Cornell University Laboratory for Intelligent Machine Systems Dynamically Variable Blade Geometry for Wind Energy Greg Meess, Michael Ross Dr. Ephrahim.

Cornell UniversityLaboratory for Intelligent Machine Systems

Dynamically Variable Blade Geometry for Wind Energy

Greg Meess, Michael RossDr. Ephrahim Garcia

Laboratory for Intelligent Machine Systems

AIAA Regional Student Conference

Boston UniversityApril 23-24, 2010


Goal: Increase wind turbine energy output by morphing blade shape to match changing wind speeds.

Pitch Chord

Twist


Outline• Motivation• Experimental Design• Airfoil Generation• Simulation• Optimization• Results

– Geometry– Power output


Motivation

• Wind turbines are constantly increasing in size– Power output is proportional to rotor swept area– The largest turbines cannot be built on land

• Blades are designed for higher wind speeds– Maximize rated power– Turbine spends little time

operating at rated power

• Little focus on low

wind speeds– Variable Pitch

http://www.terramagnetica.com/2009/08/01/why-are-wind-turbines-getting-bigger/


Problem Parameterization • Blade Element Momentum

(BEM) Theory is used• Turbine has operating regime

between 4 m/s and 20 m/s– 4 m/s is lower limit of current

turbines• Fixed speed generator of 60

rpm– Rotations vary from 30 to

120 rpm.• Rayleigh Distribution is used

to assess annual power output

• Chord, twist, and camber are examined

Vestas V90 power output vs. wind speed

Sample wind speed Rayleigh distribution


Airfoil Generation• NACA XX12 Series

– Leading edge, trailing edge follow NACA equations

– Flexible panels connect to leading edge, rest on trailing edge

– As chord extends/retracts, panels keep airfoil profile

• XFOIL Simulation– CL, CD data collected for angles of

attack between -10° and 45°

NACA 2412 original, fully extended, and fully retracted shapes

Sample data from XFOIL for modified shapes


Turbine Performance Analysis• Equations based on basic BEM theory1, WT_Perf

source code2, and Aerodyn Theory Manual3.– Blade divided into a number of elements– Power of each element is P= 1/2ρAU34a(1-a)

• Power Coefficient Cp = 4a(1-a)– Axial induction factor defined as a = (U1-U2)/U1– Need initial guess for axial induction factor– Axial induction factor calculated using relative wind

angle, coefficients of lift and drag, tip loss factor– Initial axial induction factor updated– Iterate for convergence– Calculate power

Polyamide

Nylon “Kite Wing”

1 Manwell, J.F., et al., Wind Energy Explained, John Wiley & Sons Ltd., 2002.2 Buhl, Marshall, National Renewable Energy Laboratory, 2004.3 Laino, David and A. Hansen, User’s Guide to Wind Turbine Aerodynamics Software AeroDyn, Windward Engineering, 2002.

Streamtube around wind turbine rotor, used as basis for BEM theory (Manwell 85).

Blade geometry for analysis of horizontal axis wind turbine (Manwell 108).


Parametric Study• Performance of morphing blades compared to that of a fixed blade

– Sample blade from WT_Perf optimized across all parameters at wind speed of 10 m/s

– All morphing blades begin with this shape• Each morphing blade changes one parameter

– Three chord scenarios are examined• Extension only• Extension and retraction• Retraction only

Add arrows



Optimization


Morphology Plot

Low Speed Shape

High Speed Shape

Variable Pitch

15°



Morphology Plot

Variable CamberAdd picture


Highlight new lines

Annual Output

Power Curve


Morphology Surface

Low Speed Shape

High Speed Shape

Variable Chord

Emphasize retraction over others

Define retraction factor


Highlight new lines


Morphology Surface

Low Speed Shape

High Speed Shape

Variable Twist


Highlight new lines

Clarification/vertical lines


Conclusion

• Variable Twist has the most influence on the performance– Consistent 5% improvement over current pitch control scheme– Achievable using torque tube mechanism– Shape distribution close to linear

wind speed (m/s) retracting chord

variable pitch

variable twist

variable camber

Fair (6.7) 18.64% 23.54% 29.26% 21.77%

Good (7.25) 15.51% 18.45% 23.71% 17.09%

Excellent (7.75) 13.44% 14.98% 20.14% 13.79%

Outstanding (8.4) 11.56% 11.60% 16.99% 10.47%

Superb (10.45) 9.67% 7.51% 14.22% 6.16%

Percent Improvement over Static Blade

Find V-22 paper or illustration

Cite “Fair”, “Good”,etc.

Emphasize improvement over pitch


Future Work


Acknowledgements

Find official titles


Questions & Comments?

Laboratory for Intelligent Machine SystemsAcknowledgements: Professor Sidney Leibovich, Donald Barry

Cornell University Laboratory for Intelligent Machine Systems Dynamically Variable Blade Geometry for Wind Energy Greg Meess, Michael Ross Dr. Ephrahim.

Documents

wind turbine rotor

wind turbine energy

wind energygreg meess

relative wind angle

higher wind speedsmaximize

changing wind speeds

blade shape

blade changes