Search

A model for a oscillating flap wave energy converter (WEC) is as a single degree of freedom system with a non-linear term to allow for the drag of the device through the water, known as the Morison term. The focus of this system identification is on estimating the dynamic state of the system and estimating the non-linear parameter from observations of the wave elevation and the vertical displacement of the device. It is assumed that the mass, stiffness and damping of the system, without the Morison term, are known from the physical characteristics of the device. The Kalman Filter (KF) can be used to estimate the states of a linear system, however, it is not directly applicable to a non-linear system. Various adaptations have been proposed for non-linear systems. One of the first was the extended Kalman Filter (EKF) which relied on a linearization about the current state values. However, an alternative approach, known as the unscented Kalman Filter (UKF) has been found to give a better performance and is easier to implement. We apply the UKF to estimate the dynamic states of the system together with the non-linear parameter. The fitted model can be used to predict the performance of the device in different wave environments. [ABSTRACT FROM AUTHOR]

Waves offer an attractive source of renewable energy, with relatively low environmental impact, for communities reasonably close to the sea. Two types of simple wave energy converters (WEC), the heaving buoy WEC and the oscillating flap WEC, are studied. Both WECs are considered as simple energy converters because they can be modelled, to a first approximation, as single degree of freedom linear dynamic systems. In this study, we estimate the response of both WECs to typical wave inputs; wave height for the buoy and corresponding wave surge for the flap, using spectral methods. A nonlinear model of the oscillating flap WEC that includes the drag force, modelled by the Morison equation is also considered. The response to a surge input is estimated by discrete time simulation (DTS), using central difference approximations to derivatives. This is compared with the response of the linear model obtained by DTS and also validated using the spectral method. Bendat's nonlinear system identification (BNLSI) technique was used to analyze the nonlinear dynamic system since the spectral analysis was only suitable for linear dynamic system. The effects of including the nonlinear term are quantified. [ABSTRACT FROM AUTHOR]