Loughborough University
Browse

LQG control for hydrodynamic compensation on large floating wind turbines

Download (1.42 MB)
journal contribution
posted on 2023-01-24, 16:35 authored by Qusay Hawari, Taeseong Kim, Christopher Ward, James FlemingJames Fleming

This work proposes a novel Linear Quadratic Gaussian (LQG)-based blade pitch control method for floating offshore wind turbines, in which a state-space model of the turbine and water hydrodynamics is included in the LQG design. The actuation considered is collective blade pitch control with the objective of generator power stabilization and platform motion reduction. A linear Kalman filter is used to estimate un-measurable states relating to wave excitation and radiation through measurements of generator speed, platform pitch, and wind disturbance. Controller design models were validated with the full order nonlinear model under various testing conditions. The new controller design is tested on a nonlinear high-fidelity simulation model of the 15 Mega-Watt (MW) floating semi-submersible wind turbine. In simulations with realistic stochastic wind and wave disturbances, the new controller achieves 32% lower generator speed Root Mean Square Error (RMSE) and 16% lower platform pitch RMSE compared to a standard LQG controller that does not include hydrodynamic states, for equivalent levels of pitch actuation and with a 2° /sec rate limit on pitch. The inclusion of hydrodynamics in the controller design not only reduced platform pitching fluctuation, but also had a strong effect of hub-height factors such as the generator speed.

Funding

Loughborough University

History

School

  • Mechanical, Electrical and Manufacturing Engineering

Published in

Renewable Energy

Volume

205

Pages

1 - 9

Publisher

Elsevier

Version

  • VoR (Version of Record)

Rights holder

© The Authors

Publisher statement

This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Acceptance date

2023-01-18

Publication date

2023-01-19

Copyright date

2023

ISSN

0960-1481

eISSN

1879-0682

Language

  • en

Depositor

Dr James Fleming. Deposit date: 23 January 2023

Usage metrics

    Loughborough Publications

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC