Rotationally sampled spectrum approach for simulation of wind speed turbulence in large wind turbines

This paper presents the rotational wind speed turbulence modeling with a view towards a large wind turbine simulation. The dynamic of high power wind systems is analyzed in relation to rotational wind speed turbulence acting on the wind turbine blades. Rotational wind speed turbulence generation is accomplished through a shaping filter. The paper proposes a method for this filter synthesis, using the correlation technique based on von Karman fixed point spectrum model. A “rated” non parametric frequency model of the rotationally sampled turbulence is deduced in conformity with the theoretical support provided by the correlation technique. The model uses as input data the steady state values of wind speed and rotational speed shaft, as well as site properties like turbulence intensity and turbulence length. The parametric model of the shaping filter is obtained through an optimization procedure which deals with the minimization error between the “rated” frequency characteristic of the filter and the parametric frequency characteristic of the shaping filter that must be synthesized. The paper analyzes how the power spectral density changes when the system operating point moves through different operating regions of the power-wind speed characteristic. It provides numerical results to prove the good approximation between the “rated” rotationally sampled spectrum given by the correlation technique, and the rotationally sampled spectrum given by the synthesized shaping filter. Finally, it shows how the rational shaping filter can be used in the numerical simulation of high-power wind energy conversion systems.