Aerodynamic Performance and Wake Development of Airfoils with Serrated Trailing-Edges

Xiao Liu, Hasan Kamliya Jawahar, Mahdi Azarpeyvand, Raf Theunissen
2017 AIAA Journal  
General rights This document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: A comprehensive experimental study has been performed for symmetric NACA 0012 and cambered NACA 65(12)-10 airfoils with a variety of trailing-edge serrations over a wide range of angles of attack. Results are presented for the aerodynamic force measurements, wake development and energy content of the wake
more » ... t structure at moderate chord based Reynolds numbers (Re c = 2×10 5 −6×10 5 ). The aerodynamic force measurements have shown that the use of trailing-edge serrations for cambered airfoils can lead to significant reductions in the lift coefficient at low angles of attack but does not particularly change the stall characteristics of the airfoils. The wake flow results have shown that the use of serrations can lead to considerable reduction of the wake turbulence intensity, which is shown to be due to a complex interaction between the flow field over the tip and root planes of the serration in the near-wake. Furthermore, power spectral density results pertaining to wake turbulence indicate that the implementation of serrations de-energises the wake flow over a wide range of frequencies. A IRFOIL self-noise and flow interaction noise are major components of the overall noise originating from airframes and jet engines [1] and are also a dominant noise source of modern wind turbines [2]. One of the dominant causes of airfoil self-noise is due to the convection and interaction of the airfoil's turbulent boundary layer with the trailing-edge. Consequently, boundary layer characteristics play an important role in the level of noise generated [3]. The resulting trailing-edge noise is of dipolar nature and its acoustic power varies as a power of the velocity with typical powers ranging between 4 and 5 [4]. Trailing-edge noise is highly significant at low Mach numbers due to the efficient scattering of the turbulent fluctuations over the solid trailingedges [5]. In order to reduce this dominant trailing-edge noise, several passive methods such as serrated trailing-edges [6-8], porous surface treatments [9-11], brushes [12, 13] and morphing trailing-edges [14-16] have been under investigation over the past two decades. In particular, serrated trailing-edges have received considerable interest due to their simple yet efficient noise reduction capabilities amongst all the other passive trailing-edge treatments. Recent research has also shown that implementation of add-on trailing-edge serrations along the outboard blade section can significantly reduce wind turbine aerodynamic noise [17]. Despite the numerous studies directed toward noise reduction capabilities of serrations, the flow behaviour around serration structures and their overall effect on the airfoil's aerodynamic performance have remained largely unexplored. The very first attempt in assessing the effectiveness of serrations for trailing-edge noise purposes, is perhaps that of Howe [18-20]. Howe's analytical investigations have shown that simple modifications such as sawtooth and sinusoidal serrations to trailing-edge can reduce the efficiency of the airfoil's trailing-edge noise generation by introducing destructive sound interference. It was shown that the magnitude of the noise reduction depended on the frequency, length and spanwise spacing of the serration teeth. Howe's mathematical model also showed that optimal attenuation can be achieved with trailing-edge elements having a sharpness larger than 45 • . Despite providing fundamental insight in the aero-acoustic noise generation process, Howe's model nevertheless overestimates the absolute noise reduction levels when compared to experimental data. A much more recent analytical model developed by Lyu et al. [21] has shown that in addition to the de-3 of 28
doi:10.2514/1.j055817 fatcat:zqgcfinxajgfnn7svxgmmk4l6u