## Wind field description

It has long been recognized that wind simulation should be an integral part of wind turbine structural design, and analysis. Now that the structural models become more sophisticated, adequate input to these models is essential in order to make full use of the increased accuracy. The wind input required to receive a license is laid down in design standards (e.g. IEC-1400-1 standard [112], NVN 11400-0 [198] or Germanischer Lloyd [77]). In addition, a realistic wind input is also requisite in order to be able to evaluate the performance of a wind turbine regarding the obtained fatigue load reduction. After all, the fatigue loading depends strongly on the wind characteristic used as input for the simulation [186]. Hence realistic modeling of three-dimensional wind fields is essential in the cost-effective design, and operation of a wind turbine.

From field experiments it is known that the undisturbed wind velocity is variable in space, time and direction, see Fig. 2.5. The rotor blades move at high speed (compared to the wind velocity) through this spatially, and non-uniform wind field. This gives rise to excitation of the rotor with dominant frequencies at integer multiples of the rotational speed [50, 141]. The spatial structure of the atmospheric turbulence is mainly responsible for this so-called "rotational sampling effect". Obviously, this effect is more pronounced for constant speed turbines than for variable speed turbines.

The undisturbed wind velocity is in the current design codes decomposed into a deterministic, and a stochastic (i.e. turbulent) wind description. Deterministic wind inputs are very useful to study and verify the global wind turbine behavior, while stochastic wind inputs are to be used for the prediction of the loads which a wind turbine will experience during its life-time. The state-of-the-art with regard to the deterministic part is a representation that includes a linear (or more accurately bi-linear), Power Law (or exponential) or logarithmic model of wind shear together with two, almost identical representations of the tower shadow effect, viz. the (1-cos) and the potential flow (or dipole) model.

A model of the turbulent wind field suitable for loading calculations requires good representation of both the temporal and spatial structure of turbulence [229]. Calculations based on a turbulence simulation which assumes a fully coherent cross-wind spatial structure will not take into account the crucial important "eddy slicing" transfer of rotor load from low frequencies to those associated with rotational speed and its higher harmonics. This "eddy slicing" or "rotational sampling effect", associated with rotating blades slicing through the turbulent structure of the wind, is a significant source of fatigue loading. Although early models concentrated on representation of the longitudinal component of turbulence only, the state-of-the-art is to base load calculations on a model of all three turbulent velocity components of the wind field [228].

The most "correct" method to simulate such a turbulent wind field would probably be to solve the Navier-Stokes equations of an atmospheric flow bounded from below by an aerodynamically rough surface directly by Direct Numerical Simulation (DNS) [174]. However, the computational cost of this would be enormous. A cheaper way to do it would be to use Large Eddy Simulation (LES), which is an approximate solution to the Navier-Stokes equations where the motions of the smallest scales are not solved directly, but modelled. Still, this requires supercomputers, and is usually not justified for practical engineering use.

Therefore, in wind engineering, empirical information is generally used in the methods developed for the simulation of turbulent wind. All these methods use, as a starting point, auto-spectral and coherence descriptions of the turbulence. There are several of such descriptions available with the most common being the Von Karman and Kaimal spectral models. A more comprehensive method is that due to Mann [173, 174] developed at Ris0 National Laboratory, Denmark. This method is in principle a special case of the general method of Shinozuka and Jan [263]. Another method is the Veers turbulence simulation method, often referred to as SANDIA or SNLWIND method [132] which is originally proposed by Veers, and is again based on the method of Shinozuka and Jan [263].

All these descriptions are encoded in so-called stochastic 3D wind field generators. The following stochastic 3D wind field generators are commonly used: EWS [145, 168], Mann [174], Shinozuka/Jan [263], SNLWIND-3D [132], Sosisw [69, 70], SWIFT (Simulation Wind Field in Time) [315], SWING-4 (Stochastic WINd Generator) [11, 13], Veers [296] and WIND3D [226]. In Table 2.1, and Table 2.2 on page 40, and 41 respectively, it is summarized which wind generator is used in which wind turbine design code.

## Renewable Energy 101

Renewable energy is energy that is generated from sunlight, rain, tides, geothermal heat and wind. These sources are naturally and constantly replenished, which is why they are deemed as renewable. The usage of renewable energy sources is very important when considering the sustainability of the existing energy usage of the world. While there is currently an abundance of non-renewable energy sources, such as nuclear fuels, these energy sources are depleting. In addition to being a non-renewable supply, the non-renewable energy sources release emissions into the air, which has an adverse effect on the environment.

Get My Free Ebook