LowNoise: Linear Stability and Resolvent Analysis for Prediction and Mitigation of Wind Turbine Trailing-edge Noise

• Applicants: Professor Dr.-Ing. Kilian Oberleithner; Professor Dr.-Ing. Ennes Sarradj
• Subject Area: Fluid Mechanics ; Acoustics ; Hydraulic and Turbo Engines and Piston Engines
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 458062719

The mitigation of flow noise from wind turbines is of central importance for current and future, even more powerful wind turbines. The most important noise source is the trailing-edge noise. It is caused by pressure fluctuations in the boundary layer at the trailing edge of the rotor blade. Frequency characteristics and sound levels depend largely on the coherent eddy-like structures in the boundary layer. Current methods to reduce trailing-edge noise show contradictory results, as their influence on the coherent structures is neither sufficiently understood nor reliably modelled.Current investigations on free shear flows and boundary layer flows clearly show that linear stability and resolvent analysis can be used to systematically describe the formation and control of coherent structures. Therefore, in the LowNoise project, these very successful and novel methods are applied to the flow field of a typical wind turbine airfoil. The central goal is a physical low-dimensional model which describes the essential mechanisms of trailing-edge noise.The modelling is done in several steps. The coherent structures are approximated as linear modes of the mean-flow field, which can be determined by the stability or resolvent analysis. The pressure fluctuations on the wing surface are then determined by data assimilation from the linear modes. The trailing-edge noise, then, results from the integral of the pressure field on the wing surface.The central innovation of this approach is the modelling of the sound sources using linear stability and resolvent theory. On the one hand, this enables the quantitative determination of the trailing-edge noise at significantly fewer empirical input variables than with current low-dimensional models. On the other hand, the model provides the causal mechanisms of the formation of the coherent structures, and thus, of the trailing-edge noise, which allows to optimise existing control measures and to develop new more effective control measures.The LowNoise project starts with large-eddy simulations of the entire flow field. From the flow fields the velocity and pressure fluctuations of the coherent structures can be extracted and an empirical low-dimensional model can be derived. Based on the mean-flow fields of the simulation the stability and resolvent theory model will be developed. The validation of the simulations and the model is done experimentally by means of pressure and acoustic measurements in a wind tunnel. The project ends with a model-based analysis of the control influences of the trailing-edge noise.The models developed in LowNoise will show the essential physical relationships for the generation and effective reduction of trailing-edge noise of wind turbines. Furthermore, the model and concepts developed in LowNoise can be applied to a variety of other wall-bounded flows.

• DFG Programme: Research Grants
• International Connection: Brazil, France
• Cooperation Partners: Professor Dr. André Cavalieri; Professor Lutz Lesshafft, Ph.D.