Further Discussion on the White-Noise Approach for Predicting Slow-Drifts in FAST

Making use of theoretical approximations for the computation of the wave-induced slow-drift forces is a common procedure in the early stages of design of a new floating unit. They can help reducing the computational burden in two different fronts: for generating the QTFs in a frequency domain analysis, and during the subsequent execution of time-domain simulations. In a previous paper, we have discussed a simple procedure for making use of the white-noise approximation in FAST, without the need for any modification of the software. The proposal only requires restricting the computation of the QTFs to pairs of frequencies that are indeed essential to the slow-drift dynamics. For this, however, an additional assumption is made, considering that each motion is decoupled from those in the other dofs. In the present paper, a more detailed analysis of the subject is made, in order to clarify the theoretical aspects of the procedure and supplement the previous analysis. Once again, the results are based on the data available for the OC4 FOWT. The accuracy obtained with the procedure is discussed not only in terms of the resulting motions, but also comparing its effects on the second-order force spectra. A more detailed evaluation of the dynamic couplings is presented, and comparisons with the results obtained with Newman’s approximation are made in simulations involving waves only.