VIV of slender beams at high mode order will appear as a non-stationary response process. Amplitudes, dominating frequency and mode composition are seen to vary in time, but so far we do not have a complete understanding of this process. One approach is to attempt to understand and model the physical mechanism behind the observed behaviour, but an alterative is to establish a simplified model that can be used for fatigue calculations. The purpose of the present paper is to compare the dominant response frequencies that have been observed in large scale tests of a flexible beam to the discrete response frequencies that are predicted by an empirical model for prediction of VIV. Use of wavelet and modal analyses on experimental data makes it possible to describe the time variation of the peak frequency, and hence also the relative period of time this frequency falls into discrete frequency bins. An empirical model is also proposed for calculation of relative duration of competing response frequencies. The observed frequencies can hence be compared to the results from the proposed model. The conclusion is that the model identifies the domination response frequencies with satisfactory accuracy. The range of calculated discrete response frequencies is larger than for the discrete peak frequencies identified from the experiments. But the observed response process has a broader frequency band than the variation of the peaks. Hence, the results from the analysis seem to agree well with observations. Further analyses are, however, still needed in order to verify the proposed model.

You do not currently have access to this content.