4
Wind Energy: General Powered by wind Fact Sheet 1

Wind Energy: General - Department of Energy

  • Upload
    others

  • View
    3

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Wind Energy: General - Department of Energy

Wind Energy: General

Powered

by wind

Fact Sheet 1

Page 2: Wind Energy: General - Department of Energy

HistoryHumans have been using wind powersince ancient times to propel sailboatsand sailing ships. The wind wheel ofHeron of Alexandria marks one of thefirstknowninstancesofwindpoweringamachine inhistory1,2. Thefirstelectricitygenerating wind turbine was a batterychargingmachine installed in July1887by Scottish academic, James Blythto light his holiday home in Marykirk,Scotland3.SomemonthslaterAmericaninventor Charles F. Brush built the firstautomatically operated wind turbinefor electricity production in Cleveland,Ohio3.Inthe1890s,Danishscientistand

inventor Poul laCourconstructedwindturbines to generate electricity, whichwas then used to produce hydrogen4.TheSmith-Putnamwind turbinewas thefirstmegawatt-classwind turbinewhichwas synchronized to a utility grid inVermontin1941.

Siemens and Vestas are the leadingturbinesuppliersforoffshorewindpower.Dong Energy, Vattenfall and E.on arethe leading offshore operators5. As ofOctober2010,3.16GWofoffshorewindpowercapacitywasoperational,mainlyinNorthernEurope.

Fig. 1 The first automatically operated wind turbine, built in Cleveland in 1887 by Charles F. Brush. It was 60 feet (18 m) tall, weighed 4 tons (3.6 metric tonnes) and powered a 12kW generator6.

Fig 2 It is easier to appreciate the size of new 10-MW turbines with comparisons to previous generations. The Clipper Windpower Britannia turbines, for example, will have towers about as high as 50-storey buildings7.

AsofMay 2009, 80 countries around theworld usewindpoweronacommercialbasis8withChinatakingtheleadat42,287MW(March20119,10).Thetotalcapacityofwindinstalledby2010was194,400MW11. Wind power accounts for approximately 24% ofelectricityuse inDenmark,16.4% inSpain12,15% inPortugal,10%in theRepublicof Ireland,and9½% inGermany. SouthAfrica’sfirst commercial wind farm, Darling Wind Farm (5.2 MW), wascommissionedinMay2008.

Growth trendsDespiteconstraintsfacingsupplychainsforwindturbines,theannualmarket forwindcontinued to increaseatanestimated rate of 37%, following 32% growth in 2006. Interms of economic value, the wind energy sector hasbecomeoneofthekeyplayersintheenergymarkets,withthe totalvalueofnewgeneratingequipment installed in2007reaching€25billion,orUS$36billion13.

Over the past five years, the average growth in newinstallationshasbeen27.6%perannum.Intheforecastto2013theexpectedaverageannualgrowthrateis15.7%13,14.More than 200 GW of new wind power capacity couldcomeonlinebeforetheendof2013.Windpowermarketpenetrationisexpectedtoreach3.35%by2013and8%by201813,14.

Wind turbine typesThereareprimarilythreetypesofwindturbines:

1) TheHorizontalAxisWindTurbine(HAWT),2) TheVerticalAxisWindTurbine(VAWT),and3) Othervariations.

Fig 3. Vertical Axis Wind Turbine (VAWT), Horizontal Axis Wind Turbine (HAWT) (source Wikipedia)

Fig 4. Airborne wind generator of Savonius style (source Wikipedia)

Page 3: Wind Energy: General - Department of Energy

Fig 5. Components of a typical Horizontal Axis Wind Turbine (HAWT)

Wind turbine power generationAwind turbineextractsandconverts theenergy inwindintoelectricalenergyviawindthatflowsoverandcauseslift of the wind turbine blades that turn an electricitygenerator.Thepowerofawindturbinecanbedescribedbythefollowingequation:

Fig 6 Typical Wind Turbine Power Curve15

• Powerofawindturbineincreaseswithcubeofthewindspeedv3

• PowerofawindturbineisproportionaltothesweptareaAoftheblades,henceincreasesquadraticallywiththerotordiameter D(A=π.D2/4)

P(W) = cp1/2ρAv3……………1

Ppowerinwatts,cppowercoefficient,ρairdensity (kg/m3),Asweptareaofblades(m2),vwindspeed(m/s)

From above equation it follows that:

ThetheoreticalwindenergyextractionlimitofaHAWTis59,36%(Betzlaw)(meaningthetheoreticalmaximumenergyinthewindthatcanbeextractedbyaHAWTis59,36%).Ingeneral,also taking intoaccountconversion losses forexample, thepowercoefficientcpofawindturbineisintheorderof30to40%16.

TheavailabilityofmodernHAWT is typically 98%andmore.Thisisthecapabilityofthewindturbinetooperatewhenthewindisblowingatorabovetheturbine’scutinwindspeedandwhenthewindturbineisnotundergoingmaintenance.

Thecapacityfactorofapowerplantistheratiooftheelectricalenergyproducedinagivenperiodoftimetotheelectrical

energy that could have been produced at continuousmaximumpoweroperationduringthesameperiod.

Globalwindturbinecapacityfactorstatistics17:Spain 24.6% Germany 16% Denmark 24% Sweden 19%

Capacityfactor(CF)isveryuseful,butisoftenmistakenforthetotalamountof timeawindturbineoperates(that is,iftheCFis25%,peoplethinktheturbineisonlyproducingelectricity 25%of the time). In reality,atmost sites, itwillbegeneratingatamuchhigherpercentageof the time(availability of 98% and more), but at less than its peakoutput level18(ratedwindspeed).

1 A.G.Drachmann,“Heron’sWindmill”,Centaurus,7(1961),pp.145–1512 DietrichLohrmann,“VonderöstlichenzurwestlichenWindmühle”,ArchivfürKulturgeschichte,Vol.77,Issue1(1995),pp.1–30(10f.)3 “JamesBlyth”.OxfordDictionaryofNationalBiography.OxfordUniversityPress.Retrieved2009-10-09.4 Price,TrevorJ(3May2005).“JamesBlyth-Britain’sfirstmodernwindpowerengineer”.WindEngineering29(3):191–200.5 Madsen&Krogsgaard.OffshoreWindPower2010BTMConsult,22November2010.Retrieved:22November2010.6 AWindEnergyPioneer:CharlesF.Brush.DanishWindIndustryAssociation.Retrieved2008-12-28.7 http://eetweb.com/wind/wind-turbines-go-supersized-20091001/8 “WindPowerIncreasein2008Exceeds10-yearAverageGrowthRate”.Worldwatch.org.Retrieved2010-08-29.9 GlobalWindEnergyCouncil10 Wald,MatthewL.(January11,2011).“China’sGallopingWindMarket”.TheNewYorkTimes.11 Globalwindenergycouncil12 EWEAAnnualStatistics201013 “BTMForecasts340-GWofWindEnergyby2013”.Renewableenergyworld.com.2009-03-2714 BTMConsult(2009).InternationalWindEnergyDevelopmentWorldMarketUpdate200915 http://www.pfr.co.uk/standfordhill/15/Wind-Power/119/Capacity-Factor/16 WindResourceAssessmentinAustralia–APlannersGuide,CSIRO,V1.1,October200317 Windstatsnewsletter,volume19number2(Spring2006)18 http://lightbucket.wordpress.com/2008/03/13/the-capacity-factor-of-wind-power/

Page 4: Wind Energy: General - Department of Energy

19 Measnet,EvaluationofSiteSpecificWindConditions,V1,November200920 IEC61400-12-1Powerperformancemeasurementsofelectricityproducingwindturbines21 GettingStartedwithWAsP9,Risø-I-2571(EN)

Wind resource assessmentApplyingequation1,adifferenceof5%inwindspeed(v)willresultinadifferenceof 15% power (P) output of the windturbine. It is therefore of the utmostimportance that the wind data beaccurate, representative, reliable andmeasuredatorasnearthewindturbinehub height (hub connecting blades)aspossible.Theheightofthetopwindspeed measurement shall be at leasttwothirdsoftheplannedhubheight19.Aminimumofoneyearofon-sitewinddata is typically required by financialinstitutions. Only reliable, calibratedsensors and instrumentation should beused. Appropriate arrangement ofinstrumentsonthemeasurementmastisimportantforaccuratedatacapturing.Inparticular, theanemometershallbelocated to minimize flow distortions,

especially from mast and boominfluences20. The South African WindAtlasProject(WASA)windmeasurementmastarrangement,instrumentsummaryandmastsiteinformationcanbefoundonthefollowingwebsites:http://www.wasa.csir.co.za/andhttp://wasadata.csir.co.za/wasa1/WASAData.

The output of a dedicated on-sitewind measurement program can begraphically illustrated as a wind roseforthatsite.Thewindroseisagraphicillustration of the wind conditions,direction, speed and frequency overthe wind measurement period at aspecific location. The length of each‘spoke’ around the circle is related tothe frequency of time that the wind

blowsfromaparticulardirection.Eachconcentriccircle representsadifferentfrequency,emanatingfromzeroatthecentertoincreasingfrequenciesattheoutercircles.

By making use of the wind rose inFigure7,aswellas topographicalandterraindataof thewindmeasurementmast site,anObservationalWindAtlascan be constructed for the site. Bycombining the Observational WindAtlaswithawind turbinepowercurveas represented in Figure 6, the poweroutput of thewind turbine,wind farmandwindresourcemapFigure8(AEP,Annual Estimated Power production,GWh) can be calculated for that siteandimmediatesurroundings.

Meso and micro scale modelingMeso scale models, which were developed for numericalweather prediction and combined with micro scalemodels,are increasinglyused, refined,andvalidated in thecalculation and development of (numerical) wind atlasesinEgypt,Canada,theUnitedStatesofAmericaandChina,forexample.Ithasbeenshownthat,byutilizingappropriatemesoandmicroscalemodels,itispossibletocalculateanddevelopwind atlases in half the time and at a lesser costof making traditional (observational) wind atlas based onphysicalwindmonitoring.Thisismadepossiblebecausethemesoscalemodelusesavarietyofglobal,geophysicalandmeteorologicaldatabases,suchasthereanalysisdatabasewhichisagriddedhistoricalweatherdataproducedbytheUnitedStates’NationalCenters forEnvironmentalPrediction(NCEP)andAtmosphericResearch(NCAR).Furthermore,bycombiningthemesoscalemodelwiththemicroscalemodel,and taking into account local topography and roughnessprofile,forexample,regionalwindclimatescanbecalculatedandpresentedinanumericalwindatlas.Thestandarderrorinmeanwindspeedforthesemodelsisusually7%orlessonceuncertaintiesinthedataareremoved.

Fig 7. Wind Rose (source: Wikipedia)

Fig 8. Sample Wind Resource map21

Application of the numerical wind atlas saves time andmoney in calculating local wind climates that can beusedtopredict,withoutphysicalwindmeasurements,keyparameterssuchasthewindspeed,frequency,direction,estimatedpoweroutputandwindfarmlayoutatanysitecoveredbythenumericalwindatlas.

Byintegrating(GIS)thenumericalwindatlasanddatafromelectricity networks, roads, towns, and environmentallysensitive areas, for example, it enables decision-makers,developers and financiers to identify upfront potentialviable wind ‘hot spots’ in reasonable time and beforelengthyandcostlywindmeasurementsneedstobedone,.

By‘downscaling’(microscale)themesoscalewindatlastohigher resolution itcanbeusedtoexplore thesewind‘hot spots’ for feasibility studies, motivation of bankableprojectsandfornationalresourceplanning.

Wind Resource Assessment Summary (Source: Riso, DTU)

ConclusionTheapplicationofwindenergy forelectricitygenerationhas come a long way, and wind turbine theory andtechnologyarewellunderstoodandcommercialized.Bymakinguseof stateof theartwind resourceassessment

methods, national, regional and local wind resourceplanningcanbedoneandviablewind‘hotspots’canbeidentifiedanddeveloped.