Are We Any Better Informed? Dr Antony Robotham Auckland University - PowerPoint PPT Presentation

Vertical Axis Wind Turbines Are We Any Better Informed? Dr Antony Robotham Auckland University of Technology NZWEA 16 Apr 2014

• From the mid-1970s,experimental development of vertical axis wind turbines (VAWTs) was underpinned by an incremental improvements of aerodynamic performance prediction methods  • The double actuator disk, multiple streamtube theory(DMST)with corrections for streamtube expansion, dynamic stall, blade tip effects and flow curvature was the state-of-the-art and was easier to implement than the computationally more demanding vortex methods Introduction – The Early Years

• Commercially, VAWTs have not been as successful as the three-bladed, pitch controlled horizontal axis turbine became the standard for the industry and progress in VAWT development stagnated  • However, a renewed interest in VAWTs has emerged, prompted by the development of small turbines for use in urban environments and perceived advantages for large offshore turbines Introduction – More Recently

The objective today is to: • Present trade studies of VAWTs using the double actuator disk, multiple streamtube theory • Look at some recent third party findings • Present some initial VAWT investigations using a proprietary CFD software tool that uses a mesh-less approach to fluid dynamics modelling Objective

VAWTS

• Renewed interest in VAWTs for urban and offshore applications • New generations of VAWT favour the helical configuration • Helical VAWTs are being designed with large height/diameter ratios Observations

WHY HELICAL VAWTS?

What does VAWT theory tell us ?

Ω R 𝑊 ∞ VAWT Multiple Streamtube Theory

𝑈ℎ𝑓 𝑡𝑢𝑠𝑓𝑏𝑛𝑢𝑣𝑐𝑓 𝑗𝑡 𝑐𝑝𝑣𝑜𝑒𝑓𝑒 𝑐𝑧 𝑡𝑢𝑠𝑓𝑏𝑛𝑚𝑗𝑜𝑓𝑡 𝑏𝑢 𝝒 𝑏𝑜𝑒 𝝒 + 𝜺𝝒 𝑊 𝑥 𝑊 𝑒 𝑊 𝑛 𝑊 𝑏 𝑊 𝑣 𝑊 ∞ 𝑞 𝑏 𝜘 𝑞 𝑏 𝜘 𝑞 𝑏 𝜄 𝜄 𝛾 𝜘 = 0 𝑆 Ω after Sharpe & Read (1982) VAWT Streamtube

𝑊 𝑣 𝜌 2 + 𝜘 𝛽 𝑣 Ω𝑆 Upstream Relative Velocity Triangle

𝜘 𝑣 cos 𝜘 𝑊 𝑣 𝜌 𝑊 2 + 𝜘 𝛽 𝑣 𝑊 𝑣 sin 𝜘 Ω𝑆 Upstream Relative Velocity Triangle

𝜘 𝑣 cos 𝜘 𝑊 𝑣 𝜌 𝑊 2 + 𝜘 𝛽 𝑣 𝜀𝑈 𝑊 𝑣 sin 𝜘 Ω𝑆 𝜀𝑂 Upstream Relative Velocity Triangle

𝑫 𝒒𝒏𝒃𝒚 = 𝟏. 𝟔𝟑𝟑 𝑏𝑢 𝒍 = 𝟏. 𝟒𝟔𝟗 𝜸 𝑥ℎ𝑓𝑠𝑓 𝑙 = 𝑏𝜏 Ω𝑆 4𝜌 𝑊 ∞ 𝑾 𝒗 𝑾 ∞ 𝑾 𝒆 𝑾 ∞ 𝛞 ° 𝛄 ° 𝛄 ° 0 180.0 0.821 0.0 0.463 10 172.8 0.824 12.8 0.471 𝑾 𝒗 𝑾 ∞ 𝑾 𝒆 𝑾 ∞ 20 165.4 0.832 25.4 0.495 30 157.8 0.845 37.8 0.535 40 149.9 0.863 49.9 0.589 50 141.6 0.885 61.6 0.655 60 132.9 0.911 72.9 0.732 70 123.8 0.939 83.8 0.816 80 114.3 0.969 94.3 0.907 90 104.4 1.000 104.4 1.000 after Sharpe & Read (1982) Flow Field Through Turbine k = 0.3 .358

𝜸° 0 90 180 270 360 30 𝝉 = 𝟏. 𝟒 𝝁 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 20 1.1 2.1 10 3.1 4.1 𝜷° 0 5.1 6.1 -10 7.1 8.1 -20 ADVANCING ROTOR RETREATING ROTOR -30 H-VAWT Characteristics: 𝜷° 𝒘𝒕 𝜸°

𝜸° 0 90 180 270 360 0.4 𝝉 = 𝟏. 𝟒 𝝁 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 0.3 1.1 2.1 3.1 0.2 4.1 𝑫 𝒓 5.1 0.1 6.1 7.1 8.1 0.0 -0.1 H-VAWT Characteristics: 𝑫 𝒓 𝒘𝒕 𝜸°

𝜸° 0 90 180 270 360 4.0 𝝉 = 𝟏. 𝟒 𝝁 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 3.0 1.1 2.1 3.1 2.0 4.1 𝑫 𝒐 5.1 1.0 6.1 7.1 8.1 0.0 -1.0 H-VAWT Characteristics: 𝑫 𝒐 𝒘𝒕 𝜸°

How do the different VAWTs compare ?

GEOMETRY VARIATIONS

• The derivations presented so far assume an H-type VAWT with straight blades that are parallel to the axis of rotation • The Φ -type (Troposkien Darrieus) and V-type VAWTs have blades with segments that are inclined to the vertical axis • The helical (Gorlov) VAWT have blades that are inclined to the horizontal plane • These geometry variations increase the effective area of the blade in the streamtube but modify the relative wind vectors and orientation of the aerofoil forces VAWT Geometry Variations

• The following studies are based upon the rotor geometry of a typical helical VAWT with 𝑶 = 𝟒 and 𝝉 = 𝟏. 𝟒 operating in a constant windspeed of 𝑾 ∞ = 𝟐𝟑 𝒏/𝒕 𝑾 ∞ = 𝟐𝟑 𝒏/𝒕 helical H-type V-type Φ -type Height (mm) 5300 5300 1500 3820 Diameter (mm) 3000 3000 3000 3000 Swept Area (m 2 ) 15.9 15.9 2.24 8.0 𝐝𝐩𝐭 𝟑 𝝎 = 𝟏. 𝟖𝟓 Chord (mm) 200 200 200 200 Blade Span (mm) 6161 5300 1980 5017 Angle 𝝎 ° 30.7° - - - Angle 𝝔 ° - - 45° 0° - 57° Airfoil Section NACA 0018 Trade Studies of VAWT Variations

𝝁 0 1 2 3 4 5 6 7 8 0.5 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 𝝉 = 𝟏. 𝟒 0.4 0.3 Gorlov 𝑫 𝒒 H-VAWT V-VAWT 0.2 Darrieus 0.1 0.0 -0.1 Trade Studies: 𝑫 𝒒 𝒘𝒕 𝝁

𝝁 0 1 2 3 4 5 6 7 8 0.20 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 𝝉 = 𝟏. 𝟒 0.15 Gorlov 0.10 H-VAWT V-VAWT 𝑫 𝒓 Darrieus 0.05 0.00 -0.05 Trade Studies: 𝑫 𝒓 𝒘𝒕 𝝁

𝜸° 0 90 180 270 360 0.40 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 𝝉 = 𝟏. 𝟒 0.30 𝝁 = 𝟒. 𝟕 Gorlov 0.20 H-VAWT 𝑫 𝒓 V-VAWT Darrieus 0.10 0.00 -0.10 Trade Studies: 𝑫 𝒓 𝒘𝒕 𝜸° at Tip Radius

𝜸° 0 90 180 270 360 2.00 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 𝝉 = 𝟏. 𝟒 𝝁 = 𝟒. 𝟕 1.00 Gorlov H-VAWT 𝑫 𝒐 V-VAWT Darrieus 0.00 -1.00 Trade Studies: 𝑫 𝒐 𝒘𝒕 𝜸° at Tip Radius

𝜸° 0 90 180 270 360 1.0 0.8 Gorlov 0.6 H-VAWT 𝑾 V-VAWT Darrieus 0.4 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 0.2 𝝉 = 𝟏. 𝟒 𝝁 = 𝟒. 𝟕 0.0 Trade Studies: 𝑾 𝒘𝒕 𝜸° at Tip Radius

𝜸° 0 90 180 270 360 20 15 10 Gorlov H-VAWT 5 V-VAWT 𝜷° Darrieus 0 -5 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 𝝉 = 𝟏. 𝟒 -10 𝝁 = 𝟒. 𝟕 -15 Trade Studies: 𝜷° 𝒘𝒕 𝜸° at Tip Radius

𝑫 𝒓 𝑫 𝒓 0.35 0.35 0.30 0.30 0.25 0.25 0.20 0.20 0.15 0.15 0.10 0.10 0.05 0.05 0.00 0.00 0 -0.05 0 -0.05 90 90 180 Gorlov 180 H-VAWT 270 𝜸° 𝜸° 270 360 360 𝑫 𝒓 𝑫 𝒓 0.35 0.35 0.30 0.30 0.25 0.25 0.20 0.20 0.15 0.15 0.10 0.10 0.05 0.05 0.00 0.00 0 0 -0.05 -0.05 90 90 180 180 270 270 𝜸° V-VAWT 𝜸° Darrieus 360 360 Spanwise Variation of 𝑫 𝒓 𝒘𝒕 𝜸° at 𝝁 = 𝟒. 𝟕

𝜸° 𝜸° 0 90 180 270 360 0 90 180 270 360 0.40 0.40 0.30 0.30 1 𝑫 𝒓 𝑫 𝒓 1 0.20 0.20 2 2 3 3 0.10 0.10 Rotor Rotor 0.00 0.00 -0.10 -0.10 Gorlov H-VAWT 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 𝝉 = 𝟏. 𝟒 𝝁 = 𝟒. 𝟕 𝜸° 𝜸° 0 90 180 270 360 0 90 180 270 360 0.40 0.40 1 0.30 0.30 1 2 2 𝑫 𝒓 𝑫 𝒓 0.20 3 0.20 3 Rotor Rotor 0.10 0.10 0.00 0.00 -0.10 -0.10 Darrieus V-VAWT Blade Variation of 𝑫 𝒓 𝒘𝒕 𝜸° at 𝝁 = 𝟒. 𝟕

𝜸° 𝜸° 0 90 180 270 360 0 90 180 270 360 2.5 2.5 𝑮 𝒚 𝑮 𝒚 2.0 2.0 𝑮 𝒛 𝑮 𝒛 1.5 1.5 1.0 1.0 Fx Fx Fy Fy 0.5 0.5 0.0 0.0 -0.5 -0.5 -1.0 -1.0 Gorlov H-VAWT 𝑿𝑺𝒇 = 𝟐𝟑𝟕𝟖𝟕𝟏 𝝉 = 𝟏. 𝟒 𝝁 = 𝟒. 𝟕 𝜸° 𝜸° 0 90 180 270 360 0 90 180 270 360 2.5 2.5 𝑮 𝒚 𝑮 𝒚 2.0 2.0 𝑮 𝒛 𝑮 𝒛 1.5 1.5 Fx Fx 1.0 1.0 Fy Fy 0.5 0.5 0.0 0.0 -0.5 -0.5 -1.0 -1.0 Darrieus V-VAWT Tower Forces 𝑮 𝒚 𝒘𝒕 𝜸° and 𝑮 𝒛 𝒘𝒕 𝜸° at 𝝁 = 𝟒. 𝟕

• The numerical results reflect the general observations made about how VAWT geometry influences aerodynamic performance • The advantage of the helical VAWT is its favourable cyclic loading characteristic, which is offset by a small reduction in performance efficiency Observations

AERODYNAMIC PREDICTIONS

Whilst the multiple streamtube theory is useful for VAWT trade studies, it is limited by: • Quality of available aerofoil data • Models of Dynamic Stall • Local blade geometry effects • Theory breaking down with high solidity rotors (blockage) Limitations of Streamtube Theory

Does CFD offer any substantial new insights into turbine behaviour ?

Paraschivoiu, I., Saeed, F. & Desobry, V. (2002) Numerical Simulation of Dynamic Stall

Ferreira, C.S. (2009) 2D CFD Simulation of Dynamic Stall

Ferreira, C.S. (2009) 3D CFD Tip Effects – Iso-Vorticity Surfaces

Deglaire, P. (2010) 2D CFD Velocity Field

Lanzafame, R., Mauro, S. & Messina, M. (2013) Turbulent KE and Relative Velocity

Marsh, P., Ranmuthugala, D., Penesis, I. & Thomas, G.(2013) Flow Velocity & Vortex Shedding

Scheurich, F., Fletcher, T.M., & Brown, R.E.(2011) VAWT Wakes

Meshless CFD Simulations - XFlow