Back to Search Start Over

Towards realistic non-linear receding-horizon spectral control of wave energy converters.

Authors :
Mérigaud, Alexis
Ringwood, John V.
Source :
Control Engineering Practice. Dec2018, Vol. 81, p145-161. 17p.
Publication Year :
2018

Abstract

Abstract Non-linear, power-maximising control of wave energy converters (WECs) can be achieved within a receding-horizon control framework, whereby an upper loop calculates a reference trajectory in real-time, ensuring maximal power absorption under operational constraints, while a tracking loop drives the device along the generated trajectory. This paper articulates the four fundamental components of such a control strategy: reference generation calculations, tracking loop , and wave excitation estimation and forecasting. The upper-loop optimisation problem is efficiently solved through a Fourier spectral method, taking into account non-linear dynamics and constraints. Tracking is achieved through a linear state feedback, combined with a non-linear feed-forward term. An extended Kalman filter is used for excitation force estimation, based on noisy WEC position and acceleration measurements. Finally, wave excitation forecasts are based on a linear predictor, whose coefficients are derived from the wave spectrum (on a sea-state-by-sea-state basis). The practical issues and trade-offs, which arise when the four components listed above are combined within a practical implementation, are investigated by means of realistic numerical simulations, using a WEC model comprising a combination of static and velocity-dependent non-linear forces. Highlights • Receding-horizon optimal control of wave energy converters (WECs) is studied. • The reference trajectory is updated in real time using Fourier spectral control. • Trajectory tracking is ensured using combined feed-forward and feedback terms. • The only measurements are the WEC position and acceleration, with noisy sensors. • Wave estimation and forecasting are solely based on available measurements. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
09670661
Volume :
81
Database :
Academic Search Index
Journal :
Control Engineering Practice
Publication Type :
Academic Journal
Accession number :
132755845
Full Text :
https://doi.org/10.1016/j.conengprac.2018.08.024