Structural Design of Composite Blades for Wind and Hydrokinetic Turbines
Danny Sale and Alberto Aliseda
Northwest National Marine Renewable Energy Center
Dept. of Mechanical Engineering
University of Washington
Feb. 13, 2012
Outline
2
•Previous Work
–coupled aero-structural optimization (HARP_Opt code)
–simple structural model
•Newly Developed Structural Analysis Tool (CoBlade)
–methodology & applications
•Structural Optimization
–problem formulation
–design of composite blade for tidal turbine
•Recommended Future Work
Previous Work: HARP_Opt
3
flow speed (m/s)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0
5
10
15
20
25
30
35
0
5
10
15
20
25
0
5
10
15
20
25
30
35
40
45
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Rotor SpeedBlade Pitch
Prated CPmax
ωmin
ωmax
Horizontal Axis Rotor Performance Optimization • given: turbine & environmental specifications
• optimizes: blade shape, rotor speed & blade pitch control
• satisfying: maximum Annual Energy Production (AEP)
performance constraints (power, cavitation, etc.)
Previous Work: HARP_Opt
4
Bending Strain:
Simple Structural Model
• Thin-shelled cantilever beam
•One material w/ isotropic properties
• Bending strain is only constraint
• Shell thickness is only design variable
Previous Work: HARP_Opt
5
Coupled Aerodynamic-Structural Optimization
•maximize energy production & minimize blade mass
• genetic algorithm identifies set of Pareto-efficient designs
Moving Forward: Structural Design
6
Develop a tool capable of modeling realistic composite blades
Image: www.Gurit.com
Overview of CoBlade software
7
Structural Analysis and Design of Composite Blades
• realistic modeling of composite blades –arbitrary topology & material properties
• computes structural properties –stiffnesses: bending, torsional, axial
– inertias: mass, mass moments of inertia
–principal axes: inertial/centroidal/elastic principal axes
–offsets: center-of-mass, tension-center, shear-center
• structural analysis tool –arbitrary applied loads & body forces
– recovery of 2D lamina-level strains & stresses
–blade deflection & modal analysis
– linear buckling analysis
• optimization of composite layup
Image: replica of Sandia SNL100-00 wind turbine blade using CoBlade
Methodology
8
Classical Lamination Theory (CLT)
• describes mechanical response of laminated plates
Image: G.S. Bir. “PreComp User’ Guide”
Classical Lamination Theory + Euler-Bernoulli beam model + shear flow
extensional stiffness
coupling stiffness
bending stiffness
Methodology
9
Composite Euler-Bernoulli Beam and shear flow approach
• describes global mechanical behavior of composite beam
Convert Beam Stresses into Equivalent Plate Loads
• recover 2D strains & stress at lamina level
s
f3
f1
f4
f2
Methodology
10
[1] G.S. Bir, 2005. “User’s Guide to BModes: Software for Computing Rotating Beam Coupled Modes,” NREL TP-500-38976, Golden, CO: National Renewable Energy Laboratory.
Linear Buckling Analysis
• pinned boundary conditions (conservative)
• contributions from panel stiffness, curvature, thickness, & width
shear web panels top/bottom surface panels
Modal Analysis
• BModes: Rotating Beam Coupled Modes (NREL code)
beam divided into finite elements
Optimization: Composite Layup
11
blade-root
blade-shell
spar-uni
spar-core
LEP-core
TEP-core
web-shell
web-core
Material Legend:
E11
E22
G12
ν12
ρ
σ11,fT
σ11,fC
σ22,yT
σ22,yC
τ12,y
root build-up
blade tip
LEP TEP spar
transition
to LEP
transition
to TEP
transition
to spar
shear webs
Material Properties: Failure Stresses:
Composite Layup
• all laminates balanced & symmetric
• high & low pressure surfaces symmetric
• identical shear web laminates
Optimization: Design Variables
12
Design Variables
• spar-cap width at inboard & outboard stations
• lamina thicknesses along blade length
Optimization: Objectives & Constraints
13
= BladeMass * minimize:
subject to:
penalty factors for maximum stress
penalty factors for buckling under compression & shear
penalty factor for tip deflection
penalty factor for separation of blade freqs. & rotor freq.
constraints ensure feasible geometry
Optimization: Example Design
14
0
5
10
15
20
25
30
0
400
800
1200
1600
2000
0.0 1.0 2.0 3.0
Bla
de P
itch (
deg)
Root
Bend
ing M
om
ent
(kN
-m)
Flow Speed (m/s)
Root Moment
Blade Pitch
pre
ssure
(P
a)
turbulent eddy
Image: ref [2]
[1] M.J. Lawson, Y. Li, and D.C. Sale, 2011. “Development and Verification of a Computational Fluid Dynamics Model of a Horizontal Axis Tidal Current Turbine.” The 30th International Conference on Ocean, Offshore and Arctic Engineering.
[2] G.S. Bir, M.J. Lawson, and Y. Li, 2011. “Structural Design of a Horizontal-Axis Tidal Current Turbine Composite Blade.” The 30th International Conference on Ocean, Offshore and Arctic Engineering.
Composite Blade Design for Tidal Turbine
• hydrodynamic design: Department of Energy Reference Tidal Current Turbine, ref. [1]
• design loads: extreme operating conditions in Puget Sound, WA., ref. [2]
Optimization: Results
15
CoBlade is fast: single evaluation: ~1 sec, total optimization: ~40 min
Blade mass is minimized, final iteration satisfies all constraints (no penalties)
Optimization: Results
16
normal stress, σzz (MPa)
shear stress, | τzs | (MPa)
buckling criteria, R
-400 -300 -200 -100 0 100 200 300
5 10 15 20 25 30
0 0.2 0.4 0.6 0.8 1
root build-up material
spar-cap material
core material
blade-shell material
Top Surface Lamina Stress Failure Criteria
Optimization: Results
17
0
50
100
150
200
0
50
100
150
200
0
500
1,000
1,500
2,000
Axial Stiffness
Torsional Stiffness
0
100
200
300
flapwise
edgewise
-0.14
-0.09
-0.04
0.01
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
1st mode: 8.72 Hz
2nd mode: 18.32 Hz
3rd mode: 27.73 Hz
mas
s (k
g/m
) ax
ial
stif
fne
ss (
N)
x10
6
tors
ion
al
stif
fne
ss (
N-m
2)
x10
6
be
nd
ing
sti
ffn
ess
(N
-m2)
x10
6
flap
d
isp
lace
me
nt
blade length (m)
18
Conclusions • Capable structural design tool, modeling of complex layups possible with CoBlade
•NOT a replacement for higher-fidelity FEM, but very effective for preliminary design work
• Limited validation studies
–excellent agreement for analytically obtainable results
–good agreement with ANSYS FEM model of tapered composite beam (collaboration w/ Penn. State)
Future Work • Preliminary results seem reasonable, but require further validation
–anisotropic layups
–buckling
–lamina-level strains/stresses
• Repeat coupled aero-structural optimization (HARP_Opt) with structural capabilities of CoBlade
• Include cross-coupled terms from CLT into beam equations
• Public release of CoBlade code & documentation
Thank you! Questions?
19
Acknowledgements
This work has also been made possible by
• National Science Foundation Graduate Research Fellowship under Grant No. DGE-0718124
• Department of Energy, National Renewable Energy Laboratory
• University of Washington, Northwest National Marine Renewable Energy Center
Dr. Mark Tuttle (University of Washington)
Matt Trudeau (Pennsylvania State University)
Extra
20
XR
YR
SC
CM
TC
xtc ytc
xcm
ycm xsc
ysc
θaero θelastic θinertial θcentroidal
reference plane (blade coordinate system x-axis)
chord line
centroidal principal axes
inertial principal axes
elastic principal axes
section
reference axes
R