Papadopoulou K. (2004). Aerofoil (2004). Aerofoil characteristics characteristics for wind turbine applications. applications. In: D. Kereković (ed.). Geographical Information Systems in research & practice. practice. Croatian Information Technology Association – GIS Forum, University of Silesia, Zagreb, 145-153.
AEROFOIL CHARACTERISTICS FOR WIND TURBINE APPLICATIONS Kassandra A. Papadopoulou The University of Manchester, UK e-mail: e-mail:
[email protected] [email protected]
ABSTRACT
The present study investigates experimentally the aerodynamic performance characteristics of the NACA 63-215 wind turbine aerofoil section, over a range of low Reynolds numbers and angles of attack (AoA) up to 65o. Wind tunnel tests were carried out to obtain surface pressure distributions and as well as lift, drag and moment coefficients at four Reynolds, also oil visualization technique and smoke visualization were carried out. The flow around the aerofoil was found to have three characteristic regions which were: the "attached regime", the "high lift/stall development regime, and finally the "flat plate/fully stalled" or "deep stall" regime. The most important characteristics obtained were the following: Clmax = 1.2 was reached at 14o, Cdmin = 0.007 at around 0o and the highest Cl/Cd ratio was reached at 6o for all the Reynolds numbers. The results were in good agreement with other studies, and the critical points were found to correlate with the theory of the specific wind turbine aerofoil. The contribution of this study is to expand the knowledge of the aerodynamic behaviour of the NACA 63-215 at low range of Reynolds numbers (< 524,000), since there is not sufficient data for the specific aerofoil in regard its use as a profile section for wind turbine blades and therefore its use in testing wind farm schemes with GIS applications on wind farm modelling.
1.INTRODUCTION 1.1 Introduction to wind energy
Wind turbines convert the energy from the wind to generate clean electricity (renewable energy), energy), and already over 20,000 turbines are producing electricity world-wide. Most are operating in wind farms, groups of wind turbines generating electricity on a significant scale, single wind turbines are also being used for generating electricity. The European Union has called for an i ncrease in the contribution of renewable energy sources form 4 % to 8% of total energy demand by 2005. Wind energy will play a major part in achieving this target and is assisting in reducing CO2, SO2 and NOx emissions. The installed wind energy capacity in Europe has increased by about 40% per year in the past six years (EWEA – Wind Energy, 2002).
1.2 Wind turbine turbine components components
Wind turbine rotor diameters range up to 65 metres. Wind turbines have usually three bladed which are made of glassfibre reinforced polyester or Wood-Epoxy. Most machines operate at a constant speed between 15-50 revolutions per minute. Power is controlled automatically as wind speed changes, and turbines are stopped at very high wind speeds to protect them from damage. Turbines can range in capacity from several kilowatts to several megawatts. The crucial parameter parameter is the diameter of the turbine - the longer the blades, the larger the area swept by the rotor and the greater the energy output. There are many wind turbine designs, suggesting that there is still plenty of scope for innovation and technological development (EWEA – Wind Energy Technology, 2002).
145
Papadopoulou K. (2004). Aerofoil characteristics for wind turbine applications. In: D. Kereković (ed.). Geographical Information Systems in research & practice. Croatian Information Technology Association – GIS Forum, University of Silesia, Zagreb, 145-153.
1.3 Aerofoils for wind turbines
Modern aerofoils are to a large extent development from numerical calculations and optimisation studies. Flow conditions such as separation at high angles of attack, laminar separation bubbles and transition from laminar to turbulent flow are difficult to predict accurately. Therefore, testing aerofoils is an important issue in aerofoil design (Fuglsang et al , 1998).
Rotational speed
Edgewise force
Blade rotating
Flatwise force
Wind speed
Figure 1.1 A 3-bladed upwind wind turbine and a schematic diagram of the forces acting on a rotating wind turbine blade.
1.4 Aim and Objectives
This project presents the facility for the experimental testing of a 2-dimensional aerofoil used for wind turbines. The aim of this study is to investigate experimentally the aerodynamic performance characteristics of the NACA 63-215, 2-dimensional wind turbine aerofoil section, over a range of low Reynolds numbers using different techniques. The objectives of t he study are: ➢ To measure the pressure, lift and drag distribution for Reynolds numbers between 50,000 and 524,000 at different angles of attack. ➢ To use oil visualization and smoke-wire visualization technique to study the flow distribution on and off the surface. ➢ To identify the different flow conditions at which the aerofoil performs and to obtain further information about flow around the aerofoil up to post-stall conditions. ➢ To state any differences or similarities between the results from the different tests/techniques for the same flow conditions.
1.5 Methodology
There are three basic methods of testing wind turbine rotors: field testing, tow t esting and wind tunnel testing, each method has its own advantages and disadvantages. In the current study the last one is used.
146
Papadopoulou K. (2004). Aerofoil characteristics for wind turbine applications. In: D. Kereković (ed.). Geographical Information Systems in research & practice. Croatian Information Technology Association – GIS Forum, University of Silesia, Zagreb, 145-153.
1.5 Wind tunnel techniques
The first technique used in the wind tunnel is pressure measurement and forces (lift, drag, momentum) measurements, which provide detailed information of the lift forces on the wind turbine blades. The conventional lift vectors, are determined from integration of the pressures over the surface. Where the flow is attached, the velocity via Bernoulli’s equation can be determined from the pressure. The other techniques used in the wind tunnels are the flow visualization which can be divided into two broad categories. The first is surface flow visualization when the visualization medium is applied to the surface such as oil flow. The second type is off the surface, such as smoke-wire technique. Both visualization techniques are used in this study.
2. AEROFOIL AND EXPERIMENTAL SET-UP 2.1 The aerofoil and the wind tunnel
The aerofoil section model tested was the NACA 63-215 (fig. 2.1). The span of the model was 0.254 m and the chord length was 0.15 m, the aerofoil was constructed from two aluminium blocks, which were pinned together. The aerofoil section was equipped with 46 pressure taps. At the trailing edge, a Pitot tube was introduced to measure the pressure, as the trailing edge was too sharp to allow to be drilled in. All tests were performed by mounting the aerofoil horizontally in the test section. The following plates show the aerofoil as it was tested in the wind tunnel.
Figure 2.1 The 2-dimensional aerofoil section of the NACA 63-215 (NASG, 2001), and the actual aerofoil in the Project tunnel from an upstream viewpoint.
Figure 2.2 The British Coal tunnel and the set-up at Goldstein Research Laboratories in Manchester.
147
Papadopoulou K. (2004). Aerofoil characteristics for wind turbine applications. In: D. Kereković (ed.). Geographical Information Systems in research & practice. Croatian Information Technology Association – GIS Forum, University of Silesia, Zagreb, 145-153.
2.2 Reynolds number calculation
The most significant flow factor influencing the behavior of low-speed aerofoils is that of viscosity, which indirectly causes lift and directly causes drag and flow separation. This influence is characterized by the Reynolds number of the aerofoils/fluid combination. For the NACA 63-2xx aerofoil, Reynolds number were calculated according to Wilson (found in Spera, 1994) from the formula:
Re
=
V r c
ν
=
r Ω c 30m / sec 0.5
× 10
6
where, ν = kinematic viscosity of air (m2/s) Vr = relative velocity (m/s) r Ω = local tangential velocity (m/s) c = chord length (m) So, for the maximum speed of the tunnel which is 52 m/sec the Reynolds number was 524,000. The wind speed of the tunnel for the oil and smoke visualisation in the wind tunnel was 5 m/sec (Re = 50,000), this technique was limited to a low flow velocity, since in both cases the oil and smoke would not stay attached on the upper surface of the aerofoil and the wire respectively at higher wind speeds.
2.3 Aerofoil forces from pressure distribution
Velocity and pressure are dependent on each other, Bernoulli's equation states that increasing the velocity, decreases the local pressure and vice versa (Hepperle, 2001). Instead of plotting the velocity distribution, the pressure distribution was plotted (the ratio of the local pressure to the stagnation pressure, called pressure coefficient Cp). Therefore the aerofoil pressure coefficient Cp was calculated from the relation:
Cp
=
− p ∞ 1 / 2 ρu 2
where, p = static pressure p∞ = freestream pressure ρ = density u = velocity
3. RESULTS AND DISCUSSION
The results are presented in the following sections: a) the graphs from the pressure distributions and the forces distribution, b) the oil visualization pictures, c) the smoke-wire pictures, and d) a set of results for comparison at similar conditions for all the three techniques.
3.1 Pressure coefficient
The graph shows the measured Cp curves for the different angles of attack at Re=524,000. The top part corresponds to the pressure coefficient at the top part of the aerofoil and the lower part, the pressure coefficient the bottom part of the aerofoil.
148
Papadopoulou K. (2004). Aerofoil characteristics for wind turbine applications. In: D. Kereković (ed.). Geographical Information Systems in research & practice. Croatian Information Technology Association – GIS Forum, University of Silesia, Zagreb, 145-153. Re = 524,000 -1
-0.5
0
p C
0.5
α = 18
1
α = 20 α = 25 α = 30 1.5 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
x/c
Figure 3.1 Pressure coefficient distribution for angles of attack 18o , 20 o , 25 o , 30 o.
3.2 Aerodynamic force coefficients
Below are shown the aerodynamic loads on the aerofoil section that were measured from the balance system of the Project tunnel.
�
Lift coefficient - the lift graph shows the lift coefficient Cl plotted versus the angle of attack. The graph includes the curves of Cl for three different Reynolds numbers. The distribution for Re = 50,000 is not included since the errors of most of the points of the curve up to the angle of 18o were quite large. Drag coefficient - the drag results are shown in following graph. The drag graph is similar to the lift � graph, but shows the drag coefficient Cd of the aerofoil section, versus the angle of attack. In this case the figure shows the drag distribution for Re = 262,000 and 524,000, since for Re = 50,000 and 131,000 the error in the values were large for low velocities in t he wind tunnel. Liftdistribution 1.4
1.2
1
0.8
0.6
l C
0.4
0.2
0
-0.2
Re = 131,000 Re = 262,000
-0.4 Re = 524,000
-0.6 -20
-10
0
10
20
30
40
50
60
70
AoA
Figure 3.2 The distribution of the lift coefficient for three Re numbers.
149
Papadopoulou K. (2004). Aerofoil characteristics for wind turbine applications. In: D. Kereković (ed.). Geographical Information Systems in research & practice. Croatian Information Technology Association – GIS Forum, University of Silesia, Zagreb, 145-153. Drag distribution 2
1.8
1.6
1.4
1.2
d C
1
0.8
0.6
0.4
Re = 262,000 0.2
Re = 524,000 0 -10
0
10
20
30
40
50
60
70
AoA
Figure 3.3 The distribution of the drag coefficient for two Re numbers.
3.3 Oil visualization technique results
The picture shows the results from the oil visualization experiment in the wind tunnel at AoA = 15o, at Re = 50,000. The flow direction is f rom the top to the bottom of the photograph. Plate 4.7 Aerofoil at AoA = 15o (Re = 50,000).
Laminar flow & separation bubble
Turbulent
flow
Flow separation & vortex formed
Figure 3.4 Oil visualization at an angle of 18o and Re = 50,000.
3.4 Smoke-wire technique results
These results consists of two sections, the first one shows the results at Re = 50,000 for angles of attack from -8o to 25o, the second set shows the same angle of 6o at Re = 10,000, 20,000, 30,000, 40,000 and 50,000, at this angle the separation has started for all the Reynolds numbers and the ratio of Cl/Cd has reached its maximum.
150
Papadopoulou K. (2004). Aerofoil characteristics for wind turbine applications. In: D. Kereković (ed.). Geographical Information Systems in research & practice. Croatian Information Technology Association – GIS Forum, University of Silesia, Zagreb, 145-153.
Flow separation
wake
Stagnation point
Figure 3.5 The aerofoil at Re = 10,000 at AoA = 6 o.
3.5 Comparison of results from all techniques
The figures below show a comparison of the Cp distribution along t he upper surface at AoA = 12o, the oil flow pattern and smoke visualization at the same angle and Re number. Figure 3.6 The Cp graph for the top part of the aerofoil at AoA = 12o. R e = 50,0 0 0 -3
-2 .5
-2
p C
-1 .5
-1
-0 .5
0 0
0 .1
0 .2
0 .3
0 .4
0 .5
0 .6
0 .7
0 .8
0 .9
1
x/c
Figure 3.7 The oil and smoke visualization at AoA = 12 o. The flow shown here from left to right.
151
Papadopoulou K. (2004). Aerofoil characteristics for wind turbine applications. In: D. Kereković (ed.). Geographical Information Systems in research & practice. Croatian Information Technology Association – GIS Forum, University of Silesia, Zagreb, 145-153.
4. CONCLUSION 4.1 Conclusion
Wind tunnel tests with 2-dimensional flow were carried out for the NACA 63-215 aerofoil in two different open return type tunnels. The flow characteristics of this aerofoil were identified in this study by three flow regimes that satisfied the model shown in the table below. Flow Regimes
Angles of attack
Attached Flow
-8 o to 15 o
High lift/stall development flow
15o to 30o
Flat plate/fully stalled (deep-stall)
30o to 65o
In general the correlation between the results and the techniques used was good. The oil flow pictures corresponded well to the transition / separation points and were consistent with the pressure distribution graphs. The results showed that the aerofoil behaved better at higher Re numbers. The pressure distributions were characterized by a narrow suction peak that appeared at around AoA =9oand the stagnation point moved slightly downstream on the leading edge of the pressure side. This caused the flow to accelerate around the sharp-nosed leading edge of the aerofoil. The aerofoil r eached the stall at AoA = 15o, and after the early transition and separation of the flow, the flow became turbulent and “mushroom-like” vortices were formed as seen from the oil visualization. For the pitch or stall controlled wind turbine the possible maximum power output, according to the flow conditions at the time, the NACA aerofoil needs to be positioned at the required angle of attack, 14o in this case, with the maximum Cl. When the wind is too strong or for any other reason the turbine needs to be stopped, the blades should be positioned in a stall condition usually 90o (65o in this case) achieving the maximum Cd that is required.
5. RECOMMENDATIONS FOR FURTHER WORK The Environmental wind Tunnel
Wind farms consist of number of individual wind turbines arranged on a given site so to best utilize the local wind energy. This type appears to call for the placement of most of the units at the locations of strongest flow. However, such concentrations of turbines will cause shielding to a lower wind speed. The designer is seeking an optimal situation in which most energetic regions of the site are exploited without crowding there desirable areas with so many units that array (wake) interference prevents them from achieving the best capture (Lissaman, 1994). Using more wind tunnel testing and a GIS software for wind farms applications this can be achieved. The next step of the study is to use a much larger wind tunnel like t he one showed below. A diagram/map of a wind farm is designed with a variety of arrangements regarding the wind turbines and then a model of the farm is placed in the wind tunnel for further tests.
152
Papadopoulou K. (2004). Aerofoil characteristics for wind turbine applications. In: D. Kereković (ed.). Geographical Information Systems in research & practice. Croatian Information Technology Association – GIS Forum, University of Silesia, Zagreb, 145-153.
Figure 5.1 Visualisation Module by Windfarmer (2004),
REFERENCES European Wind Energy Association – EWEA (2002), “Wind Energy – Clean Power For Generators” [Online]. Http://www.ewea.org/src/flagbro.htm European Wind Energy Association – EWEA (2002), “Wind Energy Technology” [Online]. Http://www.ewea.org/src/technology.htm Flow Science – Goldstein Research Laboratory (2003), “Environmental and Wind Engineering Services”, [OnLine]. http://www.flow-science.eng.man.ac.uk/wind.htm Fuglsang P., Antoniou I., Sørensen N.N. and Madsen H.Aa. (1998), "Validation of a Wind Tunnel Testing Facility for Blade Surface Pressure Measurements", Risø-R-981(EN), Risø National Laboratory, Roskilde. Hepperle M. (2001), "Airfoils", [On Line]. Http://members.tripod.de/MartinHepperle/Airfoils Lissaman P.B.S. (1994), "Wind Turbine Airfoils and Rotor Wakes" found in Spera D.A., Wind Turbine Technology: undamental concepts of wind turbine engineering , New York, ASME Press. Wilson R.E. (1994), "Aerodynamic Behaviour of Wind Turbines" found in Spera D.A., Wind Turbine Technology: undamental concepts of wind turbine engineering , New York, ASME Press. WindFarmer (2004), “The Wind Farm Design and Optimisation Software”, [OnLine]. Http://www.garradhassan.com/windfarmer/windfarmer.htm
153