The current trend in the wind power industry is to develop wind turbines of constantly increasing size and rated power, as well as wind farms of growing size and installed wind power. A careful study of the behaviour of the wind turbines during their operation is of crucial importance in the planning phase and in the design stage of a wind farm, in order to minimize the risks deriving from a non accurate prediction of their impact in the electric grid causing sensible faults of the system. To analyze the impact of the wind turbines in the system, motivates the development of accurate yet simple models. To be able to practically deal with this topics, a simple model of a wind turbine system is investigated and developed; it has the aim to describe the behaviour of a wind turbine in operation on a mechanical standpoint. The same reduced order simple model can also be employed for control system studies; the control system model that can’t be used in generation, can use the reduced model. Together with the analytical description of such model, is realized a MatLab code to numerically analyse the response of the system, and the results of the simulation through such code are presented. The objective of this thesis has been to provide a simple benchmark tool in MatLab for grid integration and control studies for interested researchers.

The increased fatigue loads and decreased power output of a wind turbine placed in the wake of another turbine is a well-known problem when building new wind-power farms and a subject of intensive research. These problems are caused by the velocity gradients and high turbulence levels present in the wake of a turbine. In order to better estimate the total power output and life time of a wind-power farm, knowledge about the development and stability of wind-turbine wakes is crucial. In the present thesis, the flow field around small-scale model wind turbines has been investigated experimentally in two wind tunnels. The flow velocity was measured with both hot-wire anemometry and particle image velocimetry. To monitor the turbine performance, the rotational frequency, the power output and the total drag force on the turbine were also measured. The power and thrust coefficients for different tip-speed ratios were calculated and compared to the blade element momentum method, with a reasonable agreement. The same method was also used to design and manufacture new turbine blades, which gave an estimate of the distribution of the lift and drag forces along the blades. The influence of the inlet conditions on the turbine and the wake properties was studied by subjecting the turbine to both uniform in flow and different types of boundary layer in flows. In order to study the stability and development of the tip vortices shed from the turbine blades, a new experimental setup for phase-locked measurements was constructed. The setup made it possible to introduce perturbations of different frequencies and amplitudes, located in the rear part of the nacelle. With a newly developed method, it was possible to characterize the vortices and follow their development downstream, using only the streamwise velocity component. Measurements were also performed on porous discs placed in different configurations. The results highlighted the importance of turbine spacings. Both the measurements on the turbine and the discs were also used to compare with large eddy simulations using the actuator disc method. The simulations managed to predict the mean velocity fairly well in both cases, while larger discrepancies were seen in the turbulence intensity. / <p>QC 20140424</p>

A major portion of this thesis is devoted to the Whisper 200 wind turbine model in Matlab-Simulink environment. The turbine is installed at the Department of Electrical Power Engineering, FEEC BUT. In the model, several types of simulations can be executed. On that basis, the power curve and mathematical relationships between wind speed and other physical quantities, such as RPM, electic current, and voltage, were obtained. Comparisons of the simulations' results with measurement data illustrate adequate agreement, but limitations of the model remain significant, as is exhaustively documented and commented upon in the thesis. As a partial advancement towards elimination of the model's deficiencies, conditions for substantial performance improvements of the dynamic simulation have been elaborately derived.