The role of free stream turbulence with large integral scale on the aerodynamic performance of an experimental low Reynolds number S809 wind turbine blade

https://doi.org/10.1016/j.jweia.2015.03.010Get rights and content

Abstract

Effects of free stream turbulence with large integral scale on the aerodynamic performance of an S809 airfoil-based wind turbine blade at low Reynolds number are studied using wind tunnel experiments. The case of wind turbine blades subjected to turbulence structures with large integral scale is of interest since in the atmospheric boundary layer very large-scale structures interact with the blades. A constant chord (2-D) S809 airfoil wind turbine blade model with an operating Reynolds number of 2.08×105 based on chord length was tested for a range of angles of attack representative of fully attached and stalled flow as encountered in typical wind turbine operation. The smooth-surface blade was subjected to a quasi-laminar free stream with very low free-stream turbulence as well as to elevated free-stream turbulence generated by an active grid. This turbulence contained large-scale eddies with levels of free-stream turbulence intensity of up to 6.14% and an integral length scale of about 60% of chord-length. The pressure distribution was acquired using static pressure taps and the lift was subsequently computed by numerical integration. The wake velocity deficit was measured utilizing hot-wire anemometry to compute the drag coefficient also via integration. In addition, the mean flow was quantified using 2-D particle image velocimetry (PIV) over the suction surface of the blade. Results indicate that turbulence, even with very large-scale eddies comparable in size to the chord-length, significantly improves the aerodynamic performance of the blade by increasing the lift coefficient and overall lift-to-drag ratio, L/D for all angles tested except 0°.

Introduction

Wind turbines operate in a highly turbulent atmospheric boundary layer, which gives rise to a spectrum of scales that impact wind turbines in terms of external blade loads and acoustic noise. These flow phenomena have been proven to directly impact the operating life of the blade and other components such as the gearbox and generator. Most wind tunnel-based aerodynamic studies of airfoil and wind turbine blades are performed under laminar inflow conditions or with free-stream turbulence whose integral scale is small compared to the blade chord length. It is thus of special interest to determine aerodynamic performance of a wind turbine blade specifically under the influence of free-stream turbulence that contains large-scale structures. It is important to mention that constraints placed on the experiments by the wind tunnel test section size, maximum speed, and use of an active grid (which provides some flow resistance or effective “blockage” effect on the test section inlet) limits the magnitude of the Reynolds number that can be produced. The relative low Reynolds number flow (based on chord) of 2.08×105 utilized in this study compared to industrial multi-megaWatt wind turbines producing Reynolds numbers of one to several million limits the significance of the results to smaller scale commercial and residential wind turbines. Nevertheless, from a theoretical perspective on subjecting otherwise laminar-transitional flow over a wind turbine blade with large-scale turbulence in the freestream, it is a worthwhile investigation.

The present study is thus performed to quantify the aerodynamic performance of a low Reynolds number blade subject to free-stream turbulence with large integral scale. By measuring the blade surface pressure and wake velocity deficit at various angles of attack, we can compute the lift and drag characteristics of the blade under the presence of free-stream turbulence.

The effect of turbulence on the performance of wind turbines depends on the intensity of the turbulence as well as the size of turbulence scales. Turbulence levels in the atmospheric boundary layer have been quoted between 5 and 25% depending on the topography and daily cycle (Hojstrup, 1999, Hand, 2003). Furthermore, there have been wind tunnel studies to replicate these conditions for wind turbine blades with flow turbulence levels in the range from 1.1% to 16% utilizing passive grids for fixed and rotating blades/rotor (Sicot et al., 2006, Amandolèse and Széchényi, 2004, Sicot et al., 2008). These types of grids however do not have the ability to create high Reynolds number turbulence based on Taylor micro scales compared to active grids. Thus, utilizing an active grid it is possible to study the role of large scale turbulence (on the order of the chord length) on the performance of the wind turbine blade.

In the current study, an active grid as that described in the experiments of Brzek et al. (2009) and Cal et al. (2010) is used to produce freestream turbulence leading to a turbulence intensity of about 6.14% at the location of the blade. The turbulence intensity is defined according to the ratio of the average root mean square (r.m.s.) turbulence of all three velocity components to the mean freestream velocity, U. Since the turbulence component in the spanwise direction, w was not obtained, it was accounted for by increasing the magnitude of the streamwise, u2 and wall-normal, v2 fluctuations by 50% and dividing by three. This is a valid assumption, since the turbulence generated with the active grid is considered isotropic, i.e. it is uniform in all directionsTu=13(1.5u2+1.5v2)U

The integral length scale can be evaluated according to the formula proposed by Mydlarski and Warhaft (1996) according toL0.9u3εwhere u is the r.m.s. free-stream velocity and ε is the dissipation rate. The r.m.s. velocity and dissipation rate for the current experimental conditions were previously calculated using the third-order structure function as detailed by the study of Kang (2003) and more recently Brzek et al. (2009). The turbulence intensity produced an integral length scale in the flow of about 0.15 m, which is of the same order of magnitude as the chord of the blade of 0.25 m.

One of the main goals of this study is to understand the relationship between the turbulent length scales, the aerodynamic forces (specifically lift and drag) and the development of the mean flow. The paper is organized as follows: Section 2 describes the experimental setup and methods; which consists of the wind tunnel facility with its unique active grid to generate turbulent inflow conditions, and the hot-wire anemometry and particle image velocimetry (PIV) techniques utilized to measure the wake velocity and mean flow over the blade. Section 3 discusses the results, comprised of the pressure coefficient distributions, wake velocity deficit measurements, and the numerical integration method employed to compute the aerodynamic coefficients. Finally, Section 4 is devoted to conclusions gleaned from the experimental results.

Section snippets

Wind tunnel facility

The experiments of the current investigation were performed at the Corrsin Wind Tunnel Facility at The Johns Hopkins University; a schematic of the facility is shown on Fig. 1. The wind tunnel is a two-story closed-return type wind tunnel with a 25:1 primary and a 1:1.27 secondary contraction. The test section contains a 1.22 m width, 0.91 m height, and a 10 m length where the background turbulence intensity is less than 0.1% at the end of the test section under normal operation. Additional levels

Results and discussion

Of special interest to the performance of wind turbine blades operating in the atmospheric boundary layer is the effect of turbulence, either from the atmosphere or induced from the turbine wake, on the aerodynamics and ultimately the power output of the turbine.

Fig. 4 is a schematic of the S809 airfoil with its relevant forces considered in this study. The airfoil is pitched up by an angle of attack, α from the horizontal free stream velocity, U. The blade generates aerodynamic lift, L and

Conclusions

The role of turbulence on the aerodynamic efficiency of an S809 wind turbine blade has been experimentally determined by means of static pressure measurements on the blade surfaces, hot-wire probe measurements of the wake velocity deficit, and 2-D particle image velocimetry of the mean flow. The decrease in lift and overall reduced aerodynamic efficiency (as shown by the lift-to-drag ratio) under the influence of high levels of turbulence was limited to 0° angle of attack. However, for higher

Acknowledgements

The authors would like to acknowledge the funding agencies and people who supported this study: from the Office of Naval Research (ONR), Dr. Ronald Joslin. The National Science Foundation (NSF) and its programs: the Graduate K-12 Fellowship for “Discovery-based Activities in Energy and the Environment” (support from Fall 2010–Spring 2012) and the former Alliance for Graduate Education and the Professoriate (AGEP) at Rensselaer Polytechnic Institute. Finally, we would like to thank Jason Li

References (16)

  • J. Hojstrup

    Spectral coherence in wind turbine wakes

    J. Wind Eng. Ind. Aerodyn

    (1999)
  • C. Sicot et al.

    Rotational and turbulence effects on a wind turbine blade. Investigation of the stall mechanisms

    J. Wind Eng. Ind. Aerodyn.

    (2008)
  • X. Amandolèse et al.

    Experimental study of the effect of turbulence on a section model blade oscillating in stall

    Wind Energy

    (2004)
  • B. Brzek et al.

    Effects of free-stream turbulence on rough surface turbulent boundary layers

    J. Fluid Mech.

    (2009)
  • R.B. Cal et al.

    Experimental study of the horizontally averaged flow structure in a model wind-turbine array boundary layer

    J. Renew. Sustain. Energy

    (2010)
  • H.W. Coleman et al.

    Experimentation and Uncertainty Analysis for Engineers

    (1989)
  • R.A Davis

    Practical Numerical Methods for Chemical Engineers

    (2015)
  • Goett, H.J., 1938. Experimental Investigation of the Momentum Method for Determining Profile Drag. Report No. 660...
There are more references available in the full text version of this article.

Cited by (34)

  • The influence of freestream turbulence on the temporal pressure distribution and lift of an airfoil

    2021, Journal of Wind Engineering and Industrial Aerodynamics
    Citation Excerpt :

    It is possible that flow homogeneity plays a role in the behaviour of stall angles here, as Wang et al. (2014) and Maldonado et al. (2015) conducted their experiments farther downstream of their grids (x/M ​= ​14 and 39, respectively) than Devinant et al. (2002) and Kamada et al. (2011) (x/M ​= ​5 and 6.6, respectively). The stall angle behaviour in the present study agrees with Wang et al. (2014) and Maldonado et al. (2015) in that it remains constant, and the common feature is that all three studies were sufficiently far from the grids, suggesting a grid-based dependency of the other two studies. We would like to note that this is not conclusive evidence that flow homogeneity is the most significant factor in the stall angle behaviour, as other factors exist, such as the different airfoil profiles.

  • The effect of the integral length scale of turbulence on a wind turbine aerofoil

    2020, Journal of Wind Engineering and Industrial Aerodynamics
View all citing articles on Scopus
View full text