consequences of load mitigation control strategies for a floating … · 2020. 1. 22. · 3p....

17
Consequences of load mitigation control strategies for a floating wind turbine Chern Fong Lee, NTNU Erin E. Bachynski, NTNU ([email protected] ) Amir R. Nejad, NTNU

Upload: others

Post on 07-Feb-2021

4 views

Category:

Documents


0 download

TRANSCRIPT

  • Consequences of load mitigation control strategies for a floating wind turbine

    Chern Fong Lee, NTNUErin E. Bachynski, NTNU ([email protected]) Amir R. Nejad, NTNU

    mailto:[email protected]

  • 2

    Control-induced resonance

    1. Platform pitches forward2. Nacelle sees an increase in

    relative wind speed3. Controller regulates blade

    pitch to reduce rotor speed4. Thrust is reduced

  • 3

    Load-mitigation control strategies for FWTs

    • AD: Nacelle velocity feedback (added damping)– Lackner, 2007– Modify rotor speed reference with nacelle velocity measurement

    𝐻𝐻𝑃𝑃𝑃𝑃∆Ω𝐴𝐴𝐴𝐴

    Ω𝑟𝑟

    +

    -

    𝐻𝐻𝑊𝑊𝑊𝑊

    ∆𝜃𝜃+

    K�̇�𝑥

    Ω0

  • 4

    Load-mitigation control strategies for FWTs

    • ES: Energy shaping controller– Pedersen, 2017– Modify rotor speed reference using the deviation of nacelle

    velocity from its value in equilibrium

    +

    D𝛿𝛿�̇�𝑥 𝐾𝐾𝐸𝐸𝐸𝐸

    𝐻𝐻𝑃𝑃𝑃𝑃Ω0∆Ω𝐸𝐸𝐸𝐸

    Ω𝑟𝑟

    +

    -

    𝐻𝐻𝑊𝑊𝑊𝑊

    ∆𝜃𝜃IPC

  • 5

    Load-mitigation control strategies for FWTs

    • AD: Nacelle velocity feedback (added damping)– Lackner, 2007– Modify rotor speed reference with nacelle velocity measurement

    • ES w/o IPC: Energy shaping controller– Pedersen, 2017

    • ES w/IPC: Energy shaping controller with IPC– Try to reduce individual blade root bending moments– IPC follows Lackner and van Kuik, 2009

  • 6

    Known consequences of load-mitigating control strategies• AD: reduction in pitch motion, increased variations in

    power and rotor speed• ES: stable control, expected reductions in pitch motions• IPC: reduce blade root bending moments, increase pitch

    actuator use

    What about the drivetrain?

    Image: Nejad et al., 2016

  • 7

    Outline

    • Methodology• Global analysis results• Drivetrain loads• Conclusions

  • 8

    Methodology: Decoupled simulations

    Image: Nejad et al., 2016

    Global analysis: SIMA Drivetrain analysis: SIMPACK

    EC 1 EC 2 EC 3

    Significant wave height, Hs [m] 5.0 4.0 5.5

    Peak period, Tp [s] 12.0 10.0 14.0

    Mean wind speed, U [m/s] 12.0 14.0 20.0

    Turbulence intensity, I [-] 0.15 0.14 0.12

    6x1hr1x1hr

  • 9

    Performance indicators

    • Tower base 1-hr fatigue damage– Stresses from global analysis, rainflow counting, SN curve,

    Miner’s rule• Gear root 1-hr fatigue damage

    – Forces from MBS analysis, load duration distribution method• Bearing 1-hr fatigue damage

    – Forces from MBS analysis, load duration distribution method• Standard deviation of power output

    – Direct result from global analysis

  • 10

    Global motions, EC1

    0 0.2 0.4 0.6 0.8

    Frequency [rad/s]

    0

    50

    100

    150

    200

    250

    300

    350

    400

    S() [

    m2

    s/ra

    d]

    Baseline

    ES w/o IPC

    ES w IPC

    AD

    0 0.2 0.4 0.6 0.8 1 1.2

    Frequency [rad/s]

    0

    10

    20

    30

    40

    50

    60

    S()[r

    ad2

    s/ra

    d]

    0.3 0.4 0.5 0.6

    0

    1

    2

    0.3 0.4 0.5 0.6

    Frequency [rad/s]

    0

    10

    Surge Pitchsurge

    pitch

    wave

    pitch

    wave

  • 11

    Tower base fore-aft bending moments

    wave

    tower 3p

  • 12

    Gearbox topology

    Image: Nejad et al., 2016

  • 13

    Sun gear circumferential force

    wave

    1P

    3P

    tower

  • 14

    Tower top side-side force

    tower

    wave 1P

    pitch

  • 15

    Bearing INPB

    axial

    radial1P

  • 16

  • 17

    Conclusions• Global and drivetrain responses of a spar floating wind turbine• Three control modifications

    – active damping (AD)– energy shaping control (ES w/o IPC),– energy shaping control with individual blade pitch (ES w/IPC).

    • Improved platform motion responses in surge and pitch• ES adds some responses at i.e. wave frequency • IPC reduces blade root flap-wise bending, but introduces

    excitation of tower top shear force at rotor frequency.• The reduced blade root moment therefore comes with a cost

    of increased radial load resonance in drivetrain gears and bearings.

    • Drivetrain should be considered when assessing control performance

    Consequences of load mitigation control strategies for a floating wind turbineControl-induced resonanceLoad-mitigation control strategies for FWTsLoad-mitigation control strategies for FWTsLoad-mitigation control strategies for FWTsKnown consequences of load-mitigating control strategiesOutlineMethodology: Decoupled simulationsPerformance indicatorsGlobal motions, EC1Tower base fore-aft bending momentsGearbox topologySun gear circumferential forceTower top side-side forceBearing INPBSlide Number 16Conclusions