Time-domain numerical and segmented ship model experimental analyses of hydroelastic responses of a large container ship in oblique regular waves

Real sea conditions are characterized by multidirectional sea waves. However, the prediction of hull load responses in oblique waves is a difficult problem due to numeral divergence. This paper focuses on the investigation of numerical and experimental methods of load responses of ultra-large vessels in oblique regular waves. A three dimensional nonlinear hydroelastic method is proposed. In order to numerically solve the divergence problem of time-domain motion equations in oblique waves, a proportional, integral and derivative (PID) autopilot model is applied. A tank model measurement methodology is used to conduct experiments for hydroelastic responses of a large container ship in oblique regular waves. To implement the tests, a segmented ship model and oblique wave testing system are designed and assembled. Then a series of tests corresponding to various wave headings are carried out to investigate the vibrational characteristics of the model. Finally, time-domain numerical simulations of the ship are carried out. The numerical analysis results by the presented method show good agreement with experimental results.     

Temporal variation of modulated-wave-train geometries and their influence on vertical bending moments of a container ship

A towing experiment was conducted using a modulated wave train to investigate the vertical bending responses of a hydro-structural container ship model. In the experiment, a spatially periodic modulated wave train, as a model of a freak wave in successive high waves mimicking the so-called three sisters, was generated by the recently established higher-order spectral method wave generation (HOSM-WG) method. HOSM-WG enables us to control the location and timing of the maximum crest height in a wave tank. With precise control of the towing carriage, an experiment was conducted in which the timing of the encounters between the ship model and the modulated wave train was accurately determined. The maximum sagging moment (SM) was found to increase in proportion with the encounter wave height. However, because of differences in the relative depth of the fore and aft troughs, the maximum SM is highly variable for a given wave height. The temporal wave-geometry evolution caused the relative trough-depth to vary significantly within a wave period in the vicinity of the maximum crest height. As a result, depending on the encounter timing, the SM varied considerably for a given wave height. The temporal variation of the wave geometry is a robust feature of a modulated wave train and is common between the spatially periodic and temporally periodic modulated wave trains.     

Experimental investigation of hull girder vibrations of a flexible backbone model in bending and torsion

Although the coupled horizontal–torsional vibrations of open ships have been investigated numerically for decades, the available experimental data in oblique seas seem rare. Model tests, considering natural frequencies of bending and torsional modes, have been conducted by the Centre for Ships and Ocean Structures (CeSOS) in the towing tank and ocean basin. A flexible backbone model was designed with five cut-outs on the top side of the aluminum beam to realize approximately torsional stiffness as well as vertical and horizontal bending stiffness. This paper mainly deals with measured bending and torsional vibrations in regular and irregular waves. The damping ratios, mode shapes, and modal moments were derived from experimental data, and a numerical model based on modal superimposition is established according to the measured hydroelastic properties. Some time-domain simulations are carried out considering the structural characteristics, and compared with measured results. The test data in regular waves are presented to show the possible factors of influence on the vibrations. The effect of bending and torsional vibrations on the extreme response values in irregular waves is estimated. The uncertainties in the experiments are discussed and conclusions are presented at the end of this paper.