Dynamics Modeling and Loads Analysis of an Offshore ...

Jason JonkmanJason Jonkman

IEA Annex XXIII OC3 MeetingIEA Annex XXIII OC3 Meeting

September 7, 2007September 7, 2007

Dynamics Modeling and Loads AnalysisDynamics Modeling and Loads Analysis

of an Offshore Floating Wind Turbineof an Offshore Floating Wind Turbine

2IEA Annex XXIII OC3 Meeting

IntroductionBackground — The Big Picture

Some wind turbines have been installed

in shallow water; none in deepwater

A vast deepwater wind resource

represents a potential to use offshore

floating wind turbines to power much of

the world

Numerous platform concepts have been

proposed

Simulation tools for modeling the

dynamic responses needed to be

developed & verified

The technical feasibility must be

demonstrated by characterizing the

dynamic response and identifying critical

loads & instabilitiesGE Wind Energy 3.6-MW Turbine


IntroductionPrevious Research

Henderson

et al

Fulton et al

Withee

Nielsen et al

Zambrano et al

Bulder et al

Lee

Wayman et al

Vijfhuizen

ResearchersResearchers

Limited capabilities:

– simple hydrodynamic

models

– simple aerodynamic

models

– concept-specific tools

Limited simulations in

analysis

Nonlinear representation

of wind turbine & platform

Transient events included

Flexible wind turbine

Controls included

Analysis requires many

simulations

No nonlinear

characteristics

No transient events

Rigid wind turbine

No controls

Many tools available from

offshore O&G industry

Linearized representation

of wind turbine

Analysis presented in

summary form (RAOs,

response statistics)

LimitationsLimitationsCharacteristicsCharacteristics

Fre

qu

en

cy

Do

ma

inF

req

ue

nc

y D

om

ain

Tim

e D

om

ain

Tim

e D

om

ain


IntroductionObjectives & Scope

My objectives:

– to develop a comprehensive simulation tool for modeling the

coupled dynamic response of offshore floating wind turbines

– to verify the simulation capability through model-to-model

comparisons

– to apply the simulation tool to an integrated loads analysis

for one of the promising system concepts

These objectives are important in that they:

– address the primary limitations of previous research

– are fundamental for determining the most technically-

attractive & economically-feasible design

Not included:

– economics, manufacturing, installation, optimization


Outline of Presentation

Introduction

Development of Simulation Capability:

– Overview

– Support Platform Kinematics & Kinetics

– Hydrodynamics

– Mooring System

Design Basis & Wind Turbine Model

Verification of Simulation Capability

Loads Analysis

Influence of Control on Barge Pitch

Conclusions & Recommendations


Simulation Capability DevelopmentOverview — Modeling Requirements

Fully coupled aero-hydro-servo-elastic interaction

Wind-Inflow:–discrete events

–turbulence

Waves:–regular

–irregular

Aerodynamics:–induction

–rotational augmentation

–skewed wake

–dynamic stall

Hydrodynamics:–diffraction

–radiation

–hydrostatics

Structural dynamics:–gravity / inertia

–elasticity

–foundations / moorings

Control system:–yaw, torque, pitch


Simulation Capability DevelopmentCombining Computational Methodologies

Land -Based

Wind Turbines

Characteristics :

_ Flexible & dynamically active

_ Controllable

_ Nonlinear time -domain

aero-servo -elastic analysis

Simulation Tools :

_ FAST with AeroDyn

_ MSC.ADAMS© with

A2AD & AeroDyn

Sea-Based

Oil & Gas Platforms

Characteristics :

_ Rigid & static

_ Passive

_ Linear frequency -domain

hydrodynamic analysis

Simulation Tools :

_ SWIM-MOTION -LINES

_ WAMIT©

Offshore Floating

Wind Turbines

Characteristics :

_ Flexible & dynamically active

_ Controllable

_ Coupling between turbine &

platform motions

_ Nonlinear time -domain

aero-hydro -servo-elastic

analysis needed

Simulation Development Approach :

_Start with FAST with AeroDyn &

MSC.ADAMS with A 2AD & AeroDyn

_Add support -platform DOFs

_Develop hydrodynamics module with

interface to SWIM & WAMIT

_Develop mooring -system module


Simulation Capability DevelopmentFully-Coupled Aero-Hydro-Servo-Elastics


Simulation Capability DevelopmentSupport Platform Kinematics & Kinetics

Support-platform DOFs introduced to

FAST & ADAMS:

– translational: surge, sway, heave

– rotational: roll, pitch, yaw

(assumed small rotations)

Dynamic couplings between motions

of turbine & platform included:

– all position, velocity, and acceleration

expressions are now affected by the

platform DOFs

– the wind turbine’s response to wind and

wave excitation is fully coupled through

the structural dynamics

ZY

X

Heave

Yaw

RollSurge

Sway

Pitch

Wind

ZY

X

Heave

Yaw

RollSurge

Sway

Pitch

Wind

Support Platform DOFs


Simulation Capability DevelopmentHydrodynamics Module

Time -Domain Hydrodynamics (H ydroDyn )

Wave Spectrum

& Direction

Radiation Kernel

Incident -Wave

Excitation

Plat form

Mot ions

Hydrod ynamic

For cing Lo ads

Added -Ma ss Matrix

(Radiation Problem )

Damping Matrix

(Radiation Problem )

Impu lsive

Added Mass

Restoring Matrix

(Hydrostatic Problem )

Wave-Excitation Force

(Diffraction Problem )

White G aussian

No ise

Infinite -Freq.

Limit

Cosine

Transform

Inverse FFT

Box-Muller

Method

Seed for

RNG

Platform

Geometry

Sea

Current

Time

Convolution

Buoyancy

Calculation

Sum Forces

Memory

Effect

Buoy ancy

Morison’s

Equation

Viscous

Drag

Frequency -Domain Radiation / Diffraction

Hydrodynamics Preprocessor (SWIM or WAMIT)

Incid ent -Wave

Kin ematics


Simulation Capability DevelopmentMooring-System Module

Fx

Fz

s

( )x s

( )z s

( )eT s

FH

FV

AH

AV

Fairlead

Anchor

BL, ,EA,C

Fx

Fz

s

( )x s

( )z s

( )eT s

FH

FV

AH

AV

Fairlead

Anchor

BL, ,EA,C

Mooring Lin e Properties

( , , , )

Quasi -Static Mooring -System Module (Calculations Shown for Each Line )

Newton -Raphson Iteration to

Find Fairlead TensionGlobal -to-Local

Transformation

Compute

Anchor

Tension

Compute Configuration of ,

& Tensions Within , Line

Fairlead Location Relative

to Anchor ( , )

Fairlead Tension ( , )

Local -to-Global

Transformation &

Sum Tensions

Platform Position

Mooring Restoring

Anchor Tension

( , )

Configu ration of ,

& Tensions Within , Line

[ , , ]

Fx

Fz

L

EA

BC

FH

FV



Introduction

Development of Simulation Capability

Design Basis & Wind Turbine Model:

– NREL 5-MW Baseline Wind Turbine

– ITI Energy Barge & MIT / NREL Barge

– Reference-Site Data


Loads Analysis




Design Basis & Wind Turbine ModelNREL 5-MW Baseline Wind Turbine

Specifications developedfor a representative multi-megawatt wind turbine

Heavily influenced by theREpower 5M prototype &DOWEC project turbines

Properties developed:– blade structural &

aerodynamic properties

– nacelle & hub

– drivetrain

– tower

– control system:

• generator torque

• blade pitch

REpower 5M Wind Turbine

Gross Properties Chosen for the Baseline TurbineRating 5 MW

Rotor Orientation , Con figuration Upwind , 3 Blades

Control Variable Speed, Collective Pitch

Drivetrain High Speed, Multiple Stage Gearbox

Rotor Diameter 126 m

Hub Height 90 m

Cut-In, Rated, Cut -Out Wind Speed 3 m/s, 11.4 m/s, 25 m/s

Cut-In, Rated Rotor Speed 6.9 rpm, 12.1 rpm

Rated Tip Speed 80 m/s

Overhang, Shaft Tilt, Precone 5 m, 5º, 2.5º

Rotor Mass 110,000 kg

Nacelle Mass 240,000 kg

Tower Mass 347,460 kg

Coordinate Location of Overall CM (-0.2 m, 0.0 m, 64.0 m)


Design Basis & Wind Turbine Model5-MW Baseline Control-System Properties

Low-Pass

Filter

Gene rator

Spe ed

Torque-Speed

Look-Up Table

Determine

Control Region

Control Region

(1, 1_ , 2, 2_ , or 3)

Gene rator

Tor que

Torque Limit

Saturation

Torque Rate

Saturation

Generator

Torque

+Rated

Speed

Filtered

Speed

-

Spe edErr or

Integrator

Integr ated

Speed Error

Saturate

Integral

Gain

Schedule Integral

Gain

Proportional

GainGain

Factor

+

Inte gralTe rm

Propo rtionalTe rm

Pitch Limit

Saturation

Pitch Rate

Saturation

Rotor-C ollective

Blade-Pit ch Angle

Rotor-Collective

Blade-Pitch Angle

+

++

Integr ated

Speed Error

Rotor-C ollective

Blade-Pit ch Angle

Gene rator

Tor que


Design Basis & Wind Turbine ModelFloating Platforms

ITI Energy Barge

Sponsored by ITI Energy, Scotland

Developed by the Universities of

Glasgow & Strathclyde (NAME)

Square barge with central moonpool

& oscillating water column (OWC)

Eight slack, catenary lines

MIT / NREL Barge

Developed by MIT for NREL

Cylindrical barge

Four pairs of taut lines

NREL 5-MW Turbine on ITI Energy Barge

Barge concepts chosen for simplicity in design, fabrication, & installation


Design Basis & Wind Turbine ModelReference-Site Data

Data purchased from http://www.waveclimate.com/:

– JPD of wind speed, wave height & wave period

– extreme wave-height data with 1- & 50-yr recurrence

IEC class IB values assumed for data not measured

Reference Site Location (61.3º N, 0.0º E)

0

5

10

15

20

25

3 5 7 9 11 13 15 17 19 21 23 25

Hub-Height Wind Speed, m/s

Expected Significant Wave Height, m Peak-Spectral Period Range, s



Introduction



Verification of Simulation Capability:

– Hydrodynamics Module

– Mooring-System Module

– Time Versus Frequency Domain

Loads Analysis





Introduction




Loads Analysis:

– Design Load Case Table

– Results & Discussion




Loads AnalysisDesign Load Case Table

Summary of Selected Design Load Cases from IEC61400-1 & -3

DLC Controls / Events Load

Model Speed Model Height Direction Factor

1.1 NTM Vin < V hub < V out NSS Hs = E[H s|V hub ] _ = 0º Normal operation 1.25*1.2

1.3 ETM Vin < V hub < V out NSS Hs = E[H s|V hub ] _ = 0º Normal operation 1.35

1.4 ECD Vhub = V r, V r±2m/s NSS Hs = E[H s|V hub ] _ = 0º Normal operation ; ± wind dir 'n. 1.35

1.5 EWS Vin < V hub < V out NSS Hs = E[H s|V hub ] _ = 0º Normal operation ; ± ver . & hor. shr . 1.35

1.6a NTM Vin < V hub < V out ESS Hs = 1.09*H s50 _ = 0º Normal operation 1.35

2.1 NTM Vhub = V r, V out NSS Hs = E[H s|V hub ] _ = 0º Pitch runaway _ Shutdown 1.35

2.3 EOG Vhub = V r, V r±2m/s, Vout NSS Hs = E[H s|V hub ] _ = 0º Loss of load _ Shutdown 1.10

6.1a EWM Vhub = 0.95*V 50 ESS Hs = 1.09*H s50 _ = 0º, ±30 º Yaw = 0º, ±8º 1.35

6.2a EWM Vhub = 0.95*V 50 ESS Hs = 1.09*H s50 _ = 0º, ±30 º Loss of grid _ -180 º < Yaw < 180 º 1.10

6.3a EWM Vhub = 0.95*V 1 ESS Hs = 1.09*H s1 _ = 0º, ±30 º Yaw = 0º, ±20 º 1.35

7.1a EWM Vhub = 0.95*V 1 ESS Hs = 1.09*H s1 _ = 0º, ±30 º Seized blade ; Yaw = 0º, ±8º 1.10

6) Parked (Idling)

7) Parked (Idling) and Fault

Winds Waves

1) Power Production

2) Power Production Plus Occurrence of Fault


Loads AnalysisResults & Discussion — Overview

Compared land- & sea-based ultimate loads

Problems with both land- & sea-based concepts were

identified:

– land: tower S-S instability when idling with loss of grid

– sea: platform yaw instability when idling with fault

– sea: susceptibility of barge to extreme waves

These problems dominate the load results

To make a fair comparison, the results are split into

groups & presented separately:

– normal operation load cases

– all other load cases:

• idling, faults, & 1- & 50-yr events


Loads AnalysisNormal Operation — DLC 1.1 Statistics


Loads AnalysisNormal Operation — DLC 1.1 Statistics (cont)


Loads AnalysisNormal Operation — Extreme Events Tables

Extreme Events for Blade 1 Root Moments — Land

Extreme Events for Blade 1 Root Moments — Sea

Blade Root

Coordinate System


Loads AnalysisNormal Operation — DLC 1.1 Animation


Loads AnalysisNormal Operation — Sea-to-Land Ratios

0

1

2

3

4

5

RootFMxy1 RootFzc1 RootMMxy1 RootMzc1

Rati

o o

f S

ea t

o L

an

d

0

1

2

3

4

5

GenPwr GenTq GenSpeed RotSpeed

Rati

o o

f S

ea t

o L

an

d

0

1

2

3

4

5

OoPDefl1 IPDefl1 TTDspFA TTDspSS

Rati

o o

f S

ea t

o L

an

d

5.6

0

1

2

3

4

5

TwrBsFMxy TwrBsFzt TwrBsMMxy TwrBsMzt

Rati

o o

f S

ea t

o L

an

d

6.0

6.4


Loads AnalysisNormal Operation — Effect of Wave Height

0

1

2

3

4

5


Rati

o o

f S

ea t

o L

an

d

0

1

2

3

4

5


Rati

o o

f S

ea t

o L

an

d

5.6

0

1

2

3

4

5


Rati

o o

f S

ea t

o L

an

d

0

1

2

3

4

5


Rati

o o

f S

ea t

o L

an

d

6.0

6.4

100% - Original 75% 50% 25% 0% - Still Water


-4

-2

0

2

4

0 100 200 300 400 500 600

Time, s

S-S

T-T

De

fl,

m

No Brake

Brake

Brake Engaged

Loads AnalysisOther Load Cases — DLC 6.2a S-S Instability

Problem: energy absorption exceeds tower S-S damping

Conditions:

– 50-yr wind event for land-based turbine

– idling + loss of grid; all blades = 90º; nacelle yaw error = ±(20º to

40º)

– S-S instability diminished by compliance of barge

Possible solutions:

– modify airfoils to reduce energy absorption

– apply brake to keep rotor away from critical azimuths


Loads AnalysisOther Load Cases — DLC 7.1a Yaw Instability

Problem: rotor azimuth couples with platform yaw motion

Conditions:

– 1-yr wind & wave events for sea-based turbine

– idling + fault; blade pitch = 0º (seized), 90º, 90º

– yaw instability also occurs in sea-based DLC 2.1

Possible solutions:

– reduce fully feathered pitch to allow slow roll while idling

– apply brake to stop rotor

-180

-90

0

90

180

0 100 200 300 400 500 600

Time, s

Pla

tfo

rm Y

aw

,

de

g

No Brake

Brake

Brake Engaged


Loads AnalysisOther Load Cases — 1- & 50-yr Events

Problem: large / steep waves lead

to large barge pitching

Conditions:

– 1- & 50-yr wind & wave events

– power production or idling

Linear hydrodynamic theory is

violated:

– will the real response be better or worse?

Possible solutions:

– streamline barge shape

– add pitch & roll dampers

– install barge only at less severe sites

ITI Energy Concept Moments

Before an Extreme Wave Event



Introduction




Loads Analysis

Influence of Control on Barge Pitch:

– Overview of Pitch-Damping Problem

– Influence of Conventional Turbine Control



In Region 3 pitch-to-feather speed regulation:

tilt into wind wind speed blade pitch thrust tilt more?

tilt out of wind wind speed blade pitch thrust tilt more?

May introduce negative damping of barge-pitch & large loads

Influence of Control on Barge PitchOverview of Pitch-Damping Problem

0

200

400

600

800

1,000

3 5 7 9 11 13 15 17 19 21 23 25

Wind Speed, m/s

Ro

tor

Th

rust,

kN

Region 1_ 2 2_ 3


Influence of Control on Barge PitchOverview of Pitch-Damping Problem (cont)

Hydrodynamic radiation damping, BRadiation

:

– BRadiation

depends on frequency of oscillation:

• damping actually described by time convolution

– instead, damping chosen at natural pitch frequency

Thrust sensitivity to wind speed, :

– two methods for estimating:

• ideal, closed-loop: compute as slope of S-S thrust vs. wind speed curve

“ideal” because there is no rotor-speed error

• open-loop method: perturb wind speed & quantify change in thrust

“open loop” because pitch isn’t varied with wind speed

T V

( ) ( ) ( )Mass Radiation Radiation Viscous Hydrostatic Lines HHI A B B C C L T+ + + + + =&& &

x x x

Hydrostatic LinesMass Radiation Radiation Viscous02 2 2

HH HH HH

M C K

C CI A B B Tx x x T

L L V L

++ ++ + + =��

� � � � � � � � � � � � � � � � ��

x

xn

x

K

M=

x

x

x x

C

2 K M=

HHx L=

0

TT T x

V= �+( ) ( ) ( )Mass Radiation Radiation Viscous Hydrostatic Lines HHI A B B C C L T+ + + + + =�� +


Influence of Control on Barge PitchOverview of Pitch-Damping Problem (cont)

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30

3 5 7 9 11 13 15 17 19 21 23 25

Wind Speed, m/s

Ba

rge

-Pit

ch

Da

mp

ing

Ra

tio

, -

Open-Loop

Ideal Closed-Loop


Feedback of Tower-Top Acceleration

Append speed controller with tower-feedback (TFB) loop:

TFB gains tested for x = 0.05 & 0.5

Active Pitch-to-Stall Speed Regulation

Pitch-to-feather S-S thrust with wind speed

Pitch-to-stall S-S thrust with wind speed

Speed controller redeveloped for pitch-to-stall & tested

May require adaptation of airfoils to get large thrust augmentation

Detuning of the Pitch-to-Feather Gains

Reduce pitch-to-feather gains to make system behave more open-loop

Gains chosen to reduce target response frequency

Upper bound to amount of attainable improvement

Influence of Control on Barge PitchInfluence of Conventional Turbine Control

0

T TT T x

V= +� + Px

K x=� ��Px

x

x x

K T

2 K M

=


Influence of Control on Barge PitchFeedback of Tower-Top Acceleration

0

2,000

4,000

6,000

8,000

Time (sec)

Ge

nP

wr,

kW

Baseline Baseline+TFB

0

500

1,000

1,500

2,000

Time (sec)

Gen

Sp

eed

, rp

m

-40

-20

0

20

40

Time (sec)

BlP

itc

h1

, º

-10

-5

0

5

10

0 50 100 150 200 250 300

Simulation Time, s

Ptf

mP

itc

h,

º


Influence of Control on Barge PitchActive Pitch-to-Stall Speed Regulation

0

2,000

4,000

6,000

8,000

Time (sec)

Ge

nP

wr,

kW

Baseline (Pitch-to-Feather) Pitch-to-Stall

0

500

1,000

1,500

2,000

Time (sec)

Gen

Sp

eed

, rp

m

-30

-15

0

15

30

Time (sec)

BlP

itch

1, º

-20

-10

0

10

20

0 50 100 150 200 250 300

Simulation Time, s

Ptf

mP

itch

, º


Influence of Control on Barge PitchActive Pitch-to-Stall Speed Regulation (cont)

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30

3 5 7 9 11 13 15 17 19 21 23 25

Wind Speed, m/s

Ba

rge

-Pit

ch

Da

mp

ing

Ra

tio

, -

Pitch-to-Feather - Open-Loop

Pitch-to-Feather - Ideal Closed-Loop

Pitch-to-Stall - Open-Loop

Pitch-to-Stall - Ideal Closed-Loop


Influence of Control on Barge PitchDetuning of the Pitch-to-Feather Gains

0

2,000

4,000

6,000

8,000

Time (sec)

Ge

nP

wr,

kW

Baseline Detuned Gains

0

500

1,000

1,500

2,000

Time (sec)

Gen

Sp

eed

, rp

m

0

6

12

18

24

Time (sec)

BlP

itc

h1

, º

-10

-5

0

5

10

0 50 100 150 200 250 300

Simulation Time, s

Ptf

mP

itc

h,

º


Influence of Control on Barge PitchDetuning of the Pitch-to-Feather Gains (cont)

0

1

2

3

4

5


Rati

o o

f S

ea t

o L

an

d

0

1

2

3

4

5


Rati

o o

f S

ea t

o L

an

d

5.6

0

1

2

3

4

5


Rati

o o

f S

ea t

o L

an

d

0

1

2

3

4

5


Rati

o o

f S

ea t

o L

an

d

6.0

6.4

Baseline Detuned Gains



Introduction




Loads Analysis


Conclusions & Recommendations:

– Conclusions

– Recommendations for IEC Annex XXIII OC3


Conclusions & RecommendationsConclusions

Developed fully-coupled simulation capability:

– started with FAST & ADAMS with AeroDyn

– added support-platform DOFs

– developed time-domain hydrodynamics (HydroDyn)

• interfaced SWIM & WAMIT

– developed quasi-static mooring-system module

Verified novel model features:

– hydrodynamics module

– mooring-system module

– time versus frequency domain

– all results were favorable

– gave confidence to perform more rigorous analyses

Dutch Tri-Floater


Conclusions & RecommendationsConclusions (cont)

Performed loads analysis for ITI Energy:– floating concept has similar mean loads, but larger

load excursions than the wind turbine on land

– platform motions affect the blades least, tower most

– improved loads at less severe sites

– barge mitigates S-S instability of idling turbine on land

– barge exhibits yaw instability while idling with fault

– barge susceptible to extreme waves

Examined influence of conventional control:– feedback of tower-top acceleration leads to increased

excursions in rotor speed & power

– active pitch-to-stall leads to increased barge pitching

– detuning of the pitch-to-feather gains leads to reducedbarge pitch, but limited improvement is attainable

Design changes required for technical feasibility


Conclusions & RecommendationsRecommendations for IEA Annex XXIII OC3

Simulate Spar-buoy:– “Hywind” or SWAY concept?

Define:– mass, CM location, & inertia

– diameter, draft, & water depth

– mooring system:• geometry of lines & distributed mass & stiffness or

• equivalent mooring system linear restoring matrix

– hydrodynamic parameters:• linear hydrostatic restoring matrix

• radiation / diffraction or

• equivalent Morison’s formulation

Choose load cases:– full-system natural frequencies

– NTM + NSS + normal operation

– EWM + ESS + idling

– EWM + ESS + seized blade

Identify output parameters:– typical wind turbine parameters

– platform displacements

– mooring line tensions

Spar Concept by SWAY

Dynamics Modeling and Loads Analysis of an Offshore ...

Documents

Transcript of Dynamics Modeling and Loads Analysis of an Offshore ...