Search

A method of testing the performance of a cluster of wind turbine units an data analysis equations are presented which together form a simple and direct procedure for determining the reduction in energy output caused by the wake of an upwind turbine. This method appears to solve the problems presented by data scatter and wind variability. Test data from the three-unit Mod-2 wind turbine cluster at Goldendale, Washington, are analyzed to illustrate the application of the proposed method. In this sample case the reduction in energy was found to be about 10 percent when the Mod-2 units were separated a distance equal to seven diameters and winds were below rated.

A statistical analysis of data obtained from the Technology and Engineering Information Systems was made. The systems analyzed consist of the following elements: (1) sensors which measure critical parameters (e.g., wind speed and direction, output power, blade loads and component vibrations); (2) remote multiplexing units (RMUs) on each wind turbine which frequency-modulate, multiplex and transmit sensor outputs; (3) on-site instrumentation to record, process and display the sensor output; and (4) statistical analysis of data. Two examples of the capabilities of these systems are presented. The first illustrates the standardized format for application of statistical analysis to each directly measured parameter. The second shows the use of a model to estimate the variability of the rotor thrust loading, which is a derived parameter. Previously announced in STAR as N82-23696

Experimental data, together with supporting analysis, are presented on the power conversion performance and blade loading of large, horizontal-axis wind turbines tested at electric utility sites in the U.S. Four turbine rotor configurations, from 28 to 61 meters in diameter, and data from five test sites are included. Performance data are presented in the form of graphs of power and system efficiency versus free-stream wind speed. Deviations from theoretical performance are analyzed statistically. Power conversion efficiency averaged 0.34 for all tests combined, compared with 0.31 predicted. Round blade tips appeared to improve performance significantly. Cyclic blade loads were normalized to develop load factors which can be used in the design of rotors with rigid hubs.

A method for calculating the 'guaranteed mean' power output of a wind turbine generator is proposed. The term 'mean power' refers to the average power generated at specified wind speeds during short-term tests. Correlation of anemometers, the method of bins for analyzing non-steady data, the PROP Code for predicting turbine power, and statistical analysis of deviations in test data from theory are discussed. Guaranteed mean power density for the Clayton Mod-OA system was found to be 8 watts per square meter less than theoretical power density at all power levels, with a confidence level of 0.999. This amounts to 4 percent of rated power.

A statistical analysis of data obtained from the Technology and Engineering Information Systems was made. The systems analyzed consist of the following elements: (1) sensors which measure critical parameters (e.g., wind speed and direction, output power, blade loads and component vibrations); (2) remote multiplexing units (RMUs) on each wind turbine which frequency-modulate, multiplex and transmit sensor outputs; (3) on-site instrumentation to record, process and display the sensor output; and (4) statistical analysis of data. Two examples of the capabilities of these systems are presented. The first illustrates the standardized format for application of statistical analysis to each directly measured parameter. The second shows the use of a model to estimate the variability of the rotor thrust loading, which is a derived parameter.

Cast nickel-base alloy designated as NASA TAZ-8A has been developed for use in high temperature aircraft engine components. TAZ-8A composition is 8Ta, 6Cr, 6A1, 4Mo, 4W, 2Cb, 0.5Zr, 0.125C, 0.004B, and balance Ni (weight percent). Its specific gravity at room temperature is 8.65.

Nineteen cast nickel-base alloys, five cast cobalt-base alloys, and two wrought nickel-base alloys are included in study. Five nickel-base alloys have directionally-solidified polycrystalline grain structure. Three diffusion coatings and vapor-deposited overlay coating are also included in investigation.

The comparative thermal-stress fatigue resistances of 26 nickeland cobalt-base alloys were determined by fluidized bed tests. Cycles to cracking differed by almost three orders of magnitude for these materials, with directional solidification and surface protection showing definite benefit. The alloy-coating combination with the highest thermal-stress fatigue resistance was directionally solidified NASA TAZ-8A with an RT-SP coating. Its oxidation resistance was also excellent, showing approximately a 1/2 percent weight loss after 14,000 fluidized bed cycles.

Preliminary estimates are presented of vibratory loads and stresses in hingeless and teetering rotors for the proposed NSF-NASA Mod-0 wind power system. Preliminary blade design utilizes a tapered tubular aluminum spar which supports nonstructural aluminum ribs and skin and is joined to the rotor hub by a steel shank tube. Stresses in the shank of the blade are calculated for static, rated, and overload operating conditions. Blade vibrations were limited to the fundamental flapping modes, which were elastic cantilever bending for hingeless rotor blades and rigid-body rotation for teetering rotor blades. The MOSTAB-C computer code was used to calculate aerodynamic and mechanical loads. The teetering rotor has substantial advantages over the hingeless rotor with respect to shank stresses, fatigue life, and tower loading. The hingeless rotor analyzed does not appear to be structurally stable during overloads.

A computer program called TFLIFE is described which can be used to predict the thermal fatigue life of metals and structural components from conventional metal properties. This program is used as a subroutine with a main program supplied by the user. The main program calculates input cycles of temperature and total strain for TFLIFE which then calculates a stress cycle, creep and plastic strain damage, and cyclic life. A unique feature of TFLIFE is that it calculates lives according to several different failure criteria for the same input data. These criteria are surface crack initiation, interior crack initiation, and complete fracture of both unnotched and notched fatigue specimens. Results are presented for two typical problems: thermal-mechanical fatigue of bar specimens of the tantalum alloy T-111 and thermal-stress fatigue of wedge specimens of the nickel alloy B-1900. The computer program is now ready for more extensive evaluation on structural components and additional laboratory specimens.

Thermal fatigue resistances were determined from fluidized bed tests. Cycles to cracking differed by almost three orders of magnitude for these materials with directional solidification and surface protection of definite benefit. The alloy-coating combination with the highest thermal fatigue resistance was directionally solidified NASA TAZ-8A with an RT-XP coating. It oxidation resistance was excellent, showing almost no weight change after 15 000 fluidized bed cycles.

Preliminary estimates of vibratory loads and stresses in hingeless and teetering rotors for the proposed 100-kW wind power system are presented. Stresses in the shank areas of the 19-m (62.5-ft) blades are given for static, rated, and overload conditions. The teetering rotor has substantial advantages over the hingeless rotor with respect to shank stresses, fatigue life, and tower loading. A teetering rotor will probably be required in order to achieve a long service life in a large wind turbine exposed to periodic overload conditions.

During the past five years, the goals of economy and reliability have led to a significant evolution in the basic design--both external and internal--of large wind turbine systems. To show the scope and nature of recent changes in wind turbine designs, development of three types are described: (1) system configuration developments; (2) computer code developments; and (3) blade technology developments.

Approaches to the structural analysis of wind turbine blade designs are reviewed. Specifications and materials data are discussed along with the analysis of vibrations, loads, stresses, and failure modes.

Preliminary test data on output power versus wind speed, rotor blade loads, system dynamic behavior, and start-stop characteristics on the Mod-1 wind turbine generator are presented. These data were analyzed statistically and are compared with design predictions of system performance and loads. To date, the Mod-1 wind turbine generator has produced up to 1.5 MW of power, with a measured power versus wind speed curve which agrees closely with design. Blade loads were measured at wind speeds up to 14 m/s and also during rapid shutdowns. Peak transient loads during the most severe shutdowns are less than the design limit loads. On the inboard blade sections, fatigue loads are approximately equal to the design cyclic loads. On the outboard blade sections, however, measured cyclic loads are significantly larger than design values, but they do not appear to exceed fatigue allowable loads as yet.

In an investigation of windpower plant siting, equations are presented and evaluated for a wind profile model which incorporates both roughness and wind speed effects, while retaining the basic simplicity of the Hellman power law. These equations recognize the statistical nature of wind profiles and are compatible with existing analytical models and recent wind profile data. Predictions of energy output based on the proposed profile equations are 10% to 20% higher than those made with the 1/7 power law. In addition, correlation between calculated and observed blade loads is significantly better at higher wind speeds when the proposed wind profile model is used than when a constant power model is used.

Several large (1500 kW) horizontal rotor configurations were analyzed to determine the effects on dynamic loads of upwind downwind rotor locations, coned and radial blade positions, and tilted and horizontal rotor axis positions. Loads were calculated for a range of wind velocities at three locations in the structure: (1) the blade shank; (2) the hub shaft; and (3) the yaw drive. Blade axis coning and rotor axis tilt were found to have minor effects on loads. However, locating the rotor upwind of the tower significantly reduced loads at all locations analyzed.

The development of computer codes for calculating dynamic loads in horizontal axis wind turbines was examined, and a brief overview of each code was given. The performance of individual codes was compared against two sets of test data measured on a 100 KW Mod-0 wind turbine. All codes are aeroelastic and include loads which are gravitational, inertial and aerodynamic in origin.

Seven computer codes for analyzing performance and loads in large, horizontal-axis wind turbines were used to calculate blade bending moment loads for two operational conditions of the 100 kW Mod-O wind turbine. Results are compared with test data on the basis of cyclic loads, peak loads, and harmonic contents. Four of the seven codes include rotor-tower interaction and three are limited to rotor analysis. With a few exceptions, all calculated loads were within 25% of nominal test data.

Three large (1500 kW) horizontal rotor configurations were analyzed to determine the effects on dynamic loads of upwind and downwind rotor locations, coned and radial blade positions, and tilted and horizontal rotor axis positions. Loads were calculated for a range of wind velocities at three locations in the structure: the blade shank, the hub shaft, and the yaw drive. Blade axis coning and rotor axis tilt were found to have little effect on loads. However, locating the rotor upwind of the tower significantly reduced loads at all locations analyzed.

Dynamic blade loads, including aerodynamic, gravitational, and inertial effects, are presented for two large horizontal-axis wind turbines: the ERDA-NASA 100 kW Mod-0 and 200 kw Mod-0A wind power systems. Calculated and measured loads are compared for an experimental Mod-0 machine in operation. Predicted blade loads are also given for the higher power Mod-0A wind turbine now being assembled for operation as part of a municipal power plant. Two major structural modifications have been made to the Mod-0 wind turbine for the purpose of reducing blade loads. A stairway within the truss tower was removed to reduce the impulsive aerodynamic loading caused by the tower wake on the downwind rotor blades. Also, the torsional stiffness of the yaw drive mechanism connecting the turbine nacelle to the tower was doubled to reduce rotor-tower interaction loads. Measured reductions in load obtained by means of these two modifications equaled or exceeded predictions.

Natural frequencies, as a function of power were determined using a finite element model. Operating conditions investigated were operation with a resistive electrical load and operation synchronized to an electrical utility grid. The influence of certain drive train components on frequencies and mode shapes is shown. An approximate method for obtaining drive train natural frequencies is presented.

The resistance of thin TD-NiCr sheet to cyclic deformation was compared with that of TD-Ni and a conventional nickel-chromium alloy. Strains were determined by a calibration technique which combines room-temperature strain gage and deflection measurements with high-temperature deflection measurements. Analyses of the cyclic tests using measured tensile and creep-rupture data indicated that the TD-NiCr and NiCr alloy specimens failed by a cyclic creep mechanism. The TD-Ni specimens, on the other hand, failed by a fatigue mechanism.

The alloys Nimonic 90, IN 100, coated IN 100, B 1900, coated B 1900, MAR M200, and MAR M200DS (directionally solidified) were studied. Maximum temperatures ranged from 770 C to 1120 C (1420 F to 2050 F). Specimen geometries included tapered disks, double-edged wedges, and cambered airfoils. The disks and wedges were heated and cooled in fluidized beds. The airfoil specimens were heated by a Mach 1 natural gas burner and rapid-air-cooled, with and without spanwise loading. Life calculations included two distinct failure modes: conventional low cycle fatigue and cyclic creep. Required material properties were limited to conventional thermal, tensile, and creep rupture data. The complete life calculation system included the calculation of transient temperature distributions, thermal strains, stresses, creep damage, fatigue damage, and cycles to first crack. Calculated lives were within a factor of two for 76 of the 86 data points analyzed. Cyclic creep accounted for 81% of all the calculated damage.

Simulated turbine blade thermal fatigue testing under transient and steady state heating and spanwise loading, obtaining crack initiation and propagation data

Thermal fatigue tests of high temperature Ni and Co base alloys by fluidized bed technique

The significance of the role that creep can play in governing high-temperature, low-cycle fatigue resistance is investigated by conducting strain cycling tests on two high-temperature stainless steel alloys and making concurrent measurements of stress, temperature, and strain at various frequencies. The results are then analyzed in terms of damage imposed by creep and fatigue components. It is shown that creep can play an important and sometimes dominant role in low cycle fatigue at high temperatures. The results of the study include the findings that: (1) the simple life-fraction theory described is adequate for calculating creep damage when the cyclic creep rupture curve is used as a basis for analysis; (2) a method of universal slopes originally developed for room temperature use is sufficiently accurate at high temperature to be used to calculate pure fatigue damage; and (3) a linear creep-fatigue damage rule can explain the transitions observed from one failure mode to another.

Creep rupture data for predicting thermal fatigue cracking of turbine blades or disks

Cyclic life calculations of materials sujected to isothermal stress considering fatigue and creep

Crack initiation prediction in high temperature components subjected to arbitrary thermal- mechanical cycling

Design considerations for spin stabilized satellites with rigid expandable structures

Stress analysis of elastic-plastic shells of revolution containing discontinuities

Analysis of the common junction of three dissimilar general shells of revolution

Presentation of results for the discontinuity stresses in pressure vessels due to changes in geometry or thickness

Analysis of several typical junctions in space vehicle shell structures

Analysis of several typical junctions in space vehicle shell structures

An overview is presented of the many studies at NASA-Lewis that form the turbine component life prediction program. This program has three phases: (1) development of life prediction methods for major failure modes through materials studies, (2) evaluation and improvement of these methods through a variety of burner rig studies on simulated components in research engines and advanced rigs. These three phases form a cooperative, interdisciplinary program. A bibliography of Lewis publications on fatigue, oxidation and coatings, and turbine engine alloys is included.