Aeroacoustics and Aerodynamic Performance of a Rotor with Flatback Airfoils

usefultenchMechanics

Feb 22, 2014 (2 years and 9 months ago)

75 views

Sandia is a
multiprogram

laboratory operated by Sandia Corporation, a Lockheed Martin Company,

for the United States Department of Energy’s National Nuclear Security Administration


under contract DE
-
AC04
-
94AL85000.

Matthew Barone, Josh Paquette

Sandia National Laboratories, Albuquerque, NM

Eric Simley

University of Colorado, Boulder, CO

Monica Christiansen

Penn State University, State College, PA

Aeroacoustics

and Aerodynamic
Performance of a Rotor with
Flatback

Airfoils

2010 European Wind Energy Conference

Warsaw, Poland

23 April 2010

Outline


Flatback

Airfoils: Motivation and Introduction


Flatback

Airfoil Wind Tunnel Tests


Field Tests of the BSDS Rotor


Modeling of
Flatback

Rotor Noise




Motivation


Wind turbine blade design is a multi
-
disciplinary optimization
problem


Cost of Energy is the ultimate objective function


Optimal aero
-
structural design may differ markedly from the
optimal aerodynamic and optimal structural designs


Basic Design Question


Blade aerodynamics dominate the outboard design


Structural requirements dominate the inboard design


What inboard blade shape provides an optimal structural
design without sacrificing too much aerodynamic
performance?

Sandia Blade System Design Study (BSDS)


Employed a multi
-
disclipinary
, iterative design process


Integrated blade design for aerodynamic performance, low
weight, and manufacturability


Innovative blade design features


Flatback

airfoils


Optimal design of a carbon fiber spar cap


9 m blades were fabricated by TPI Composites for testing

Flatback

Airfoils


S
tructural

Advantages


Structural benefit of larger sectional stiffness for
given chord and thickness.


Results in higher blade strength,

lower blade
weight
.


Aerodynamic

Advantages/Disadvantages


Sensitivity

of lift to leading edge soiling is
reduced.


Drag is increased (although

L/D may
still
increase
)
.


Increased

aerodynamic noise due to blunt
trailing edge.

Flatback

Airfoil Research at Sandia

Wind Tunnel Tests

Field Tests

Computational Modeling

Goal:

Predict and
Quantify Noise
and Drag of
Flatback

Airfoils

Flatback

Airfoil Wind Tunnel Tests

Wind Tunnel Experiments


Virginia Tech Stability Wind Tunnel


Aeroacoustic

test section


Beamforming

microphone array


Airfoil surface pressure taps


Pitot

tube wake surveys


DU97
-
W
-
300 and DU97
-
flatback (10%
trailing edge thickness)


Flatback

tested with/without splitter plate


Chord Reynolds numbers from 1.5 to 3.2
million


Several angles of attack

Facilities and Instrumentation

Tests Performed

Wind Tunnel Noise Measurements

U = 56 m/s

Findings from the wind tunnel tests


Flatback

noise generated a
prominent tone


Tonal frequency and amplitude is
relatively insensitive to


Angle of attack


Boundary layer transition location


Simple splitter plate attachment
reduced noise by ~12 dB

Field Tests of the BSDS Rotor

c

h

BSDS Blade Geometry

flatback

34-m Pad
CTL B
Terrace
Terrace Channel
WATERWAY
Terrace Channel
RESERVOIR
ROAD
ROAD
N
0 100 200
Scale, ft
Prevailing
Wind
2.5 Dia Lateral Spacing
Turbine
Anemometer Tower
Test Turbine and Instrumentation


Site


8.7 m/s average wind speed
at 80 m


Turbine


Modified
Micon

65


19 m Rotor Diameter


23

m Hub Height


Instrumentation


Inflow


Center and off
-
axis met
towers, and nacelle


Wind speed and direction


Power


Loads


Tower, hub, and blade


Noise


32
-
microphone array
centered one hub height
upwind

USDA/SNL
Micon

Test
Turbine

Acoustic Instrumentation

Microphone Array Schematic

45 Total Sensor Locations, configurable to
either a low
-
frequency or high
-
frequency
array.

High
-
frequency
Microphone Ellipse

Low
-
frequency
Microphone Ellipse

Tower

Acoustic Measurements

Averaged Noise Maps at Different Blade Azimuth Positions

250 Hz


Entire Rotor

1250 Hz


Single Blade

Modeling of
Flatback

Rotor Noise

Rotor Performance Model

WTPerf

Performance Model with
CFD
-
generated airfoil tables

Modeling of
Flatback

Noise Source

Observer


Brooks, Pope, and
Marcolini

(BPM) model
for blunt trailing edge noise


Empirical model based on wind tunnel
measurements


Peak amplitude depends on the ratio of
blunt trailing edge thickness to boundary
layer thickness,
h/
d*
.


BPM only had data for
h/
d* < 1.


Modified BPM model


Scaling with flow velocity and blade
dimensions unchanged


Spectral shape function unchanged


Amplitude function modified based on
Virginia Tech wind tunnel data


more
applicable to large
h/
d*


Low
-
frequency directivity function used

Trailing Edge

Boundary
Layer

Turbulent
Wake

h

d*

Modeling of Rotor Noise


Blunt Trailing Edge Noise


Blade divided into span
-
wise sections


Local relative flow velocity obtained
from
WTPerf

model


Modified BPM model applied for each
section with a
flatback

airfoil


Inflow Turbulence Noise


Empirical model of Hubbard and
Shepherd


Other airfoil self
-
noise sources are
not currently considered


Turbulent boundary layer trailing edge


Laminar vortex
-
shedding


Separation

Flatback

Noise Source

Low
-
frequency
noise

BSDS Rotor Noise Predictions

Rotor
-
averaged Noise spectra for a ground observer one hub height upwind

Wind Speed = 8 m/s

Wind Speed = 12 m/s

Utility
-
Scale Rotor Noise Predictions


WindPACT

1.5 MW Reference
Turbine


Rotor Diameter = 72 m


Hub height = 85.3 m


Wind Speed = 11 m/s


Blade Pitch = 2.6 deg.


Rotational Speed = 20 RPM

Rotor
-
averaged Noise spectra for a
ground observer one hub height upwind

Summary


Flatback

airfoil technology can lead to lighter, more efficient
rotors


Flatback

rotor noise is being measured in a subscale field
test


Challenging due to competing hub noise


Noise associated with the blunt trailing edge of
flatbacks

has
been studied using models informed by wind tunnel data


Existing BPM model may be over
-
conservative for
flatback

airfoil noise


Flatback

airfoil noise is predicted to be lower than inflow turbulence
noise for both the subscale BSDS rotor and a reference 1.5 MW rotor
with
flatbacks

Ongoing Work


Acoustically absorbing foam
panels will be added to the
test turbine nacelle


Attenuate hub noise


Isolate inboard blade noise


Splitter plate trailing edge
attachment will be added to
the BSDS blades.


Examine effects on
performance and noise.

BSDS blade with trailing
edge splitter plate