This paper presents a nonlinear frequency domain model and uses this to assess the performance of a wave energy converter (WEC) array with a nonlinear power take-off (PTO). In this model, the nonlinear PTO forces are approximated by a truncated Fourier series, while the dynamics of the WEC array are described by a set of linear motion equations in the frequency domain, and the hydrodynamic coefficients are obtained with the boundary element method. A single heave absorber is firstly investigated to establish the accuracy of the new model in capturing the nonlinear behavior of the pumping system. Subsequently, simulations of a 2D array with 18 WECs and a pillar in the center (representing the tower of a wind turbine) are carried out to understand wave interference effects. Several optimization strategies are proposed to improve the overall performance of the WEC array. These results demonstrate a computationally effective method for accounting for nonlinear effects in large WEC arrays. The proposed approach may potentially be applied for developing control algorithms for the adaptability of a 2D array to incoming wave excitation.

A mixed time-domain/frequency-domain method is proposed for modelling dense wave energy converter (WEC) arrays with non-linear power take-off (PTO). The model is based on a harmonic balance method which describes the system response in the frequency domain, while evaluating the non-linear PTO force and solving the system equations of motion in the time domain. The non-linear PTO force is computed with Lagrange multipliers. In order to apply the proposed method for WEC array responses in real sea states, the time series is split into time windows and the simulation is carried out individually per window. The method is demonstrated by investigating the dynamics of the Ocean Grazer WEC array (OG-WEC) with an adaptable piston pumping system. The key parameters thought to possibly influence model accuracy, including the number of harmonic components, the length of the time window and overlay, are discussed. It is shown that the proposed model can significantly reduce the computational cost with an acceptable accuracy penalty.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

No abstract

No abstract