TY - GEN

T1 - Evolutionary algorithms for the development and optimisation of wave energy converter control systems.

AU - Gunn, K.

AU - Taylor, C. James

AU - Lingwood, C.

PY - 2009

Y1 - 2009

N2 - Many strategies have been proposed for the control of wave energy converters (WECs). In order to evaluate these control strategies, they need to be optimised for realistic operating conditions. This paper develops a generic approach for WEC optimisation based on the use of Evolutionary Algorithms (EAs). A new evolutionary algorithm is developed to efficiently resolve problems found in WEC control. Simulation results are presented for tuning an illustrative device in both sinusoidal and real waves; and for optimisation of slow tuning, latching and fast tuning control systems. These results show an increase in the power capture of the device using the optimised control, and demonstrate a convergence to an optimum solution within the constraints presented. In contrast to conventional methods, the proposed EA successfully optimises the control algorithms for realistic seas without prior assumptions. The capabilities of EAs in a machine learning setting, in which the control algorithm continues to evolve after installation, are then considered.

AB - Many strategies have been proposed for the control of wave energy converters (WECs). In order to evaluate these control strategies, they need to be optimised for realistic operating conditions. This paper develops a generic approach for WEC optimisation based on the use of Evolutionary Algorithms (EAs). A new evolutionary algorithm is developed to efficiently resolve problems found in WEC control. Simulation results are presented for tuning an illustrative device in both sinusoidal and real waves; and for optimisation of slow tuning, latching and fast tuning control systems. These results show an increase in the power capture of the device using the optimised control, and demonstrate a convergence to an optimum solution within the constraints presented. In contrast to conventional methods, the proposed EA successfully optimises the control algorithms for realistic seas without prior assumptions. The capabilities of EAs in a machine learning setting, in which the control algorithm continues to evolve after installation, are then considered.

KW - Wave Energy Converters

KW - Optimisation

KW - Evolutionary Algorithm

KW - Latching

KW - Control

M3 - Conference contribution/Paper

BT - 8th European Wave and Tidal Energy Conference

T2 - 8th European Wave and Tidal Energy Conference

Y2 - 7 September 2009 through 10 September 2009

ER -