Paul S. Veers

To Capture the Wind

From the birth of modern electricity-generating wind turbines in the late 1970s to now, wind energy technology has dramatically improved. Capital costs have plummeted, reliability has improved, and efficiency has increased. High-quality turbine manufacturers exist around the world, and wind plants of 300 MW and larger are being integrated into the electrical grid to exacting utility specifications. These modern wind plants are now routinely produced by multinational manufacturing companies at a cost of energy approaching, and in some cases below, that of fossil-fuel generating plants. At the end of 2006, the total U.S. wind energy capacity had grown to 11,603 MW, or enough to provide the electrical energy needs of more than 2.9 million American homes. Wind capacity in the United States and in Europe has grown at a rate of 20% to 30% per year over the past decade. Despite this rapid growth, wind currently provides less than 1% of total electricity consumption in the United States. The vision of the wind industry in the United States and in Europe is to increase winds fraction of the electrical energy mix to more than 20% within the next two decades.

AEROELASTIC BEHAVIOR OF TWIST-COUPLED HAWT BLADES

As the technology for horizontal axis wind turbines (HAWT) development matures, more novel techniques are required for the capture of additional amounts of energy, alleviation of loads and control of the rotor. One such technique employs the use of an adaptive blade that could sense the wind velocity or rotational speed in some fashion and accordingly modify its aerodynamic configuration to meet a desired objective. This could be achieved in either an active or passive manner, although the passive approach is much more attractive due to its simplicity and economy. As an example, a blade design might employ coupling between bending and/or extension, and twisting so that, as it bends and extends due to the action of the aerodynamic and inertial loads, it also twists modifying the aerodynamic performance in some way. These performance modifications also have associated aeroelastic effects, including effects on aeroelastic instability. To address the scope and magnitude of these effects a tool has been developed for investigating classical flutter and divergence of HAWT blades. As a starting point, an adaptive version of the uniform Combined Experiment Blade will be investigated. Flutter and divergence airspeeds will be reported as a function of the strength of the coupling and also be compared to those of generic blade counterparts.

The Use of Twist-Coupled Blades to Enhance the Performance of Horizontal Axis Wind Turbines

This paper reviews issues related to the use of aeroelastic tailoring as a cost-effective, passive means to shape the power curve and reduce loads. Wind turbine blades bend and twist during operation, effectively altering the angle of attack, which in turn affects loads and energy production. It is possible to build a small amount of desirable twisting into the load response of a blade with proper asymmetric fiber lay up in the blade skin. The tailored twisting can create an aeroelastic effect that has payoff in either better power production or in vibration alleviation, or both. Several research efforts have addressed different parts of this issue. Research and development in the use of aeroelastic tailoring on helicopter rotors is reviewed. Potential energy gains as a function of twist coupling are reviewed. The effects of such coupling on rotor stability have been studied and are presented. Fatigue damage estimates due to turbulent inflow have been computed for rotors employing several different control schemes, with and without twist-coupled blades. Energy otput and maximum loads are also computed and compared.

PARAMETRIC MODELS FOR ESTIMATING WIND TURBINE FATIGUE LOADS FOR DESIGN

International standards for wind turbine certification depend on finding long-term fatigue load distributions that are conservative with respect to the state of knowledge for a given system. Statistical models of loads for fatigue application are described and demonstrated using flap and edge blade-bending data from a commercial turbine in complex terrain. Distributions of rainflow-counted range data for each ten-minute segment are characterized by parameters related to their first three statistical moments (mean, coefficient of variation, and skewness). Quadratic Weibull distribution functions based on these three moments are shown to match the measured load distributions if the non-damaging low-amplitude ranges are first eliminated. The moments are mapped to the wind conditions with a two-dimensional regression over ten-minute average wind speed and turbulence intensity. With this mapping, the short-term distribution of ranges is known for any combination of average wind speed and turbulence intensity. The longterm distribution of ranges is determined by integrating over the annual distribution of input conditions. First, we study long-term loads derived by integration over wind speed distribution alone, using standard-specified turbulence levels. Next, we perform this integration over both wind speed and turbulence distribution for the example site. Results are compared between standarddriven and site-driven load estimates. Finally, using statistics based on the regression of the statistical moments over the input conditions, the uncertainty (due to the limited data set) in the long-term load distribution is represented by 95% confidence bounds on predicted loads.

Application of Measured Loads to Wind Turbine Fatigue and Reliability Analysis

Cyclic loadings produce progressive damage that can ultimately result in wind turbine structural failure. There are many issues that must be dealt with in turning load measurements into estimates of component fatigue life. This paper deals with how the measured loads can be analyzed and processed to meet the needs of both fatigue life calculations and reliability estimates. It is recommended that moments of the distribution of rainflow-range load amplitudes be calculated and used to characterize the fatigue loading. These moments reflect successively more detailed physical characteristics of the loading (mean, spread, tail behavior). Moments can be calculated from data samples and functional forms can be fitted to wind conditions, such as wind speed and turbulence intensity, with standard regression techniques. Distributions of load amplitudes that accurately reflect the damaging potential of the loadings can be estimated from the moments at any wind condition of interest. Fatigue life can then be calculated from the estimated load distributions, and the overall, long-term, or design spectrum can be generated for any particular wind-speed distribution. Characterizing the uncertainty in the distribution of cyclic loads is facilitated by using a small set of descriptive statistics for which uncertainties can be estimated. The effects of loading parameter uncertainty can then be transferred to the fatigue life estimate and compared with other uncertainties, such as material durability.

Damping measurements using operational data

The authors have measured modal damping using strain-gauge data from an operating wind turbine. This new technique for measuring modal damping is easier and less expensive than previously used methods. Auto-correlation and cross-correlation functions of the strain-gauge data have been shown to consist of decaying sinusoids which correspond to the modal frequencies and damping ratios of the wind turbine. The authors have verified the method by extracting damping values from an analytically generated data set. Actual operating response data from the DOE/Sandia 34-m Test Bed has been used to calculate modal damping ratios as a function of rotor rotation rate. This capability will allow more accurate fatigue life prediction and control.

A Comparison of Standard Coherence Models for Inflow Turbulence With Estimates from Field Measurements

The Long-term Inflow and Structural Test (LIST) program, managed by Sandia National Laboratories, Albuquerque, NM, is gathering inflow and structural response data on a modified version of the Micon 65/13 wind turbine at a site near Bushland, Texas. With the objective of establishing correlations between structural response and inflow, previous studies have employed regression and other dependency analyses to attempt to relate loads to various inflow parameters. With these inflow parameters that may be thought of as single-point-in-space statistics that ignore the spatial nature of the inflow, no significant correlation was identified between load levels and any single inflow parameter or even any set of such parameters, beyond the mean and standard deviation of the hubheight horizontal wind speed. Accordingly, hence, we examine spatial statistics in the measured inflow of the LIST turbine by estimating the coherence for the three turbulence components (along-wind, across-wind, and vertical). We examine coherence spectra for both lateral and vertical separations and use the available ten-minute time series of the three components at several locations. The data obtained from spatial arrays on three main rowers located upwind from the test turbine as well as on two additional towers on either side of the main towers consist of 291 ten-minute records. Details regarding estimation of the coherence functions from limited data are discussed. Comparisons with standard coherence models available in the literature and provided in the International Electrotechnical Commission (IEC) guidelines are also discussed. It is found that the Davenport exponential coherence model may not be appropriate especially for modeling the coherence of the vertical turbulence component since it fails to account for reductions in coherence at low frequencies and over large separations. Results also show that the Mann uniform shear turbulence model predicts coherence spectra for all turbulence components and for different lateral separations better than the isotropic von Karman model. Finally, on studying the cross-coherence among pairs of turbulence components based on field data, it is found that the coherence observed between along-wind and vertical turbulence components is not predicted by the isotropic von Karman model while the Mann model appears to overestimate this cross-coherence.

Performance of twist-coupled blades on variable speed rotors

The load mitigation and energy capture characteristics of twist-coupled HAWT blades that are mounted on a variable speed rotor are investigated in this paper. These blades are designed to twist toward feather as they bend with pretwist set to achieve a desirable twist distribution at rated power. For this investigation, the ADAMS-WT software has been modified to include blade models with bending-twist coupling. Using twist-coupled and uncoupled models, the ADAMS software is exercised for steady wind environments to generate C{sub p} curves at a number of operating speeds to compare the efficiencies of the two models. The ADAMS software is also used to generate the response of a twist-coupled variable speed rotor to a spectrum of stochastic wind time series. This spectrum contains time series with two mean wind speeds at two turbulence levels. Power control is achieved by imposing a reactive torque on the low speed shaft proportional to the RPM squared with the coefficient specified so that the rotor operates at peak efficiency in the linear aerodynamic range, and by limiting the maximum RPM to take advantage of the stall controlled nature of the rotor. Fatigue calculations are done for the generated load histories using a range of material exponents that represent materials from welded steel to aluminum to composites, and results are compared with the damage computed for the rotor without twist-coupling. Results indicate that significant reductions in damage are achieved across the spectrum of applied wind loading without any degradation in power production.

Using Environmental Contours to Predict Extreme Events on Wind Turbines

This paper presents and explores estimating design loads on wind turbines using the environmental contour method. Contours promise to provide both practical reliability estimation and valuable information about the combination of joint environmental variable values, e.g. wind speed and turbulence, most critical to each specific wind turbine. We present the background of the development of environmental contours as applied to wind energy systems, and apply this theory, in three examples, to develop contours based either (1) on design code description of environmental conditions, or (2) on measured data for a site-specific application. The site-specific case is used for both stall and pitch controlled turbine examples. From these contours, and a functional description of the short-term response of the turbine, implicit FORM estimates are made for the turbine response; these estimates are compared with results obtained from numerical integration of the short-term response of the turbine over the joint distribution of wind speed and turbulence. We find that the environmental contour method provides reasonable estimates of the expected extreme load, compared with the full integration method.

The development of confidence limits for fatigue strength data

Over the past several years, extensive databases have been developed for the S-N behavior of various materials used in wind turbine blades, primarily fiberglass composites. These data are typically presented both in their raw form and curve fit to define their average properties. For design, confidence limits must be placed on these descriptions. In particular, most designs call for the 95/95 design values; namely, with a 95% level of confidence, the designer is assured that 95% of the material will meet or exceed the design value. For such material properties as the ultimate strength, the procedures for estimating its value at a particular confidence level is well defined if the measured values follow a normal or a log-normal distribution. Namely, based upon the number of sample points and their standard deviation, a commonly-found table may be used to determine the survival percentage at a particular confidence level with respect to its mean value. The same is true for fatigue data at a constant stress level (the number of cycles to failure N at stress level S{sub 1}). However, when the stress level is allowed to vary, as with a typical S-N fatigue curve, the procedures for determining confidence limits are not as well defined. This paper outlines techniques for determining confidence limits of fatigue data. Different approaches to estimating the 95/95 level are compared. Data from the MSU/DOE and the FACT fatigue databases are used to illustrate typical results.