Dr Gerard Schepers
- Wind Energy Technology
However, although 10MW+ turbines are seen as a way to reduce the overall ‘Cost of Energy’ for off-shore wind power, the design of the required very large rotor blades fell outside the validated range of current state-of-the-art aerodynamic and aero-elastic tools in various aspects. Firstly, very large blades operating at high tip speeds mean high Reynolds and Mach numbers for which the effects are neither certain nor adequately validated. Secondly, thick(er) airfoils need to be assessed in terms of aerodynamic performance while increased flexibility will lead to larger deflections and more pronounced non-linear aero-elastic behaviour with unknown aerodynamic implications, etc. All this is further complicated by the desired implementation of active and/or passive flow devices.
As a result, a wide variety of aerodynamic models was considered, ranging from low-complexity/computationally efficient models (i.e. Blade Element Momentum - BEM) to high-fidelity/computationally demanding models (e.g. Computational Fluid Dynamics - CFD), including intermediate models (e.g. free vortex wake models-FVW). This enabled an improvement of the fast, low-complexity tools by calibrating in terms of results from high-fidelity models. The model assessment was carried out on two 10 MW reference wind turbines (RWT’s), one originating from the INNWIND.EU project, and another one designed in AVATAR.
The improvement and validation of models was also based on suitable experimental data, mainly wind tunnel measurements and a selected number of field measurements such as wind tunnel measurements on a DU 00-W-212 airfoil up to a Reynolds number of 15 million in a pressurised tunnel. Wind tunnel measurements on flow devices (e.g. vortex generators and flaps) were also performed.
AVATAR resulted in a long list of model improvements and lessons learned on the use of models along with several recommendations, one of the most important being that more validation material databases are needed, preferably experimental but also of results from high/intermediate confidence codes which can serve as validation material for low-fidelity codes. More information can be found on http://www.eera-avatar.eu/
The AVATAR consortium consisted of: