General data: Frequency domain

Solution frequencies

The frequency domain solver can be used to solve at the following solution frequencies:

Wave frequency: solves for the response of the linearised system at each wave component frequency.
Low frequency: solves for the response of the linearised system at wave drift frequencies and wind component frequencies.
Low frequency, combined linearisation: initially conducts a wave frequency solve and then, subsequently, a low frequency solve. The initial wave frequency solve is conducted purely to account for the coupling across the response frequencies when linearising quadratic viscous drag. This will, for instance, capture the increase in drag, at low frequency, on mooring lines due to the presence of wave frequency motion. The results of the initial wave frequency solve are not made available. The reported results are for the low frequency solve.

Note:

In the rest of the documentation when we refer to low frequency solves we mean both of the low frequency options above.

Solution parameters

Maximum number of iterations

The frequency domain solver uses an iterative method to linearise the quadratic viscous drag load. The calculation is abandoned if convergence has not been achieved after this number of iterations.

Tolerance

A non-dimensional value used by the frequency domain solver to determine whether the drag linearisation has converged. Increasing this value can result in reduced computation time since fewer iterations are needed to find the linearised drag load. However, doing so may also result in inaccurate results.

Convergence damping

A non-dimensional factor used to damp the convergence of the solution. If we denote by $x_i^*$ the 'pure' theoretical solution from the frequency domain solver, and $x_{i-1}$ the (damped) solution from the previous iteration, then the convergence damping $d$ is applied to obtain the damped solution for the present iteration, $x_i$, as \begin{equation} x_i = (1-d) x_i^* + d x_{i-1} \end{equation} If $d{=}0$, no convergence damping is applied. Increasing values correspond to increasing levels of convergence damping, up until a maximum value of 0.5.

Depending on the model, choosing higher or lower levels of convergence damping can reduce the number of iterations required before a converged solution is found.

Numerical integration sub-interval counts

Only required when low frequency solution frequencies are being used. Specifies the number of sub-intervals used when conducting numerical integration to calculate the wave drift load process and the wave drift damping matrix.

Note:

The wave drift load sub-interval count can have a significant impact on the calculation time, and you may find that reducing it speeds up your simulation. Should you do so, you may find however that the accuracy of the solution is reduced: we recommend that you perform sensitivity studies to confirm the accuracy of your results.