Numerical and experimental analysis of small scale horizontal-axis wind turbine in yawed conditions

Abstract

In wind farms a considerable amount of production losses are due to the presence of wakes: therefore, in order to optimize the efficiency, many control strategies have been conceived and tested recently. Among the most important methods investigated for wind farm control, yawing seems to be one of the most effective. Anyway, this topic has mostly been treated by an energetic point of view: there is still a lack of studies focusing on yaw effects on mechanical and vibrational behavior of wind turbines. On these grounds, this work is devoted to yawed wind turbines and a comparison is conducted between experimental data, collected in wind tunnel tests on a small scale turbine, and aeroelastic numerical models. The experimental setup is composed by a 2 m diameter wind turbine equipped with accelerometers, load cells and tachometers in order to collect vibration, thrust and rotational speed data. With this arrangement, tests have been performed in a closed loop, open chamber wind tunnel, at University of Perugia. The wind turbine has been subjected to steady wind time series and its mechanical behavior has been studied for yaw angles of ±45◦, ±22.5◦ and 0◦ . As concerns simulations, two models are implemented using the FAST software by NREL and an internally developped code based on BEM theory. The simulations are performed with the same wind speed and yaw angles as the experimental wind tunnel tests. The first step of this study is comparing power and thrust coefficients from experimental tests and numerical models. Since numerical codes tend to overestimate the thrust, further studies are conducted about blade deflections, vibrations and tower shadows. The study of experimental thrust depending on the azimuth angle in different yaw configurations revealed that, for vanishing yaw angles, the thrust force has the highest oscillations. This behavior can be ascribed to tower blockage effect and blade deflections. Under this circumstance, an outlook on vibration power spectrum is accomplished. From this it can be showed that 3P (blade passing) frequency, induced by tower blockage, decreases in correspondence of higher yaw angle, demonstrating that tower blockage has a lower effect. In addition, for wind speed lower than 10 m/s at 0◦ yaw, thrust oscillations decrease because of the correspondingly lower blade deflection.