1、12FTM07AGMA Technical PaperValidation of a Model ofthe NREL GearboxReliability CollaborativeWind Turbine GearboxBy C. Halse, Z. Wright, andA. Crowther, Romax TechnologyInc.Validation of a Model of the NREL Gearbox ReliabilityCollaborative Wind Turbine GearboxChristopher Halse, Zachary Wright and Ash
2、ley Crowther, Romax Technology Inc.,The statements and opinions contained herein are those of the author and should not be construed as anofficial action or opinion of the American Gear Manufacturers Association.AbstractGearboxesinthewindindustryhavebeensufferingfromapoorreputationduetomajorissueswi
3、threliability.There has been a long list of issues; e.g. grind temper, material inclusions, axial cracking in bearings, poorloadsharingonshaft-bearingarrangements,significantgearmisalignment,bearingringcreep,gearscuffing,gear and bearing micro-pitting; all of which are common and often serial proble
4、ms. There has beenimprovement in the last few years for some of the products, yet it is not uncommon for wind sites built asrecently as 2008 to have 20 - 40% of gearboxes requiring a component replacement (such as a high speedpinion or intermediate shaft bearing) already (by 2012) and 5-10% complete
5、 gearbox failures. An importantprogram for the industry, “The Gearbox Reliability Collaborative” (GRC), has been funded by the USDepartment of Energy and run by the National Renewable Energy Laboratory for several years to aid theindustry in improving the reliability of this key component. The colla
6、borative has brought togethermanufacturers, academia, national laboratories, engineering consultants and gear and bearing softwareproviders as part of a program to model, build, simulate and test gearboxes with a goal to improve reliabilityand reduce the cost of energy.The team at NREL have instrume
7、nted two gearboxes with over 125 channels, for measurements such asplanetarytoothloaddistributions,ringgearhoopstrains,planetbearingloaddistribution,sunorbitandcarrierdeflection. They were then subjected to a rigorous testing regime, both up-tower and on the NREL 2.5MWdynamometer. Romax Technology h
8、as been a collaborator in the GRC Analysis Group and has developeddetailed computer simulation models of the gearbox including gear macro and micro-geometry, bearingmacro and micro-geometry, structural stiffness of gearbox housing, carrier and ring gear, system clearancesandpreloads,andsurroundingbo
9、undaryconditions(suchasmainshaft,rotorhubandbedplate).Themodelis used for accurate simulation of the whole system deflections and the prediction of the resulting gear andbearing contact conditions under various loading conditions.Copyright 2012American Gear Manufacturers Association1001 N. Fairfax S
10、treet, Suite 500Alexandria, Virginia 22314October 2012ISBN: 978-1-61481-038-43 12FTM07Validation of a Model of the NREL Gearbox ReliabilityCollaborative Wind Turbine GearboxChristopher Halse, Zachary Wright, and Ashley Crowther, Romax Technology Inc.IntroductionWind power has grown to be an importan
11、t part of the energy mix in the U.S., contributing 35% of all newgenerating capacity constructed in the last 5 years, and with a total of 48,611MW of installed capacity and8,900MWunderconstructionasofMay2012AmericanWindEnergyAssociation. Thestatewiththelargestinstalled capacity, Texas, can sometimes
12、 generate over 20% of demand from wind New York Times. Windpowerhasthelowest“lifecycleemissions”ofallenergyproductiontechnologies,payingbacktheenergyusedin turbine manufacturer, installation, operation, maintenance and decommissioning in three to six monthsEuropean Wind Energy Association.Most wind
13、turbines contain speed-increasing gearboxes to convert the slow main rotor speed into the1000 rpm range for convenient generator operation. However, many wind turbine gearboxes have a poorreputationduetopoorreliability. Therehavebeenmanyissues;grindtemper,materialinclusions,axialcrack-ing in bearing
14、s, poor load sharing on shaft-bearing arrangements, significant gear misalignment leading topremature wear, bearing ring creep, gear scuffing, and gear or bearing micro-pitting. These are all commonand often serial problems. There has been improvement in the last few years for some products as commo
15、nfailure modes have been addressed, yet it is not uncommon for wind sites built as recently as 2008 to have20-40%ofgearboxesrequiringacomponent replacement(such asa highspeed pinionor intermediateshaftbearing) by 2012 and 5-10% complete gearbox failures.SomeOEMsaremanufacturingdirect-drivewindturbin
16、es,wherethegeneratorrotorturnsatthespeedofthemain rotor. This can lead to challenging structural engineering in maintaining a small generator air gap at alarge diameter, inconvenient assembly and difficult transportation. Capital costs are high for direct drive andlongtermreliabilityisunknown. Howev
17、er,somemanufacturersaccepttheseincreasedrisksasthereductionin operating expenditure from removingthe gearboxis perceivedto beso great. Withoutan improvedunder-standing of gearbox failure modes and more reliable gearbox products, this important market for the gearindustry is in danger of being eroded
18、 by growth in the use of direct drive machines.Animportantprogramfortheindustry,“TheGearboxReliabilityCollaborative”(GRC),hasbeenfundedbytheDepartment of Energy and run bythe NationalRenewable EnergyLaboratory (NREL)for severalyears. Thecollaborative has brought together OEMS, component suppliers, a
19、cademia, national laboratories, engineer-ing consultants and software providers as part of a program to test gearboxes, validate software models andverify design assumptions. The overall goal is to improve gearbox reliability and reduce the cost of energyproduced by wind turbines.TheNRELGRCteamhasde
20、liveredanextensiveinstrumentationandtestingeffortontwo750kWgearboxes,described further below. The contribution of the authors to the project has been as members of the AnalysisTeam as well as designers of revisions to the next gearbox (being built in 2013). The focus of this paper is acomparisonbetw
21、eenmeasurementandsimulationofkeyparameters;gearloaddistributions,annulusdeflec-tion,carrierdeformationandplanetarygearloadsharing. Thesimulationresultsthatarerobustandthosethatare sensitive to hard-to-predict parameters such as manufacturing and assembly variations will beoutlined.Lessons learned in
22、 how best to apply computer-aided-engineering tools to improve wind turbine gearboxreliability will be described.Test articlesTestswerecarriedoutontwo750kWwindturbinegearboxesobtainedbyNREL. Theyweremodifiedtohavetypical wind turbine bearing arrangements and using typical design practices. The aim o
23、f this original4 12FTM07modification was not to solve all problems in the design, ratherit wasto creategearboxes thatwere similartothe fleet of modern wind turbine gearboxes in operation. One of the gearboxes was tested in the NRELdynamometer, the other was operated in the field in a wind turbine.Th
24、e gearboxes are speed increasers, with the input shaft attached to the wind turbine rotor and the outputshaft attached to a high speed coupling which in turn drives a generator. The first (input) stage is a planetaryhelicalgearstagewiththreeplanets,withinputonthecarrier,theringfixedandtheoutputonthe
25、sun. Thesunis splined into a hollow shaft to allow some float, and then the second and third stages are parallel helicalgears. The carrier bearings and planet gear bearings are cylindrical roller bearings, whilst each of the hollowshaft, intermediate shaft and high speed shaft are supported by a dou
26、ble-row taper roller bearing and a cyl-indricalrollerbearing. Oilisfedintotherotatingcarrierandrotatinghollowshaftvia“oilfeedrings”toenableoiltoreachtheplanetbearingsandthesuntohollowshaftspline. Thelabelsusedforthegearsandbearingsareshown in Figure 1.Measurement campaignNREL smeasurementcampaignwas
27、extensivewithparticularfocusonmeasurementofsystemdeflectionsand load distributions in planet-ring meshes and planet bearings. The gearboxes were instrumented withover 125 channels, for measurements including planetary tooth load distributions, ring gear hoop strains,planet bearing load distribution,
28、 sun gear orbit and carrier deflection. They were then subjected to arigoroustestingregime,bothup-towerandontheNREL2.5MWdynamometer. Themeasurementsusedsofarforthevalidation work are briefly described below, but for more details of these and all the other measurements thefullGRCreportisfromPhases1&2
29、ispubliclyavailablehttp:/www.nrel.gov/docs/fy11osti/51885.pdf,NREL/TP-5000-51885 Gearbox Reliability Project Report: Findings from Phase 1 & Phase 2 Testing.The ring gear strain was measured in the tooth roots at several positions across the face width and at threepositions around the circumference
30、to simultaneously capture the loading on each planet.Figure 1. Illustration of gearbox model with part labels (gearbox housing hidden)5 12FTM07Strain gauges were also placed externally on the ring gear. This allowed the hoop stress to be captured, butalsostraingaugeswereplacedacrosstheface widthwhic
31、h allowedthe toothload distributionto beobservedexternally from the gearbox.Grooves were machined circumferentially in the planet bearing inner rings at various angles relative to thepeak part of the load zone in order to capture the extent of the load zone, the load on the upwind bearingrelative to
32、 the downwind and also the overall planet load sharing.Proximity probes were attached to the housing at locations detecting the proximity of machined rings on theplanet carrier in the radial and axial directions. This allowed detection of the deflection shape of the planetcarrier.Gearbox simulationT
33、hegearboxwasmodelledintheRomaxWindsoftwarepackage. Thepowerof thesystem isthat itanalysesallcomponents(gears,bearings,housings,carrier,spline,bushings,mainframe)inparallel,includingsystemnonlinearities. The model is solved with an iterative method to calculate static equilibrium. This enablesaccurat
34、e prediction of system deflections and misalignments, which can then be fed into component lifecalculations using AGMA or ISO methods, included in the same software.The gearbox mounting arrangement in the wind turbine nacelle is the so-called “three-point mounted”. Themain shaft that connects the ma
35、in wind turbine rotor to the gearbox is only supported by a single sphericalroller bearing, and the gearbox is supported by two elastomer bushings mounted to the main frame. In thisarrangement over-turning and yawing moments from main rotor weight and aerodynamic loading are trans-mitted to the gear
36、box, as the spherical roller bearing has no reaction to an applied moment. Therefore themodelincludesarepresentationofthemainframe,mainbearingandmainshaftasshowninFigure 2. Loadsare applied upwind of the main bearing flange.The shafts, housing, torque arm, main frame, ring gear and planet carrier ar
37、e modelled with finite-elements.ShaftsaremodelledasTimoshenkobeamelements,whilstallotherpartsaresolid-meshedandthenreducedto super-elements using static condensation.Figure 2. Illustration of the model showing main bearing, main frame, main shaft and gearbox6 12FTM07Gear meshes are modelled with too
38、th contact analysis. The face width is broken down into to multiplefinite-widththin-strips. Thebendingstiffnessofeachstripiscalculatedusingafiniteelementmeshbasedonthetoothandrootgeometry. LocaldeformationsarecalculatedusingHertziancontacttheory. Thenumberofstrips in contact depends on the load, mis
39、alignment, tooth micro-geometry modifications and twist along theface width (important for the sun gear). The approach is similar to the many tooth analysis codes on themarket,butherethetoothcontactissolvedinparalleltothefullsystemdeflections. Thepositionoftheresult-ant load on the face width change
40、s the loads on the bearings and hence the misalignment of the gears, soiterations are required to achieve the force and moment balance.The gearbox spline is modelled in a similar way to gears, by breaking down into strips and calculating thedeflection of each strip including the effect of crowning a
41、nd a finite-element mesh to get the tooth stiffness.Loadsharingbetweenteethinasplineiscriticallyaffectedbypitch errorsbetween theteeth, andtypically itisassumedthatonlyhalftheteetharecarryingload. Theeffectofpitcherrorswasnotaccuratelycapturedinthemodel, as the calculated spline stiffness was varied
42、 to checkthe sensitivityof theresults. Variations insplinestiffnessdidnotchangetheresultsofinterestinthispapersignificantly,sofocuswasnotputonimprovingthemodel in this area.The roller bearings are modelled based on a similar approach to gears, with a combined numerical and ana-lytical approach. Bear
43、ing manufacturers typically do not share bearing internal details so roller size, numberof rollers, roller profile modification and so on are estimated based on the published load capacity and pastexperience. This gives a reasonable estimate of bearing stiffness. The operating radial and/or axial in
44、ternalclearances are calculated based on the bearing clearance as supplied (i.e., CN, C1, C2 etc.), the shaft andhousing fits, and thermal expansions based on estimated temperature differences across each bearing, assuggestedbyANSI/AGMA/AWEA6006-A03. Meshmisalignmentspredictedbythemodelwereshowntobe
45、sensitive to bearing clearance, so this is an area of the model in which it is important to get a goodprediction.The model may be used for investigations of the sensitivity tomanufacturing tolerancesand othervariations.Allmechanicaldimensionswillhaveatoleranceandthereareamultitudeofunknowns(forinsta
46、nce,theexactoperatingtemperaturesandthereforethe operatingbearing clearance). Although“virtual” investigationsintothe effect of these variables are straightforward, the investigator may be overwhelmed with the number ofpossibilities and studies that can be carried out. The validation exercise guides
47、 these studies and highlightswhere tolerances or unknowns are particularly important.Model validationRing gear load distributionThe measurements of ring-planet load distribution showed clearly that there was a different load distributionat each of the azimuth positions. This suggests a misalignment
48、between the planet carrier and the ring gear.The model (with no manufacturing tolerances or differences between each planet) showed a good match attwoofthepositions,andareasonablematchatoneposition,seeFigure 3. Thisoccursinthemodelonlywhenthecorrectoff-axisloadduetotheappliedweightwasappliedtothemai
49、nshaft. Theeffectof thisweight istodeflect the carrier relative to annulus such that the gear misalignment varies with carrier azimuth.Improvedvalidationwas achievedby modifyingthe connectiondimensions betweenthe meshload pointandthe ring gear solid finite-element mesh. The annulus is reduced to a stiffness matrix (super-element) withnodes at the gear mesh points. The nodeat thegear meshpoint isconnected tothe ringgear finiteelementsby RBE2 (fully rigid beam) connections. Judgment is required when making these connections as to howmanyfiniteelem
