Determination of local flow conditions and preliminary investigation on the validity of the PNoise code for large-size wind turbine

Abstract

This paper summarizes the determination of more realistic local Reynolds and Mach flow numbers at the blades of a large-diameter, horizontal-axis wind turbine designed specifically for this purpose with the aid of the blade element momentum method, and the subsequent effort to extend the validation range of the modified-BPM airfoil trailing edge noise method against a single high Reynolds number acoustic dataset available. The validation extension effort proved unfruitful, and the reasons are discussed in detail in relation to the dataset employed. After coupling the BPM model to a hybrid boundary layer solver, the resultant modified method, called PNoise, was embedded into the TU Berlin (HFI) wind turbine design, open code environment, QBlade, available under General Public License. During the time period past since the release of the first integrated version (v0.95), the open-source code has been downloaded by users more than 20,000 times, prompting the trial for further validation of the proposed method. Although the validation extension is not sanctioned based on this preliminary investigation, the calculation of more realistic flow conditions over the blades of current utility-size wind turbines could be helpful to other researchers. Also, this effort highlighted the significant uncertainties associated with methods employed to obtain acoustic spectra from one specific aeroacoustic wind tunnel, especially when the acoustic signal obtained experimentally is later subjected to transforming algorithms. The findings also stress the fact that more reliable experimental data is needed under high Reynolds numbers (notice: all Reynolds numbers described in the text are local, chord-based Reynolds numbers.) in order to support TE and other airfoil self-noise model development and validation.