Back to the programme printer.gif Print




Delegates are invited to meet and discuss with the poster presenters in this topic directly after the session 'Aerodynamics and rotor design' taking place on Wednesday, 12 March 2014 at 09:00-10:30. The meet-the-authors will take place in the poster area.

Vanessa Del Campo UPC , Spain
Co-authors:
Vanessa Del Campo (1) P Toni Sant (2) F Daniel Micallef (2) Carlos Simao Ferreira (3)
(1) UPC , Terrassa, Spain (2) University of Malta, Msida, Malta (3) TUDelft, Delft, Netherlands Antilles

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2014 and are published here unedited

Dr. Vanessa Del Campo has been working for the Aerospace Department at the UPC (Barcelona) for 5 years now. During this period she has been both teaching (Aerospace Technology and Aerodynamics) and researching. Her PhD research work was focused on the Aerodynamics of Wind Turbines. As part of her PhD research, she has collaborated with TUDelft - DUWIND (Delft) and Rutgers (New Jersey). The work done for her PhD thesis was mostly experimental, based on the PIV Technique. She studied Aerospace Engineering at the UPM (Madrid) and RWTH (Aachen).

Abstract

Assessment of the lifting line approximation for wind turbine blade modelling

Introduction

Lifting line vortex approaches have been used numerous times to predict rotor flow fields. Nonetheless there could be severe deficiencies in the flow field close to the blade due to the assumption that blade vorticity is concentrated on a line. If the loading is prescribed as a boundary condition on the flow, the approximation can be thoroughly assessed by comparing the flow field with experimental results and data generated from 3D panel free-wake vortex approach where the blade geometry is fully modelled.

Approach

This study comprises Stereo Particle Image Velocimetry (SPIV) measurements at the TU Delft Open Jet Facility as well as numerical calculations based on potential flow free-wake methods. A 2-bladed, 2m diameter model rotor was used for the investigation. The chord and twist distributions are shown in Figure 1. Further details can be found in Micallef et al. [1]. Velocity measurements were carried at different blade span cross-sections. A measurement sample is shown in Figure 2. For this study only the axial flow condition at a tip speed ratio of 7 and wind speed of 6m/s is considered.
A 3D potential flow, free-wake panel method uses source and doublet elements to represent the 3D blade geometry and the released wake. Results from the simulation are used to compare the loads obtained with experimental estimates as well as to understand the effects of tip three-dimensionality compared to a lifting line approximation.
In order to assess the validity of the lifting line approximation, an inverse free-wake lifting line model is used (refer to Sant et al. [4, 5] for more details). The model requires the input of blade loading in order to determine the angle of attack and the resulting flow field. Since the model can use both the experimental loads as well as the loads obtained from the 3D panel method, the results from the lifting line model can be used to assess the validity of the approximation of concentrating vorticity on a line.
The approach taken in this work can be summarized as shown in Figure 3 which shows how the loads from the experimental results and the 3D panel code are prescribed to the inverse free-wake lifting line code to re-generate a flow field which can then be compared with the SPIV measurements and panel code predictions.


Main body of abstract

Del Campo et al. [2] utilized the velocity fields and its derivatives to derive the wind turbine blade loads by means of a momentum approach, using 3D methodology and a second order Poisson solver in order to reconstruct pressure fields (see Ragni et al. [3] and Van Oudheusden et al. [6] for further references on the methodology). The loads derived from experiment were compared with those calculated using a 3D, potential-flow free-wake vortex panel method. These are shown in Figure 4.
The derived loads from the SPIV results are input as boundary conditions to an inverse free-wake lifting line model. A sample of some preliminary results of the flow field are shown in Figure 5. In order to be able to discriminate between differences due to the body influence and due to the 3D flow effects resulting from the blade tip the analysis is subdivided into two major sections. The lifting line approximation is first tested on blade span positions which are assumed to behave in a predominantly 2D manner. Secondly the lifting line approximation is checked in regions of high three-dimensionality with the use of the panel model.


MID BOARD BLADE SECTIONS (0.4<z/R<0.8)

Considering first the predominantly 2D sections (0.4<z/R<0.8), the loci around which the error in the flow velocities V = (Vradilal, Vtheta, Vaxial) is less than 20% is calculated. A sample figure is shown in Figure 6 which compares the axial flow component differences at a radial position of z/R = 0.61. The results show that over most of the upstream region, the percentage discrepancies between the lifting line results and experimental results are far greater than 20%. Just ahead of the leading edge, the percentage differences reduce. At the trailing edge there is a region of large percentage discrepancies due to the trailing vorticity sheet thickness. For the region shown, extending up to 0.09m downstream, most of the axial flow cannot be determined from the lifting line model. In the full paper, the maximum permissible distance close to the blade will be determined in order to ensure that the error in the velocity estimate is below the 20% threshold.

OUTBOARD BLADE SECTIONS (z/R>0.8)
1. Considering sections which are predominantly 3D (z/R>0.8), the differences between the flows calculated using the lifting line approach and the experimental results are determined.
2. In order to assess further the 3D influences, the loads determined from the panel method are this time used as the load boundary conditions to the inverse lifting line method.
3. The resulting flows from the lifting line procedure are then compared with the panel method solution.

Figure 2 shows the relative velocity contours as obtained from the experiment and from the panel model results. In the full paper the lifting line results will be shown for the case were the loads from the experiment and from the panel model are applied. The percentage differences will also be shown in order to determine discrepancies which are purely due to the inability of the lifting line model to capture 3D effects and those due to the fact that the blade is modelled as a line carrying vorticity.


Conclusion

The lifting line approximation of modelling wind turbine blades has been thoroughly assessed by means of an inverse lifting line free-wake vortex solver coupled with experimental SPIV results and a more sophisticated 3D potential flow panel solver. The loads (determined from the experimental flow velocities and the 3D panel solver) were used as boundary condition to the lifting line solver.
The results show that there are substantial discrepancies up to at least 12cm (1c) upstream of the blade and 9cm (0.75c) downstream of the blade. This result is true for the mid-board region of the blade which is predominantly two-dimensional.
In the tip region, the velocity field resulting from the lifting line calculations were compared with the panel model after applying the loads calculated by the panel solver. The results of this analysis will be provided in the full-paper. The most interesting outcome from this study is to establish whether or not the lifting line is still suitable in singularity regions involving high pressure jumps.
The outcome of this paper is relevant to two audiences. For the rotor blade designer who requires flow information in the vicinity of the rotor, the outcomes of these results have shown that body influences extend to relatively high distances downstream and hence the lifting line does not continue to be a suitable element of approximation. To the scientific community working in the field of blade wake interactions such as for the case of floating offshore rotors, establishing the extent to which the blade effects the local flow is important when the rotor is oscillating in and out of the wake. From the results of this study, the use of lifting line approximations have been shown to be relatively crude for such applications.



Learning objectives
1. To establish the extent of applicability of the lifting line approximation to investigate rotor blades
2. To establish the deficiencies of lifting line approaches in singularity regions such as the tip.



References
[1] Micallef, D. (2012). 3D flows near a HAWT rotor: A dissection of blade and wake contributions. PhD thesis, Delft University of Technology & University of Malta.

[2] Del Campo, V., Ragni, D., Micallef, D., Akay, B., Diez, F. J., and Sim˜ao Ferreira, C. (2013). 3D load estimation on a horizontal axis wind turbine using SPIV. Wind Energy, pages n/a–n/a.

[3] Ragni, D., van Oudheusden, BW. and Scarano, F. (2012). 3D pressure imaging of an aircraft propeller blade-tip flow by phase-locked stereoscopic PIV. Experiments in fluids, 52(2):463_477,

[4] Sant, T., van Bussel, G., and van Kuik, G. (2006). Estimating the angle of attack from blade pressure measurements on the NREL phase vi rotor using a free wake vortex model: Axial conditions. Wind Energ., 9:549–577.

[5] Sant, T., van Bussel, G., and van Kuik, G. (2009). Estimating the angle of attack from blade pressure measurements on the NREL phase vi rotor using a free wake vortex model: Yawed conditions. Wind Energ., 12:1–32.

[6] Van Oudheusden, B. W., Scarano, F., Roosenboom, E.W.M. ., Casimiri, E.W.F., and Souverein, L.J. (2007). Evaluation of integral forces and pressure fields from planar velocimetry data for incompressible and compressible flows. Experiments in Fluids, 43(2):153_162.