5MW風電齒輪增速箱設計 (一級行星輪系)【說明書+CAD】
5MW風電齒輪增速箱設計 (一級行星輪系)【說明書+CAD】,說明書+CAD,5MW風電齒輪增速箱設計,(一級行星輪系)【說明書+CAD】,MW,齒輪,增速,設計,一級,行星,說明書,CAD
Wind Turbine Testing in theNREL Dynamometer Test BedJune 2000 NREL/CP-500-28411Walt MusialNational Renewable Energy LaboratoryBrian McNiffMcNiff Light IndustryPresented at AWEAs WindPower 2000 ConferencePalm Springs, CaliforniaApril 30May 4, 2000National Renewable Energy Laboratory1617 Cole BoulevardGolden, Colorado 80401-3393NREL is a U.S. Department of Energy LaboratoryOperated by Midwest Research Institute Battelle BechtelContract No. DE-AC36-99-GO10337NOTICEThe submitted manuscript has been offered by an employee of the Midwest Research Institute (MRI), acontractor of the US Government under Contract No. DE-AC36-99GO10337. Accordingly, the USGovernment and MRI retain a nonexclusive royalty-free license to publish or reproduce the publishedform of this contribution, or allow others to do so, for US Government purposes.This report was prepared as an account of work sponsored by an agency of the United Statesgovernment. Neither the United States government nor any agency thereof, nor any of their employees,makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy,completeness, or usefulness of any information, apparatus, product, or process disclosed, or representsthat its use would not infringe privately owned rights. Reference herein to any specific commercialproduct, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarilyconstitute or imply its endorsement, recommendation, or favoring by the United States government or anyagency thereof. The views and opinions of authors expressed herein do not necessarily state or reflectthose of the United States government or any agency thereof.Available electronically at http:/www.doe.gov/bridgeAvailable for a processing fee to U.S. Department of Energyand its contractors, in paper, from:U.S. Department of EnergyOffice of Scientific and Technical InformationP.O. Box 62Oak Ridge, TN 37831-0062phone: 865.576.8401fax: 865.576.5728email: reportsadonis.osti.govAvailable for sale to the public, in paper, from:U.S. Department of CommerceNational Technical Information Service5285 Port Royal RoadSpringfield, VA 22161phone: 800.553.6847fax: 703.605.6900email: ordersntis.fedworld.govonline ordering: http:/www.ntis.gov/ordering.htmPrinted on paper containing at least 50% wastepaper, including 20% postconsumer waste1 WIND TURBINE TESTING IN THE NREL DYNAMOMETER TEST BEDWalt MusialNational Wind Technology CenterNational Renewable Energy Laboratory1617 Cole BoulevardGolden, CO 80401Brian McNiffMcNiff Light Industry43 Dog Island RoadHarborside, ME 04642ABSTRACTA new facility has recently been completed at the National Renewable Energy Laboratory that allowsfull-scale dynamometer testing of wind turbine components, from generators to complete wind turbines.This facility is equipped with a 2.5 MW motor, gearbox, and variable speed drive system to deliver shafttorque. To simulate other aspects of wind turbine loading an MTS fatigue-rated loading system is fullyintegrated into the facility. This will allow actuators to cyclically load the structure in a variety of ways.Enron formally Zond Wind Corporation has installed the first test article in the facility to help maturethe Z-750 series wind turbine design. Tests include brake and control system tuning, endurance testing ofgear elements and bearings, and structural testing. Some aspects of the power converter will also betested. This paper describes the Dynamometer Test Bed and its capabilities. Also, an overview of theZond testing program is presented.INTRODUCTIONWind turbine systems have grown larger during the past decade. Although the cost of wind energycontinues to decrease, the per-turbine costs have increased making the empirical approach of prototypefield-testing increasingly risky. For new turbine designs, megawatt scale prototypes are expensive tobuild and long, single-unit production lead times consume the margins of production schedules. With thetypical rush from prototype to production, manufacturers often identify field problems too late to becorrected before large volume manufacturing runs are started. This can lead to premature field failuresand costly retrofits.Wind turbine drive problems are well-documented and may include gearbox failures and wear, bearinglife issues, generator failures, and controller induced system failures.1-3 Problems might arise duringinitial start-up, under extreme wind conditions, or may be related to long-term operation. In any case, thetime to find and solve these problems is when the first units are tested.For years, turbine manufacturers have relied on laboratory testing critical components, such as the rotorblades, to reduce their potential exposure to catastrophic field events.4 Until now they have not hadaccess to facilities capable of testing the whole turbine drive system, which is at the core of every turbinedesign.2At the National Renewable Energy Laboratory (NREL) a new Dynamometer Test Bed(DTB) was built totest wind turbine drive train systems and components. Born from industry requests, this new facility coulddramatically alter the approach to evaluating new wind turbine systems. It is now being used to verify thedrive train design of the Enron formally Zond Wind Corporations Z-750 turbine system.FACILITY DESCRIPTIONGeneral DescriptionThe DTB is located at NRELs National Wind Technology Center (NWTC) and is dedicated to testingwind turbine drive systems. These are normally comprised of some combination of gears, couplings,bearings, shafts, lubricant systems, gearboxes, generators, controllers, and power conversion systems.The size of the systems that can be tested range from 100 kW to 2.5 MW.The dynamometers prime mover is a variable-torque, variable-speed motor that is controlled by avariable frequency drive system. This 4160 VAC motor is coupled to a 2.5 MW, three-stage epicyclicgearbox. Test articles can be connected to the motor high-speed shaft, the low speed shaft of the gearbox,or an intermediate shaft depending on the speed and torque requirements.TABLE 1 - DYNAMOMETER SHAFT OUTPUT RANGESLocationRatio toMotorMax Speed(rpm)Max Power(kW)Speed at MaxPower (rpm)Low Speed Shaft51.38:143.5250023Intermediate15.4:1146250078High Speed1:1225025001200Table 1 shows the operational capabilities at each shaft. System control logic allows the torque and speedto be continuously varied via manual operator command signals or in an automated manner bySupervisory Control and Data Acquisition (SCADA) command.Figure 1 shows the two operating envelopes, one for each drive shaft option, and a practical wind turbinedesign limit curve that was used to guide the dynamometer design. A representative group of actual windturbine designs (existing or planned) are represented as data points on this figure. These data are plottedbased on their rated output multiplied by a factor of between 1.3 and 1.7. This represents the approximatescaling factors that would be required to accelerate a full lifetime of operation into a few months ofcontinuous operation during an endurance test.The drive motor and gearbox assembly is mounted to a 169-kN (38-ton) elevating drive table that has avertical adjustment span of 3.05-m (10-ft), with tilting capabilities from 0 to 6 degrees. A 1.52-m (5-ft)deep pocket under the drive table allows the top of the table to be lowered 0.61-m (2-ft) in below gradelevel. This range will accommodate shaft elevations of up to 3.05-m (10 ft). Vertical table motion is3Zond Z-75001000200030000.0010.0020.0030.0040.0050.0060.0070.0080.0090.00100.00Dynamometer Output Shaft Speed (RPM)Power (kW)Intermediate ShaftLow Speed ShaftTurbine Endurance TestsPractical Design EnvelopeFIGURE 1 NRELS 2.5 MW DYNAMOMETER OPERATING ENVELOPEactuated using four 222-kN (50-ton) motorized screw jacks that are located at each corner of the table.The screw jacks are operated in front and back pairs with an absolute position accuracy of .25-mm(0.010-in). This fine adjustment capability was built into the system to allow precise alignment betweenthe drive system shafts and the test article shaft. When the proper table position is reached, the table isfastened to the vertical faces of two reaction walls on either side of the drive table. This connection ismade using 10 steel angle brackets that bolt through the drive table and attach to vertical T-slots that aremachined into steel plates mounted to the wall.The table motion is controlled by a programmable logic controller (PLC) that receives exact positioninformation from absolute position encoders located at each jackscrew. This information is translatedinto shaft position and angles by the PLC.The facility is equipped with a 222-kN (50-ton) electric overhead bridge crane that runs the length of the27.4-m (90-ft) test bay. The crane is pendant controlled with multiple speed capabilities.All test articles are mechanically coupled to one of the three shafts in the drive system. Test articles canbe mounted to slotted base plates imbedded in the floor of the south end of the facility or to a grid ofimbedded fasteners north and south of the reaction walls.4FIGURE 2 NRELS 2.5 MW DYNAMOMETER POWER LOOPFigure 2 shows a schematic layout of the power loop for the DTB. Shaft power is typically convertedback to electrical energy by the test article and regenerated back into the NWTC power grid at either 480volts or 4160 volts, where it is reused by the dynamometer. Only the power losses of the drive systemneed to be added locally at the PMH-10 junction outside the facility to keep the system running at steadystate. Figure 3 shows a photo of the DTB.Shaft Bending LoadsMost of the loading experienced by a wind turbine is introduced by wind forces at the mainshaft in theform of shaft torsion and bending. Although numerous dynamic, aerodynamic, and inertial effectscombine to create the various load cases, each of the cases can usually be reduced to the torsional andbending reactions at the mainshaft/hub interface. Shaft torsion alone is often sufficient when testing thegear teeth of a given design. However, if a test of the gear case, bearings, shafts, bedplate, yaw support,or couplings is desired, additional shaft bending loads must be applied. The dynamometer drive producesthe torsional input, but the bending loads must be applied by side-loading the jack-shaft system.In the NREL dynamometer, shaft bending is applied using a servo-hydraulic system, with an actuatorcapable of delivering up to 110 kips. Using the hydraulic actuator system, shaft-bending loads can besynchronized with the rotation of the turbine shaft to give periodic, per-rev shaft loading, representativeof true field conditions.5FIGURE 3 PHOTO OF 2.5 MW DYNAMOMETERRegenerative PowerDuring in-field transient conditions the components of a wind turbine drive system are either acceleratingor decelerating. For a true simulation of the operating wind turbine the inertia and stiffness of thedynamometers rotating components should match the rotor system of the test article. This is not easy todo since these properties are fixed in the dynamometers design. However, inertia matching can beachieved by increasing or decreasing the dynamometer drive power under closed-loop torque control tomimic expected acceleration rates for representative rotor mass properties during rotational speedchanges. Stiffness matching may be done by clever jack-shaft design.Torque Versus Speed ControlThe dynamometer drive can be operated under either torque control or speed control. In speed controlmode, the dynamometer can be operated using a manual throttle control dial, or it can be driven throughthe SCADA system according to a prescribed speed time-history. This mode has been useful for testingthe Z-750 variable speed system. However, speed control may not be as useful when testing a fixed speedinduction or synchronous turbine system, or when inertial effects are critical.6Torque control mode has not been fully implemented yet, but it may prove to be more useful forgenerating a fixed power output or for generating a variable torque load on constant speed generators.Transient load testing will require torque control to balance the rotating inertia. The ultimate goal is tomake the dynamometer behave like the wind, and force the test articles control system to respond in realtime to random turbulent fluctuations. Under closed-loop control, a random wind time history such asSNLWIND3D 5 will be used to compute real-time torque values through a simple Cp versus ? (powercoefficient versus tip-speed-ratio) transfer function, or through a more complex aerodynamics code suchas AeroDyn 6. Turbine responses would be monitored to assess proper controller function.TYPES OF TESTSThe primary advantage of dynamometer testing is that loads to the test turbine can be controlled. Thetypes of testing are divided into wind turbine system tests and component tests. In most cases, one mustinstall a representative drivetrain of the wind turbine to be tested to transmit the loads from thedynamometer. Some of the different types of testing that may be conducted are described below, but theauthors do not presume that this is an exhaustive list.Wind Turbine System TestsDrivetrain Endurance Testing - Endurance testing is done to demonstrate the fatigue life of a particulardrive system to ensure that it can survive its full operating life. Endurance testing will involve 3 to 6months of continuous operation at elevated loads, combined with the application of the significanttransient load cases that are part of long-term operating spectrum. Transverse shaft loads are applied tosimulate actual operating rotor bending loads that can influence the tooth contact stresses, and the fatiguelife of bearings, shafts, and housings. Such testing requires high, steady-state power levels that vary from1.3 to 1.7 times the rated capacity of the test article.Turbulent Wind Simulation Testing - This type of testing can demonstrate the proper operation of theturbines systems under design conditions. Operating the dynamometer under closed-loop torque control,random wind turbulence data are converted into shaft torque by a computer algorithm using feedbackfrom the turbines own control system. Severe stochastic torque conditions can be simulated demandingthat the turbine control system perform at its operational limits under all conditions.Load Event Testing - This type of testing allows measuring the turbine system response under extremeoperating events. Transient torsion and bending loads are applied to the drive train system to simulate thetrue field environment. System inertia can be matched by using the regenerative power system of theelectric variable speed drive. Transverse bending loads can be applied to drive shafts, bearing housings,or yaw systems to recreate specific design load cases. This type of testing also includes self-induced loadsthat arise from the test turbines own components and do not enter through the rotor system. Examples ofself-induced loads are brake system tests, generator failures and grid faults, and normal generatortransients.7Component TestingTesting of individual turbine components is commonly the focus of testing versus the evaluation of awhole system. The dynamometer can be used for qualifying and optimizing various other componentsprior to field installation or retrofit. Tests can be conducted on gears, shafts, housings, bearings,generators, isolated control systems, brakes, or power electronics components, with or without arepresentative drive train assembly.Gearbox Testing - Gearboxes have been the source of widespread field failure throughout the windindustry. It is often necessary to test, evaluate, and optimize the individual effects of operation on thegearbox sub-components to increase life or decrease turbine cost. 7 Some gear tests include: evaluating and mitigating gear tooth loads, evaluating and mitigating tooth wear mechanisms, optimizing gear lubricant properties, optimizing oil system levels, temperatures, and filtration, gear case stress measurements, bearing life testing, and evaluating proper tooth meshing under specific loads to establish optimum lead and profilemodifications.Generators The generator is a critical part of the drivetrain system. It contributes significantly to themechanical loads generated in the entire driveline both in normal operation and when faulted. Thegenerator performance under a range of conditions can be isolated and tested on the dynamometer. Somegenerator tests include: evaluating electrical characteristics and fault settings, evaluating bearing and winding temperatures, measuring torque-speed characteristics, determining breakaway torque and extreme operating performance, determining system efficiencies, and measuring transient response characteristics.Control systems can also be tuned and evaluated during this process. This can be done on high or low-speed (direct-drive) generators. To accommodate direct-drive generators that may have oversized radialdimensions, the dynamometer test bay has a pit built into the test section floor. In these instances, part ofthe generator may be mounted in the pit below the surface of the test bay floor to allow proper heightmatching with the dynamometer output shafts.Z-750 TEST PROGRAMZ-750 BackgroundUnder the cost-shared U.S. Department of Energy (DOE)/NREL sponsored Near-term Research andTesting subcontract, Zond has endeavored to reduce fabrication and installation costs of the Z-750 serieswind turbine through value-engineering methods. To support that activity, NREL has installed a fully8configured Z-750 wind turbine in the DTB to evaluate measures for reducing costs while maintaininghigh quality and the ability to tolerate the various operational and design loads. The components wefocused on in this test program include the gear-case, all gear elements, shafts and couplings, the brake,and, to some extent, the bearings.The testing was broken down into four phases: contact tests, brake event tests, micro-pitting tests, andendurance tests. Each of these is discussed below to describe the goals and the process involved.Tooth Contact TestsContact tests are performed to verify that proper meshing is occurring during elastic loading of the gears.This is done at various torque load levels using a simple gear tooth painting technique that displays thewear pattern. A dynamometer is required to do this properly because it is very difficult to select accurateand repeatable load levels in the field. The results are used in the design, manufacturing, and qualityprocess to optimize the gear geometry and materials. This phase has been successfully completed for theZ-750s current gearset. Figure 4 shows and example of the wear pattern observed by running thedynamometer at a constant load for 1 hour.FIGURE 4 EXAMPLE OF CONTROLLED WEAR GEAR TOOTH CONTACT PATTERN9Brake Event TestsThe Z-750 is equipped with a high-speed, shaft mounted, spring applied, hydraulically released brake toallow for stopping the rotor in the event of pitch system malfunction. Such a redundant system is calledfor in the International Electrotechnical Commission (IEC) standards 8 and the Germanischer-Lloydapproval documents. The second phase of testing reduced the brake application reaction loads throughthe drive train though various mitigation techniques. These techniques were focused on decreasing thebrake application response time while providing for gentler torque rise and drive train reaction. Asdescribed by McNiff et al 3, a reduction in these drive train loads in a demonstrable and repeatablemanner can allow for concomitant reduction in system elements required to properly transfer these loads.This, of course, leads to a lower cost structure. This phase has also been successfully completed after theevaluation of four such methods.Micro-pitting TestsThe micro-pitting studies in Phase 3 is currently in process, and it is, by far, one of t
收藏