ABSTRACT

There is a growing interest in impact loads by extreme waves which can be found in DNV-GL's new guideline OTG-14 for horizontal impact loads of column stabilized units. Up to now, the estimation of extreme wave impact loads has been mostly performed through model tests which still have many uncertainties especially in measuring the impact pressure or load. Moreover, high cost and long program of model test seems to be needed because many realizations with seed variations are required to estimate impact pressure level properly. Therefore, many attempts have been made to solve this problem by numerical analysis based on CFD as an alternative to the model test. However, the reliability of CFD simulation result for the extreme wave generation and impact load estimation has not been established properly yet.

Recently, as a benchmark study, KRISO(Korea Research Institute of Ships and Ocean Engineering) has conducted a series of experimental study on impact loads acting on a fixed cylinder with three different types of focusing wave; plunging, spilling and steep at a 2D wave flume.

In the present paper, based on the results of the model test performed by KRISO, the realization of focusing waves and associated impact loads on a fixed cylinder was investigated using RANS based CFD. In the CFD simulation, a piston-type numerical wavemaker using the mesh deformation method was applied. The motion of wavemaker was forced to reproduce the time series of the physical wavemker measured in the model tests. To obtain the proper focusing waves, a numerical calibration with a series of mesh and time step size was performed and the quality of the waves was investigated by the wave characteristics in terms of wave crest and focused wave rise time. With the calibrated focusing waves the CFD simulations for wave impact loads on the fixed cylinder were carried out and the results were also compared with the model test results.

This content is only available via PDF.
You can access this article if you purchase or spend a download.