Wind turbulence input model for horizontal axis wind turbines

Graduation date: 1985In order to estimate wind turbine residL2;characteristics\ud in the presence of atmospheric turbulence, the important atmospheric\ud sources for the force excitations felt by the wind turbine system\ud must be identified and characterized. The goal of this research\ud is to develop the wind turbulence input model for horizontal axis\ud wind turbines.\ud The wind turbulence inputs are determined in three basic modeling\ud steps. First, the turbulent velocity field is assumed to be\ud stationary, locally homogeneous, isotropic, and satisfying Taylor's\ud frozen field hypothesis. The Von Karman model is used to characterize\ud the correlation between velocity components at different spatial\ud positions and at different times. The procedure for finding the\ud aerodynamic forces and moments which are caused by the turbulence,\ud is complicated by the distribution of the turbulent velocity across\ud each of the wind turbine blades. In the second step, the spatial\ud variation of the turbulence is approximated by a few terms in a\ud series expansion. Here, uniform terms across the rotor disk, gradient\ud terms which vary linearly across the disk, and terms which vary\ud quadratically across the disk are retained. All of these terms are\ud time dependent. In the third step, the correlation of these turbulent\ud velocity terms can be closely approximated by low order linear\ud systems forced by white noise.\ud The resulting turbulence input model is then validated using\ud the forward square root information filter and the backward square\ud root information smoother. The residuals between the model outputs\ud and actual field tests are calculated and presented in terms of power\ud spectral density functions