-1037.

[13]Leclerc,C.,Masson,C..AbnormalHighpoweroutputofwindturbineincold

weather:dingsoftheCanWEASeminarandthe15th

CanWEAConference&TradeShow'ber27-28-29,Rimouski,Canada.

CanadianWindEnergyAssociation,1999.P.190-199.

[14]Lemström,B.,Mannila,P.,Marjaniemi,M.,Operationalenvironmentfor

generatorsinwindturbines,1999EuropeanWindEnergyConference,1-5

March1999,Nice,France,pp.825-828.

[15]Stiesdal,H.,Kru.H.,10Yearswitharcticmodificationsamanufacturer’s

experience,ProceedingsoftheBOREASIVconference,Hetta,Finland1998,Finnish

MeteorologicalInstitute.

[16]Hughes.“ColdClimateTestingofDouble-endedFiberglass/SteelRootStud

SubstructuresforWindTurbineBlades”PrentationattheAmericanWindEnergy

AssociationConference,WashingtonDC,June3-7,2001

[17]/

[18],archfor“cableties”.

50

[19]Tammelin,B.,Cavaliere,M.,Holttinen,H.,Morgan,C.,Seifert,H.,Säntti,K.,

WindEnergyProductioninColdClimate,MeteorologicalPublicationsNo.41,Finnish

MeteorologicalInstitute,Helsinki.41p.,2000.

[20]Tammelin,B.,Säntti,K.,IcinginEurope,ProceedingsoftheBOREASIV

conference,Hetta,Finland1998,FinnishMeteorologicalInstitute.

[21]Laakso,T.,Holttinen,H.,WindenergystatisticsofFinland,Yearreport2001,

VTTProcess,Espoo.39p.+/T7511/02.

[22]Tammelin,B.,Seifert,mantaras,K.,roceedings

ofBOREASIV(inetal.).FinnishMeteorologicalInstitute,Helsinki.

Inprint.

[23]Tammelin,B.,ormationonanemometersandfrostprevention

hmeteorologicalinstitute.34pp.

[24]Tammelin,B.,dings

ofBOREAS(inetal.).Finnishmeteorologicalinstitute.p.241-261.

[25]Kenyon,P.R.,etryinNewEnglandmountainicingenvironments.

InproceedingsofBOREASII(inetal.).Finnishmeteorological

institute.154-163.

[26]Botta,aliere,anemometerperformanceinicing

dingsoftheEWEC’99,Nice,t.

[27]UrsmanualofLabkoLID-3200.

[28]InstallationandoperationmanualofInstrumarIM101v2.4.

[29]Makkonen,L.,Jäändetektointikosteusmittauknavulla,Reporttothefoundation

ofVilho,YrjöandKalleVäisälä/FinnishScienceAcademy,2000,inFinnish.

[30]TammelinB.,Icingofanemometeranditffectonestimationofwindenergy

IIconferenceproceedingspp.85-96.21–25March1994,

Pyhätunturi,hMeteorologicalInstitute,Helsinki.

[31]Tammelin,B.,Joss,palainen,J.,eportontheEUMETNET

project“SpecificationofSevereWeatherSensors”.FinnishMeteorologicalInstitute,

Helsinki.154p.

[32]MaissanJ.F.,Windpowerdevelopmentinsub-arcticconditionswithvererime

icing,,Canada,May2000.

[33]CraigD.F.,CraigD.B.,MonitoringoficingeventsonfjellsinnorthernCanada.

InproceedingsofBOREASII(inetal.).Finnishmeteorological

institute.154-163.

51

[34]IECstandard61400-1:WindTurbineGeneratorSystems,Part1:Safety

Requirements,IEC1998.

[35]GermanischerLloyd,RulesandRegulationsIV-NonMarineTechnologyPart-1

WindEnergy,1993Edition.

[36]GermanischerLloyd,1993EditionofRulesandRegulations

IV-NonMarineTechnologyPart-1WindEnergy,March1998.

[37]TheDanishEnergyAgency,RecommendationsforTechnicalApprovalof

OffshoreWindTurbines,December2001.

[38]IECstandard61400-12:WindTurbineGeneratorSystems,Part1:Windturbines

powerperformancetesting,IEC1997.

[39]GoodrichSensorSystems:Model0871LH1IceDetector,SpecificationsSheet.

[40]Wright,W.B.(1999)UrManualfortheNASAGlennIceAccretionCode

/CR-1999/209409

[41]Messinger,B.L.(1953)EquilibriumTemperatureofanUnheatedIcingSurfaceas

lofAeronauticalScience27(1):29-42

[42]TroenI.,PeternE.L.,EuropeanWindAtlas,fortheEuropenCommission-Ris∅

NationalLaboratory,Roskilde,Denmark,1989.

[43]MakkonenL.,EstimationofWetSnowAccretiononStructures,ColdRegions

ScienceandTechnology,Vol.17(1989)83-88,ElvierSciencePublishersB.V.

Amsterdam.

[44]d,en,Proceedings,6thInternationalWorkshopon

AtmosphericIcingofStructures,IWAIS,Budapest(1993)79.

[45]d,en,Improvednumericalmodelforwindturbineicing,

Proceedings,7thInternationalWorkshoponAtmosphericIcingofStructures,IWAIS,

Chicoutimi,Quebec(1996)373.

[46]en,Modelsforthegrowthofrime,glaze,iciclesandwetsnowdeposits

ophicalTransactionsA358(2000)2913.

[47]en,gionsScience

andTechnology10(1985)105.

[48]en,rass,Iceaccretiononcylindersandwires,National

,TechnicalReportno.23649(1984)44pp.

[49],,Experimentalevidenceformodyfyingthecurrentphysical

modelforiceaccretiononaircraftsurfaces,Proceedings,ThirdInternationalWorkshop

onAtmosphericIcingofStructures,IWAIS,Vancouver(1991)58.

52

53

[50]Tammelin,B.,Cavaliere,M.,Holttinen,H.,Morgan,C.,Seifert,H.,Säntti,K.,

WECOFinalreport,2000,EuropeanCommission,Brusls.142p.

[51]e-mailfromNordexEnergyGmbH.

[52]e-mailfromNordicWindpowerAB.

[53]e-mailfromLagerwey.

[54]Makkonen,L.,Laakso,T.,Marjaniemi,M.,Finstad,K.J.,Modellingand

PreventionofIceAccretiononWindTurbines,WindEngineering25(2001)3.

[55]/varme/underlag/

[56]/

[57]RonstenG.“Candelayedstallbecaudbyiceaccretionontheleadingedgeofan

airfoil?”,FFAP-A-981,Stockholm,1993

[58]GananderH,“WP3:ModellingofIceloads”,ProceedingsoftheBoreasVI

conference,Pyhätunturi,Finland,April9thto11th,2003