Using Wavelet Analysis to Assess Turbulence/Rotor Interactions

Abstract

Large loading events on wind turbine rotor blades are often associated with transient bursts of coherent turbulent energy in the turbine inflow. These coherent turbulent structures are identified as peaks in the three-dimensional, instantaneous, turbulent shearing stress field. Such organized inflow structures and the accompanying rotor aeroelastic responses typically have timescales of only a few seconds and therefore do not lend themselves to analysis by conventional Fourier spectral techniques. Wavelet analysis offers the ability to study more closely the spectral decomposition of short-period events such as the interaction of coherent turbulence with a moving rotor blade. In this paper we discuss our initial progress in the application of wavelet analysis techniques to the decomposition and interpretation of turbulence/rotor interaction. We discuss the results of using both continuous and discrete wavelet transforms for our application. Several examples are given of the techniques applied to both observed turbulence and turbine responses and those generated using numerical simulations. We found that the presence of coherent turbulent structures, as revealed by the inflow Reynolds stress field, is a major contributor to large load excursions. These bursts of coherent turbulent energy induce a broadband aeroelastic response in the turbine rotor as it passes through them.the maximum elevation of the rotor. The vertical stability is expressed by the gradient Richardson number defined aswhere g is the gravitational acceleration, z is the height (m), Â m the layer mean thermodynamic potential temperature (K) given byand T z and p z are the temperature (K) and barometric pressure (hPa) at height z.The covariance u 0 w 0 represents one off-diagonal component of the mean Reynolds stress tensor in turbulent flow. The remaining off-diagonal components include u 0 v 0 and v 0 w 0 , where v 0 is the zero-mean, lateral wind component. In a well-mixed, homogeneous boundary layer over flat terrain, only u 0 w 0 is generally non-zero. However, when turbulence contains motions that are both spatially and temporally organized, i.e. there is a definite phase structure between the velocity components, all three of the off-diagonal Reynolds stress components can be finite. Instantaneously, all three components can take on large positive or negative values. The passage of an organized patch of turbulence can be identified using an anemometer capable of rapid, simultaneous measurement of multi-axis wind components. Coherent turbulent structures can contain large velocity shears over small distances as well as significant local vorticity. Moderate-to high-speed flows that are just slightly or weakly stable (Ri small but positive) often contain small, energetic turbulence structures that, when ingested by wind turbine rotors, increase fatigue damage. 2 In the discussion following, references to the turbulent Reynolds stresses include only the off-diagonal terms u 0 w 0 , u 0 v 0 and v 0 w 0 .The examination of many load histories with...

Synergistic Effects of Turbine Wakes and Atmospheric Stability on Power Production at an Onshore Wind Farm

“…These coherent structures are produced by strong shearing from nocturnal low-level jets and can be damaging to turbine rotor components (Kelley et al 2001, Hand et al 2003. The wind farm studied here does not appear to experience strong "topdowned" turbulence on stable nights as low-level jets are very infrequent.…”

On Space-Time Resolution of Inflow Representations for Wind Turbine Loads Analysis

“…As has been discussed by Kelley [10], time-frequency analysis using continuous wavelet transforms can help study energetic peaks that occur coincidentally with higher-order modes that might not be detected through spectral analysis. Wavelet analysis of the loads data was performed to determine whether cutting off high frequencies in the inflow turbulence would affect turbine load characteristics in any significant way.…”

“…Wavelet analysis of the loads data was performed to determine whether cutting off high frequencies in the inflow turbulence would affect turbine load characteristics in any significant way. Also, non-stationary characteristics of loads from aeroelastic simulations such as flapwise bending loads that can change based on azimuthal position of blades may be lost by relying only on spectral analysis [10]; then, wavelet analysis can again be useful. Figure 3(a,c,e) summarizes results of the wavelet analysis for the FBM process resulting from an unfiltered inflow (13 × 13 grid and 32 Hz sampling) and an inflow filtered at 1 Hz on a 9 × 9 grid (for a hub-height mean wind speed of 12 m/s).…”

Analysis of Wind Speed Measurements using Continuous Wave LIDAR for Wind Turbine Control

“…b In addition, u * has been shown to be a good indicator of the number of high-loading events on a turbine 14,15 and may therefore be a useful value for analyzing the performance of a controller based on wind preview measurements. Results of RMS errors between true and b While the GP LLJ spectra scale with u * , other spectral models such as the IEC models scale with σu.…”

“…Analysis of measurements of coherent structures, such as Kelvin-Helmholtz billows, in the wind fields is needed. Studies have shown that coherent structures cause severe turbine loads, 14 and it is a goal of wind preview based controllers to mitigate the loads caused by coherent structures. Therefore, it is essential to determine the fidelity that LIDAR measurements of coherent structures can achieve.…”