Comparative Accuracy of Numerical Kelvin Wake Code Predictions - 'Wake Off'
Abstract
Wave patterns predicted by several numerical codes were evaluated by comparing them with data from model basin experiments on two ship hulls at three Froude number each. In general, the codes all underpredict the amplitude of divergent waves springing from the ships' bow. High wave number detail is also lacking in the vicinity of the bow wave cusp line compared to model free-surface wave patterns. Conversely, the codes tend to overpredict the amplitudes of wave close to the ships' stern and in the transverse wave system behind. The codes are ranked according to how closely they stimulate the empirical results. Within the ranking, adjacent codes give similar wave predictions and the rank might be interchanged for a few Froude number cases. Higher ranked codes gave consistently better predictions than the lower ranked codes. Problems for the lower ranked codes included: excessive wave damping such that waves are attenuated near the edges of the computational domain; reflections appear at the outer boundaries; high frequency noise exists beyond the 19-deg envelope of the spreading wave train; and wave energy has been severely underpredicted. Standardized graphic representations of the numerical and experimental data are included in the appendixes. The predictions were made blind, without prior knowledge of the specific experimental results. In view of the inherent difficulties including nonlinear wave breaking and real fluid effects (present in the experiments), the agreement achieved by the better codes is encouraging.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1991
- Accession Number
- ADA241076
Entities
People
- Arthur M. Reed
- Toby J. Ratcliffe
- William Lindenmuth