Calculating the sensitivity of wind turbine loads to wind inputs using response surfaces

This paper presents a methodology to calculate wind turbine load sensitivities to turbulence parameters through the use of response surfaces. A response surface is a highdimensional polynomial surface that can be calibrated to any set of input/output data and then used to generate synthetic data at a low computational cost. Sobol sensitivity indices (SIs) can then be calculated with relative ease using the calibrated response surface. The proposed methodology is demonstrated by calculating the total sensitivity of the maximum blade root bending moment of the WindPACT 5 MW reference model to four turbulence input parameters: a reference mean wind speed, a reference turbulence intensity, the Kaimal length scale, and a novel parameter reflecting the nonstationarity present in the inflow turbulence. The input/output data used to calibrate the response surface were generated for a previous project. The fit of the calibrated response surface is evaluated in terms of error between the model and the training data and in terms of the convergence. The Sobol SIs are calculated using the calibrated response surface, and the convergence is examined. The Sobol SIs reveal that, of the four turbulence parameters examined in this paper, the variance caused by the Kaimal length scale and nonstationarity parameter are negligible. Thus, the findings in this paper represent the first systematic evidence that stochastic wind turbine load response statistics can be modeled purely by mean wind wind speed and turbulence intensity.

[1]  Lance Manuel,et al.  On the propagation of uncertainty in inflow turbulence to wind turbine loads , 2008 .

[2]  Dayton A. Griffin,et al.  WindPACT Turbine Design Scaling Studies Technical Area 1-Composite Blades for 80- to 120-Meter Rotor , 2001 .

[3]  G.J.W. Van Bussel,et al.  Influence of atmospheric stability on wind turbine loads , 2013 .

[4]  I. Sobola,et al.  Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates , 2001 .

[5]  Anand Natarajan,et al.  Model of wind shear conditional on turbulence and its impact on wind turbine loads , 2015 .

[6]  J. Peinke,et al.  The Science of Making Torque from Wind (TORQUE 2016) , 2014 .

[7]  Jennifer M. Rinker An Empirically Based Stochastic Turbulence Simulator with Temporal Coherence for Wind Energy Applications , 2016 .

[8]  B. Iooss,et al.  Global sensitivity analysis for a numerical model of radionuclide migration from the RRC “Kurchatov Institute” radwaste disposal site , 2008 .

[9]  J. L. Tangler,et al.  NREL airfoil families for HAWTs , 1995 .

[10]  D. J. Malcolm,et al.  WindPACT Turbine Rotor Design Study , 2006 .

[11]  Nilay Shah,et al.  The identification of model effective dimensions using global sensitivity analysis , 2011, Reliab. Eng. Syst. Saf..

[12]  Neil D. Kelley,et al.  Turbulence-Turbine Interaction: The Basis for the Development of the TurbSim Stochastic Simulator , 2011 .

[13]  Herbert J. Sutherland Analysis of the structural and inflow data from the list turbine , 2002 .

[14]  Herbert J. Sutherland,et al.  Statistical Analysis of Wind Turbine Inflow and Structural Response Data from the LIST Program , 2003 .

[15]  Bertrand Iooss,et al.  Response surfaces and sensitivity analyses for an environmental model of dose calculations , 2006, Reliab. Eng. Syst. Saf..

[16]  A. Saltelli,et al.  Making best use of model evaluations to compute sensitivity indices , 2002 .

[17]  Henri P. Gavin,et al.  High-order limit state functions in the response surface method for structural reliability analysis , 2008 .

[18]  H. F. Veldkamp,et al.  Chances in wind energy: A probalistic approach to wind turbine fatigue design , 2006 .

[19]  R.A.S. Ratchliffe,et al.  Book reviewBoundary-layer meteorology: A new quarterly journal of physical and biological processes in the atmospheric boundary layer. R. E. Munn (Editor). D. Reidel Publishing Company, Dordrecht, 1969, Volume 1(1). , 1971 .

[20]  J. Peinke,et al.  Atmospheric turbulence and its influence on the alternating loads on wind turbines , 2011 .

[21]  Michael McAleer,et al.  Mathematics and Computers in Simulation: Preface , 2011 .