an mdo framework for the design of wind turbines...cp max: an mdo framework for the design of wind...

Cp-Max: an MDO framework for the design of wind turbines

L. Sartori, Politecnico di Milano

5° Wind Energy Systems Engineering Workshop, Pamplona, Spain October, 4th 2019

-

RESEARCH FRAMEWORK

L. Sartori

Cp Max: an MDO framework for the design of wind turbines

-

Research framework

• Wind turbines are steadily upscaling

• Some design challenges:

o High flexibility of elements

o Unsteady, non-uniform inflow

o Increased aero-elastic couplings

o Increased dynamic complexity

L. Sartori


-

Research framework

• Multi-disciplinary design algorithms (MDO):

o Improve physical description (modelling fidelity)

o Implement a system-level design based on COE

o Manage a wide number of different variables

o Significant research in the last years

L. Sartori


-

CP-MAX:

A MULTI-DISCIPLINARY DESIGN PHILOSOPHY

L. Sartori


-

Cp-Max Timeline

2014: Roll-out of the 330kW-16m blade

2011: Roll-out of the 2MW-45m blade

2012: Roll-out of the 700kW-23m blade

2007: First implementation, parametric design only

2011: External 3D FEM module

2012 - 2015: Studies about aero-structural design

2016: Cp-Max in its new multi-level architecture

2017: Prebend design module

2017-2019: Extensive testing and design studies

L. Sartori


-

L. Sartori


Cp-Max development team

Prof. Carlo L. Bottasso

Chair of Wind Energy

Technisce Universität München

[email protected]

Prof. Alessandro Croce

Head of PoliWind Lab

Politecnico di Milano

[email protected]

Luca Sartori

Post-doc researcher

Politecnico di Milano

[email protected]

Carlo R. Sucameli

PhD Candidate


[email protected]

Helena Canet Tarrés

PhD Candidate


[email protected]

-

L. Sartori


Software architecture

-

Multi-body modelling of wind turbines

Simulation tool: Cp-Lambda

• Multibody solver

• Lifting line + BEM Inflow

• Spatial grid + wind time history

• Nonlinear beam formulation (Bauchau et al. 2001)

• Fully-populated mass/stiffness matrix

L. Sartori


-

MACRO DESIGN LOOP

L. Sartori


-

Macro Design Loop (MDL)

𝛾

• Optimizes global features of the turbine to minimize COE

• Macro design variables:

𝑔 𝑔 𝑔 𝑔 𝒑𝒈= [ 𝑅, 𝐻, 𝜑, 𝛾, 𝜎𝑐 , 𝜏𝑐 , 𝜎𝑡 , 𝜏𝑡 ]

𝜑

Shape Rotor radius 𝐻 parameters

Hub height

Rotor tilt angle

𝑅

Blade coning

L. Sartori


-


• The shape parameters are computed from chord and thickness distributions:

Ch

ord

c(r

) [m

]

r/R [-]

η [-]

Th

ickn

ess τ

(η)

[-]

Rotor solidity 𝑔 𝜎𝑐 =

𝑔 = Rotor tapering 𝜏𝑐

𝑔 Thickness solid area 𝜎𝑡

𝑔

=

Thickness weighted area 𝜏𝑡 =

𝑅 3 𝑐 𝑟 𝑑𝑟

0

𝜋𝑅2

𝑅 𝑑𝑟 𝑟 𝑟𝑐

0𝑅

𝑑𝑟 𝑟 𝑐 0

1 1 𝜂𝑑𝜂 𝑡

100 0

1 𝜂𝑑𝜂 𝑡𝜂

01

𝜂𝑑𝜂 𝑡 0

• They link the MDL with the submodules

L. Sartori


-


• Simplified workflow of the MDL:

MDL (𝐶𝑂𝐸∗, 𝒑∗𝒈) = 𝑀𝑖𝑛𝐶𝑂𝐸(𝒑𝒂, 𝒑𝒃, 𝒑𝒔, 𝒑𝒈,𝑫)

∗ ADS ( 𝒑𝒂, 𝐴𝐸𝑃

∗) = 𝑀𝑎𝑥𝐴𝐸𝑃(𝒑𝒂, 𝒑𝒃, 𝒑𝒔, 𝒑𝒈, 𝑫) ∗ CST ( 𝒓ε

∗) = 𝐶𝑟𝑒𝑎𝑡𝑒𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝐿𝑎𝑤𝑠( 𝒑𝒂, 𝒑𝒃, 𝒑𝒔, 𝒑𝒈, 𝑫) ∗ ∗ PDS ( 𝒑𝒃, 𝐴

∗𝛿) = 𝑂𝑝𝑡𝑖𝑚𝑖𝑧𝑒𝑃𝑟𝑒𝑏𝑒𝑛𝑑( 𝒑𝒂, 𝒑𝒃, 𝒑𝒔, 𝒑𝒈, 𝑫, 𝒓ε

∗ ) SDS ∗ ∗ ∗ ∗ ) ( 𝒑𝒔, 𝐼𝐶𝐶

∗) = 𝑀𝑖𝑛𝐼𝐶𝐶( 𝒑𝒂, 𝒑𝒃 , 𝒑𝒔, 𝒑𝒈,𝑫, 𝒓ε

∗ ∗ ∗ 𝐶𝑂𝐸 = 𝐶𝑜𝑠𝑡𝑀𝑜𝑑𝑒𝑙(𝐴𝐸𝑃∗, 𝐼𝐶𝐶∗, 𝒑𝒂 , 𝒑𝒃 , 𝒑𝒔 , 𝒑𝒈, 𝑫, 𝒓ε∗ )

• Available cost models: Scaling (NREL, 2006), Industrial (SANDIA), Refined scaling (INNWIND)

L. Sartori


-

DESIGN SUBMODULES

L. Sartori


-

Aerodynamic Design Submodule (ADS)

• Optimizes the aerodynamic design variables 𝒑𝒂 of the rotor to maximize AEP

𝒑𝒂 ∗ , 𝐴𝐸𝑃∗ = max

𝒑𝒂 (𝐴𝐸𝑃 𝒑𝒂 , 𝒑𝒃, 𝒑𝒔, 𝒑𝑔 )

s.t.:

𝒗𝒕𝒊𝒑 ≤ 𝒗𝒕𝒊𝒑 𝒎𝒂𝒙

𝒈𝒂 𝒑𝒂 , 𝒑𝒈 ≤ 𝟎

Design variables:

• Chord

• Twist

• Thickness (position of airfoils)

L. Sartori


-


Chord, twist and thickness are parameterized by multiplicative/additive bumping functions

𝑐 η = 𝑠𝑐(η)𝑐𝑏𝑙(η) 𝜃 η = 𝑠𝜃(η)𝜃𝑏𝑙(η) 𝑡 η = 𝑠𝑡(η)𝑡𝑏𝑙(η)

𝑠𝑐(η) = 𝒏𝑐 η 𝒄 𝑠𝜃(η) = 𝒏𝜃(η)𝜽 𝑠𝑡(η) = 𝒏𝑡 η 𝒕

n are spline shape functions

θ, c and t are the associated nodal parameters

L. Sartori


-


• Theoretical AEP from flexible Cp-TSR curves

• The structure is frozen

• Blade shape and thickness constrained

by the MDL: 𝐴𝑒𝑟𝑜(𝑐 η , 𝑅) = 𝑔 𝜎𝑐 𝜎𝑐 𝐴𝑒𝑟𝑜(𝑐 η , η, 𝑅) = 𝑔 𝜏𝑐 𝜏𝑐 𝐴𝑒𝑟𝑜(𝑡 η , 𝑅) = 𝑔 𝜎𝑡 𝜎𝑡 𝐴𝑒𝑟𝑜(𝑡 η , η, 𝑅) = 𝑔 𝜏𝑡 𝜏𝑡

• Other constraints on max chord, tip speed, shape regularity

Ch

ord

c(r

) [m

]

r/R [-]

η [-]

Th

ickn

ess τ

(η)

[-]

L. Sartori


-

Control Synthesis Tool (CST)

∗ • Determines the control laws 𝒓𝜺 of the turbine

• Required to perform DLC simulations

• Available options:

o PID on pitch + LUT on torque

o PID on pitch + PID on torque (DTU)

o MIMO-LQR model based

o Individual Pitch Control

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0 2 4 6 8 10 12 14 16 18 20

No torque to promote rotor acceleration

L. Sartori


-

𝒑𝒃 ∗ , 𝐴𝛿

∗ = min 𝒑𝒃

(𝐴𝛿 ∗ 𝒑𝒂

∗ , 𝒑𝒃, 𝒑𝒔, 𝒑𝑔, 𝑟𝜀 ∗ )

s.t.:

𝒈𝒂 𝒑𝒂 , 𝒑𝒈 ≤ 𝟎

Design variables:

• Prebend shape

Prebend Design Submodule (PDS)

• Optimizes prebend to maximize the deformed swept area

L. Sartori


-

Prebend Design Submodule (PDS)

• Prebend is parameterized by a Bézier curve

• Static analysis at rated conditions

• Possible constraints on:

o Max tip prebend

o Max steepness

o Shape regularity

Max (+) prebend

Max (-) prebend

Max steepness

• No associated variables in the MDL Horizontal tangent

L. Sartori


-

Structural Design Submodule (SDS)

L. Sartori


• Optimizes internal structure of blade, tower to minimize the ICC

𝒑𝒔 ∗ , 𝐼𝐶𝐶∗ = min

𝒑𝒔 (𝐼𝐶𝐶 𝒑𝒂

∗ , 𝒑𝒃 ∗ , 𝒑𝒔, 𝒑𝑔, 𝑟𝜀

∗ )

s.t.:

𝒈𝒔 𝒑𝒂 ∗ , 𝒑𝒃

∗ , 𝒑𝒔, 𝒑𝑔, 𝑟𝜀 ∗ ≤ 𝟎

Design variables:

• Thickness of blade structural components

• Wall thickness of tower segments

• Diameter of tower segments

-


• Thickness of elements is defined at • Sectors defined along tower height selected stations

• Mass, stiffness, stress/strain and • Sectional mass, stiffness computed by buckling computed analytically

ANBA

• Loads and deformations from

Cp-Lambda sensors

L. Sartori


-


Computation of DLC

Aero-elastic optimization

3D FEM verification

Coarse solution

Loads Displacements

Refined solution

Constraints:

Blade tip deflection

Frequencies

Manufacturing limits

Ultimate stress, strain

Fatigue damage

Buckling

L. Sartori


-

AN EXAMPLE: DESIGN OF THE POLIMI

20MW BASELINE WT

L. Sartori


-

Definition of the initial 20 MW model

• Initial aeroelastic definition obtained from upscaling the INNWIND.EU 10 MW

Rated power 20 MW Rotor radius 126 m

IEC Class IC Hub radius 3.95 m

Rotor orientation Clockwise, upwind Hub height 168 m

Control Variable speed Tower height 163 m

Cut-In speed 4 m/s Blade mass 118 ton

Cut-Out speed 25 m/s Hub mass 278 ton

Rated wind speed 11.4 m/s Nacelle mass 1098 ton

Max tip speed 90 m/s Tower mass 1779 ton

Cone angle 2.5° Generator Inertia 8488 kg/m2

Tilt angle 5° Generator efficiency 94%

Reference: INNWIND Deliverable 1.25: PI-based assessment (application) on the results of WP2-WP4 for 20 MW wind turbines, June 2017

L. Sartori


http:INNWIND.EU

-

Design assumptions

• Classic spar-box layout

• No initial prebend

• Fixed-radius optimization

L. Sartori


-

Baseline vs. Redesign comparison

Chord Prebend

L. Sartori


-


PS Spar Cap thickness SS Spar Cap thickness

L. Sartori


-

Structural solution of the redesigned rotor

Shell Panels LE Panels

Spar Panels Shear Webs

L. Sartori


-


Units Baseline 20 MW Redesign 20 MW Var. %

Rotor speed [RPM] 6.77 6.82 + 0.74

Regulation Max tip speed

Optimal TSR

[m/s]

[-]

89.4

7.73

89.93

7.86

+ 0.59

+ 1.68

Rated wind speed [m/s] 11.56 11.45 - 0.95

Design parameters

Max chord

Prebend at tip

Spar caps fiber angle

[m]

[m]

[deg]

8.77

0

0

8.27

4

6

- 5.70

-

-

Total blade mass [ton] 113.5 107.8 - 5.05

KPIs AEP [GWh/yr] 91.63 91.74 + 0.12

COE [€/MWh] 84.92 84.56 - 0.42

L. Sartori


-


Ultimate loads Fatigue loads

BR Blade Root

HC Hub Center

TT Tower Top

TB Tower Base

L. Sartori


-

STATE OF THE ART AND OUTLOOK

L. Sartori


-

Completed research projects

• Definition of 10~20 MW reference wind turbines

• Definition of a 3.4 MW reference WT for the IEA-Task 37 project

• Integration of active/passive load mitigation in the design

• Development of upwind/downwind two-bladed rotors

• Study of the impact of WF control on a single WT design

• Global WT design with noise emissions constraints

L. Sartori


-

L. Sartori


Completed industrial projects

2MW - 45m

300kW - 16m 700kW - 24m

100kW - 10m

1m (for wind tunnel)

-

Ongoing activities

• Integration of airfoil design

• Development of numerical tools for aero-acoustic modelling and design

• Application of the design to down-scaled WTs for wind tunnel

• Integration of WF-level simulation and design modules

• Integration of additional sub-component design modules (DT & generator, support structure)

L. Sartori


-

Selected references

P. Bortolotti, C.L. Bottasso, A. Croce, L. Sartori: Integration of multiple passive load mitigation technologies by automated design

optimization—The case study of a medium‐size onshore wind turbine. Wind Energy, 2019, 22: 65–79, doi:10.1002/we.2270

H. Canet, P. Bortolotti, C.L. Bottasso: Gravo-aeroelastic scaling of very large wind turbines to wind tunnel size.

J. Phys. Conf. Ser., 2018, 1037, 042006, doi:10.1088/1742-6596/1037/4/042006

P. Bortolotti, C.R. Sucameli, A. Croce, C.L. Bottasso: Integrated design optimization of wind turbines with noise emission constraints.


L. Sartori, P. Bortolotti, A. Croce, C.L. Bottasso: Integration of prebend optimization in a holistic wind turbine design tool.


P. Bortolotti, A. Croce, and C.L. Bottasso: Combined preliminary–detailed design of wind turbines.

Wind Energ. Sci., 1, 1–18, 2016, doi:10.5194/wes-1-1-2016

L. Sartori


-

THANK YOU FOR YOUR ATTENTION

L. Sartori


Cp-Max: an MDO framework for the design of wind turbines Cp-Max: an MDO framework for the design of wind turbines L. Sartori, Politecnico di Milano 5° Wind Energy Systems Engineering Workshop, Pamplona, Spain October, 42019 th

FigureRESEARCH FRAMEWORK L. Sartori Cp Max: an MDO framework for the design of wind turbines Research framework• • Wind turbines are steadily upscaling

• • • Some design challenges:

o o o High flexibility of elements

o o Unsteady, non-uniform inflow

o o Increased aero-elastic couplings

o o Increased dynamic complexity

FigureL. Sartori Cp Max: an MDO framework for the design of wind turbines Research framework • Multi-disciplinary design algorithms (MDO): o o o Improve physical description (modelling fidelity)

o o Implement a system-level design based on COE

o o Manage a wide number of different variables

o o Significant research in the last years

FigureL. Sartori Cp Max: an MDO framework for the design of wind turbines FigureCP-MAX: A MULTI-DISCIPLINARY DESIGN PHILOSOPHY L. Sartori Cp Max: an MDO framework for the design of wind turbines Cp-Max Timeline 2014: Roll-out of the 330kW-16m blade 2011: Roll-out of the 2MW-45m blade 2012: Roll-out of the 700kW-23m blade 2007: First implementation, parametric design only 2011: External 3D FEM module 2012 -2015: Studies about aero-structural design 2012 -2015: Studies about aero-structural design 2016: Cp-Max in its new multi-level architecture 2017: Prebend design module 2017-2019: Extensive testing and design studies L. Sartori Cp Max: an MDO framework for the design of wind turbines L. Sartori Cp Max: an MDO framework for the design of wind turbines Cp-Max development team Prof. Carlo L. Bottasso Chair of Wind Energy Technisce Universität Mchen [email protected] Prof. Alessandro Croce Head of PoliWind Lab Politecnico di Milano [email protected] Luca Sartori Post-doc researcher Politecnico di Milano [email protected] Carlo R. Sucameli PhD Candidate Technisce Universität Mchen [email protected] Helena Canet Tarrés PhD Candidate Technisce Universität Mchen helena.canL. Sartori Cp Max: an MDO framework for the design of wind turbines Software architecture Multi-body modelling of wind turbines Simulation tool: Cp-Lambda • • • Multibody solver

• • Lifting line + BEM Inflow

• • Spatial grid + wind time history

• • Nonlinear beam formulation (Bauchau et al. 2001)

• • Fully-populated mass/stiffness matrix

FigureL. Sartori Cp Max: an MDO framework for the design of wind turbines FigureMACRO DESIGN LOOP L. Sartori Cp Max: an MDO framework for the design of wind turbines FigureMacro Design Loop (MDL) 𝛾 • • • Optimizes global features of the turbine to minimize COE

• • Macro design variables:

𝑔𝑔 𝑔𝑔 𝒑= [ 𝑅, 𝐻, 𝜑, 𝛾, 𝜎, 𝜏, 𝜎, 𝜏] Figure𝒈𝑐 𝑐 𝑡 𝑡

𝜑 Figure

Shape Rotor radius 𝐻 parameters Hub height Rotor tilt angle Figure𝑅 L. Sartori Cp Max: an MDO framework for the design of wind turbines Blade coning Blade coning Blade coning

Macro Design Loop (MDL) • The shape parameters are computed from chord and thickness distributions: Chord c(r) [m] r/R [-] η [-] Thickness τ(η) [-] Rotor solidity 𝑔

𝜎= 𝑐

𝑔 = Rotor tapering 𝜏𝑔 𝑐

Thickness solid area 𝜎𝑔 𝑡

= Thickness weighted area 𝜏= 𝑡

𝑅 3 𝑐 𝑟 𝑑𝑟 Figure

00

𝜋𝑅2

𝑅 𝑑𝑟 𝑟 𝑟𝑐 Figure0

𝑅 𝑑𝑟 𝑟 𝑐 Figure0

11 𝜂𝑑𝜂 𝑡Figure

100 0 1 𝜂𝑑𝜂 𝑡𝜂Figure0

1 𝜂𝑑𝜂 𝑡 Figure0

• They link the MDL with the submodules L. Sartori Cp Max: an MDO framework for the design of wind turbines Macro Design Loop (MDL) • Simplified workflow of the MDL: MDL (𝐶𝑂𝐸, 𝒑) = 𝑀𝑖𝑛𝐶𝑂𝐸(𝒑,𝒑,𝒑,𝒑,𝑫) ∗ ∗∗𝒈𝒂𝒃𝒔𝒈

( 𝒑, 𝐴𝐸𝑃) = 𝑀𝑎𝑥𝐴𝐸𝑃(𝒑, 𝒑, 𝒑, 𝒑, 𝑫) ADS 𝒂∗𝒂𝒃𝒔𝒈

∗ CST ( 𝒓) = 𝐶𝑟𝑒𝑎𝑡𝑒𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝐿𝑎𝑤𝑠( 𝒑, 𝒑, 𝒑, 𝒑, 𝑫) ε∗𝒂𝒃𝒔𝒈

∗∗ PDS (𝒑,𝐴) = 𝑂𝑝𝑡𝑖𝑚𝑖𝑧𝑒𝑃𝑟𝑒𝑏𝑒𝑛𝑑(𝒑,𝒑,𝒑,𝒑,𝑫, 𝒓) ∗ ∗∗ ∗𝒃∗𝛿𝒂𝒃𝒔𝒈ε∗SDS )

(𝒑,𝐼𝐶𝐶) = 𝑀𝑖𝑛𝐼𝐶𝐶(𝒑, 𝒑,𝒑,𝒑,𝑫, 𝒓𝒔∗𝒂𝒃𝒔𝒈ε

∗∗∗ 𝐶𝑂𝐸 = 𝐶𝑜𝑠𝑡𝑀𝑜𝑑𝑒𝑙(𝐴𝐸𝑃,𝐼𝐶𝐶, 𝒑, 𝒑, 𝒑,𝒑,𝑫, 𝒓) ∗∗𝒂𝒃𝒔𝒈ε∗

• Available cost models: Scaling (NREL, 2006), Industrial (SANDIA), Refined scaling (INNWIND) L. Sartori Cp Max: an MDO framework for the design of wind turbines FigureDESIGN SUBMODULES L. Sartori Cp Max: an MDO framework for the design of wind turbines Aerodynamic Design Submodule (ADS) • Optimizes the aerodynamic design variables 𝒑of the rotor to maximize AEP 𝒂

𝒑𝒂 ∗ , 𝐴𝐸𝑃∗ = max 𝒑𝒂 (𝐴𝐸𝑃 𝒑𝒂 , 𝒑𝒃, 𝒑𝒔, 𝒑𝑔 ) s.t.: 𝒗𝒕𝒊𝒑 ≤ 𝒗𝒕𝒊𝒑 𝒎𝒂𝒙 𝒈𝒂 𝒑𝒂 , 𝒑𝒈 ≤ 𝟎 Design variables: • Chord • Twist • Thickness (position of airfoils) L. Sartori Cp Max: an MDO framework for the design of wind turbines Aerodynamic Design Submodule (ADS) Chord, twist and thickness are parameterized by multiplicative/additive bumping functions 𝑐 η Figure

𝑐𝑏𝑙𝜃 η = 𝑠(η)𝑐(η)

𝜃𝑏𝑙𝑡 η = 𝑠(η)𝜃(η)

𝑡𝑏𝑙= 𝑠(η)𝑡(η)

𝑠(η) = 𝒏η 𝒄 𝑐𝑐 Figure

𝑠(η) = 𝒏(η)𝜽 𝜃𝜃

𝑠(η) = 𝒏η 𝒕 𝑡𝑡 Figure

Figuren are spline shape functions θ, c and t are the associated nodal parameters FigureL. Sartori Cp Max: an MDO framework for the design of wind turbines Aerodynamic Design Submodule (ADS) • • • Theoretical AEP from flexible Cp-TSR curves

• • The structure is frozen

• • Blade shape and thickness constrained by the MDL:

𝐴𝑒𝑟𝑜(𝑐 η , 𝑅) = Figure𝑔

𝜎𝑐 𝐴𝑒𝑟𝑜𝜎𝑐 (𝑐 η , η, 𝑅) = Figure𝑔

𝜏𝑐 𝐴𝑒𝑟𝑜𝜏𝑐 (𝑡 η , 𝑅) = Figure𝑔

𝜎𝜎𝑡

𝐴𝑒𝑟𝑜𝜎𝑡 (𝑡 η , η, 𝑅) = Figure𝑔

𝜏𝜏𝜏𝑡

𝜏𝜏𝑡

• Other constraints on max chord, tip speed, shape regularity Chord c(r) [m] r/R [-] η [-] Thickness τ(η) [-] L. Sartori Cp Max: an MDO framework for the design of wind turbines Control Synthesis Tool (CST) ∗ • • • Determines the control laws 𝒓of the turbine 𝜺

• • Required to perform DLC simulations

• • • Available options:

o o o PID on pitch + LUT on torque

o o PID on pitch + PID on torque (DTU)

o o MIMO-LQR model based

o o Individual Pitch Control

No torque to promote rotor acceleration L. Sartori Cp Max: an MDO framework for the design of wind turbines 𝒑𝒃 ∗ , 𝐴𝛿 ∗ = min 𝒑𝒃 (𝐴𝛿 ∗ 𝒑𝒂 ∗ , 𝒑𝒃, 𝒑𝒔, 𝒑𝑔, 𝑟𝜀 ∗ ) s.t.: 𝒈𝒂 𝒑𝒂 , 𝒑𝒈 ≤ 𝟎 Design variables: • Prebend shape Prebend Design Submodule (PDS) • • • Optimizes prebend to maximize the deformed swept area

• • Prebend is parameterized by a Bézier curve

• • Static analysis at rated conditions

• • • Possible constraints on:

o o o Max tip prebend

o o Max steepness

o o Shape regularity

L. Sartori Cp Max: an MDO framework for the design of wind turbines Prebend Design Submodule (PDS) FigureMax (+) prebend Max (-) prebend Max steepness Figure• Horizontal tangent No associated variables in the MDL

L. Sartori Cp Max: an MDO framework for the design of wind turbines Structural Design Submodule (SDS) L. Sartori Cp Max: an MDO framework for the design of wind turbines • Optimizes internal structure of blade, tower to minimize the ICC 𝒑𝒔 ∗ , 𝐼𝐶𝐶∗ = min 𝒑𝒔 (𝐼𝐶𝐶 𝒑𝒂 ∗ , 𝒑𝒃 ∗ , 𝒑𝒔, 𝒑𝑔, 𝑟𝜀 ∗ ) s.t.: 𝒈𝒔 𝒑𝒂 ∗ , 𝒑𝒃 ∗ , 𝒑𝒔, 𝒑𝑔, 𝑟𝜀 ∗ ≤ 𝟎 Design variables: • Thickness of blade structural components • Wall thickness of tower segments • Diameter of tower segments Structural Design Submodule (SDS) • • • • Thickness of elements is defined at • Sectors defined along tower height selected stations

• Mass, stiffness, stress/strain and

• • Sectional mass, stiffness computed by

buckling computed analytically ANBA • Loads and deformations from Cp-Lambda sensors FigureFigureFigureL. Sartori Cp Max: an MDO framework for the design of wind turbines Structural Design Submodule (SDS) Computation of DLC Aero-elastic optimization 3D FEM verification Coarse solution Loads Displacements Refined solution Constraints: Blade tip deflection Frequencies Manufacturing limits Ultimate stress, strain Fatigue damage Buckling L. Sartori Cp Max: an MDO framework for the design of wind turbines FigureAN EXAMPLE: DESIGN OF THE POLIMI 20MW BASELINE WT L. Sartori Cp Max: an MDO framework for the design of wind turbines Definition of the initial 20 MW model • Initial aeroelastic definition obtained from upscaling the INNWIND.EU 10 MW

Rated power Rated power Rated power 20 MW Rotor radius 126 m

IEC Class IEC Class IC Hub radius 3.95 m

Rotor orientation Rotor orientation Clockwise, upwind Hub height 168 m

Control Control Variable speed Tower height 163 m

Cut-In speed Cut-In speed 4 m/s Blade mass 118 ton

Cut-Out speed Cut-Out speed 25 m/s Hub mass 278 ton

Rated wind speed Rated wind speed 11.4 m/s Nacelle mass 1098 ton

Max tip speed Max tip speed 90 m/s Tower mass 1779 ton

Cone angle Cone angle 2.5° Generator Inertia 8488 kg/m2

Tilt angle Tilt angle 5° Generator efficiency 94%

Reference: INNWIND Deliverable 1.25: PI-based assessment (application) on the results of WP2-WP4 for 20 MW wind turbines, June 2017 L. Sartori Cp Max: an MDO framework for the design of wind turbines Design assumptions • • • Classic spar-box layout

• • No initial prebend

• • Fixed-radius optimization

FigureFigureL. Sartori Cp Max: an MDO framework for the design of wind turbines Baseline vs. Redesign comparison Chord Prebend FigureL. Sartori Cp Max: an MDO framework for the design of wind turbines Baseline vs. Redesign comparison PS Spar Cap thickness SS Spar Cap thickness FigureL. Sartori Cp Max: an MDO framework for the design of wind turbines Structural solution of the redesigned rotor Shell Panels LE Panels FigureSpar Panels Shear Webs FigureL. Sartori Cp Max: an MDO framework for the design of wind turbines Baseline vs. Redesign comparison Units Units Units Baseline 20 MW Redesign 20 MW Var. %

Rotor speed Rotor speed [RPM] 6.77 6.82 + 0.74

Regulation Regulation Max tip speed Optimal TSR [m/s] [-] 89.4 7.73 89.93 7.86 + 0.59 + 1.68

TRRated wind speed [m/s] 11.56 11.45 -0.95

Design parameters Design parameters Max chord Prebend at tip Spar caps fiber angle [m] [m] [deg] 8.77 0 0 8.27 4 6 -5.70 --

TRTotal blade mass [ton] 113.5 107.8 -5.05

KPIs KPIs AEP [GWh/yr] 91.63 91.74 + 0.12

TRCOE [€/MWh] 84.92 84.56 -0.42

L. Sartori Cp Max: an MDO framework for the design of wind turbines Baseline vs. Redesign comparison FigureUltimate loads Fatigue loads

BR Blade Root HC Hub Center TT Tower Top TB Tower Base FigureFigureL. Sartori Cp Max: an MDO framework for the design of wind turbines FigureSTATE OF THE ART AND OUTLOOK L. Sartori Cp Max: an MDO framework for the design of wind turbines Completed research projects • • • Definition of 10~20 MW reference wind turbines

• • Definition of a 3.4 MW reference WT for the IEA-Task 37 project

• • Integration of active/passive load mitigation in the design

• • Development of upwind/downwind two-bladed rotors

• • Study of the impact of WF control on a single WT design

• • Global WT design with noise emissions constraints

• • Integration of airfoil design

• • Development of numerical tools for aero-acoustic modelling and design

• • Application of the design to down-scaled WTs for wind tunnel

• • Integration of WF-level simulation and design modules

• • Integration of additional sub-component design modules (DT & generator, support structure)

FigureL. Sartori Cp Max: an MDO framework for the design of wind turbines L. Sartori Cp Max: an MDO framework for the design of wind turbines Completed industrial projects 2MW -45m 300kW -16m 700kW -24m 100kW -10m 1m (for wind tunnel) Ongoing activities L. Sartori Cp Max: an MDO framework for the design of wind turbines Selected references FigureP. Bortolotti, C.L. Bottasso, A. Croce, L. Sartori: Integration of multiple passive load mitigation technologies by automated design optimization—The case study of a medium‐size onshore wind turbine. Wind Energy, 2019, 22: 65–79, doi:10.1002/we.2270 H. Canet, P. Bortolotti, C.L. Bottasso: Gravo-aeroelastic scaling of very large wind turbines to wind tunnel size. J. Phys. Conf. Ser., 2018, 1037, 042006, doi:10.1088/1742-6596/1037/4/042006 P. Bortolotti, C.R. Sucameli, A. Croce, C.L. Bottasso: Integrated design optimization of wind turbines with noise emission constraints. J. Phys. Conf. Ser., 2018, 1037, 042005, doi:10.1088/1742-6596/1037/4/042005 L. Sartori, P. Bortolotti, A. Croce, C.L. Bottasso: Integration of prebend optimization in a holistic wind turbine design tool. J. Phys. Conf. Ser., 2016, 753, 062006, doi:10.1088/1742-6596/753/6/062006 P. Bortolotti, A. Croce, and C.L. Bottasso: Combined preliminary–detailed design of wind turbines. Wind Energ. Sci., 1, 1–18, 2016, doi:10.5194/wes-1-1-2016 L. Sartori Cp Max: an MDO framework for the design of wind turbines FigureTHANK YOU FOR YOUR ATTENTION L. Sartori Cp Max: an MDO framework for the design of wind turbines

an mdo framework for the design of wind turbines...cp max: an mdo framework for the design of wind...

Documents