Impact of the wake deficit model on wind farm yield: A study of yaw-based control optimization

Abstract

The exploitation of offshore wind resources is considered to have considerable potential in providing carbon-free energy. To increase the economic viability of wind farms, improvement in power generation is sought by mitigating the wake losses. While the industrial standards still favour turbine-level power maximization, the concept of collaborative yaw-based plant-level control has gained significant attention in recent years. The present work investigates the potential of such a wind farm control strategy employing different wake deficit models, for a range of atmospheric conditions and plant layouts. The utilized wake velocity deficit models are the top-hat Jensen model, the Gaussian-shaped Bastankhah model and its novel extension, termed Gauss–Curl Hybrid model, which accounts for secondary steering effects. The yaw control optimization is conducted on a row of eight NREL 5-MW turbines using the FLORIS modelling utility and the SLSQP optimization algorithm. Generally speaking, the Jensen model shows a lack of robustness and is not recommended for yaw control studies. In contrast, the two Gaussian-shaped models are well handled by the optimization algorithm and produce consistent results. More specifically, the Bastankhah model prefers yaw offsets of nearly equal magnitude throughout the whole wind farm except for the most downstream machine that remains aligned with the freestream. On the other hand, the GCH model suggests a large offset at the most upstream turbine, which is gradually reduced at consecutive machines. For the reference wind farm considered, the total power improvement was 3.59% and 14.66% for the Bastankhah and GCH models, respectively.