Search

An Innovative Semi-submersible platform Optimization Program (ISOP) has been developed to solve the multi-objective optimization problem for semi-submersible platforms (SEMI). Three types of SEMIs, including semi-submersible floating production unit (SEMI FPU), heave and vortex induced motion (VIM) suppressed semi-submersible (HVS) and semi-submersible floating drilling unit (SEMI FDU) are selected for case studies. The hydrodynamic performances of three types of semi-submersible platforms are analyzed by using panel method and Morison's equation. In order to improve the computing efficiency, the hydrodynamic performances for different hull forms during optimization process are estimated by the surrogate models, which are built by artificial neural network prediction method and Inverse Multi-Quadric (IMQ) radial basis function (RBF). The accuracy of surrogate models is ensured by performing leave-one-out cross validation (LOOCV). The most probable maximum (MPM) heave motion and total weight, representing the safety and economy, respectively, are chosen as the two objectives for optimization. The transverse metacentric height, the MPM surge motion, and the most probable minimum (MPMin) airgap are selected as constraints. Based on surrogate models, multi-objective particle swarm optimization (MOPSO) is employed to search for the Pareto-optimal solutions. A Computational Fluid Dynamics (CFD) tool is adopted to validate the proposed model for the prediction of the motion responses. By comparing the obtained Pareto-optimal solutions with the initial design using simple panel method plus Morison's equation, it is confirmed that the MPM heave motions for SEMI FPU, HVS and SEMI FDU can be suppressed by up to 12.68%, 11.92%, and 14.96%, respectively, and the total weights can be reduced by up to 12.16%, 13.00%, and 24.91%, respectively. Through the detailed analyses of optimization results, the most efficient design strategies for semi-submersible platforms are discussed and proposed.

Due to the existence of the side walls in a towing tank, the measured hydrodynamic forces would present some discrepancies compared to the open sea results. This phenomenon is referred to as the side wall effect. The objective of the present study is to investigate the side wall effects on ship model testing in a towing tank. The method used in the present study involves a 3D panel method based on the Rankine type Green function. Both the steady and unsteady problems were investigated numerically. The numerical results were validated against ship model test results. After the validations, a large scale computations were performed to investigate the parameters which could determine the side wall effects. Two diagrams of side wall effects (one in calm water and the other one in waves), were obtained which showed whether the side wall effect was less than the permissible error to be included in the measured values.

The paper presents the study on hydro-elastic responses of a floating raft with a mooring system in both regular and irregular waves. The floating raft consists of a number of buoys, plates and a mooring system, which are modeled by Morison type elements with proper coefficients, segmented line elements with stiffness and several mooring cables, respectively. The added mass coefficients of the buoys are computed by using the panel code WAMIT. Damping coefficients are obtained by CFD simulations of the forced surge/sway motions of a buoy with a free surface. Due to the nonlinear performance of the mooring system, the periods of maximum responses of the structure decrease as the wave height increases. For regular waves, the most of dangerous wave conditions are found in the wave periods 6 ∼ 10 s with the incident wave angles 0 ∘ ∼ 2 0 ∘ or 7 0 ∘ ∼ 9 0 ∘ , while the floating raft is relatively safe for the incident wave angle being around 45°. For irregular waves, the impacts of randomness are analyzed by using different wave seeds. Most of the results are not sensitive to the randomness of the phase angles, except for the maximum mooring tensions which show noticeable fluctuations ( ∼ 15 % ). By comparing the results of the long-crested wave and short-crested wave, it is found that the predictions of the mooring tension and the responses of the floating raft are more conservative in long-crested wave than those in short-crested wave. Besides, the impacts of different frequency ranges on the time histories of the responses are discussed. The maximum structural responses are compared with the most probable maximum (MPM) values obtained by assuming Rayleigh distribution for the responses. The most vulnerable locations of the structure are analyzed based on the maximum dynamic loads on the plates of the floating raft. When the zero-crossing period exceeds 5 s, the relatively large tensions on the plates occur in the north and south sides of the floating raft. The relatively large shear force, bending moment, normal stress, torque and shear stresses on the plates always occur in the four corners of the floating raft.

This paper presents numerical studies on vortex-induced motions (VIM) of a multi-column floating oil and gas exploration platform. Numerical computations are performed using an improved delayed detached eddy simulation (IDDES) together with a moving grid approach. The transverse (sway) and yaw motion responses, motion trajectories, motion frequencies and power spectral density of motions are computed and analyzed systematically. After extensive comparisons with experiments, it is confirmed that the present numerical solutions using IDDES agree well with the experimental results and are better than those via delayed detached eddy simulation (DDES). The differences of transverse motion responses between computational results and experiments are less than 10% in the lock-in region. The numerical simulations reveal that the transverse VIM responses occur in a range of reduced velocities from 7.0 to 14.0 at H / D = 1.44 ( H and D are the column height and width, respectively). The largest nominal transverse amplitude, around 35 % of the column width, occurs for 22 . 5 ∘ current incidence. It is found that the VIM responses mainly perform along the platform diagonals for 1 5 ∘ , 22 . 5 ∘ and 4 5 ∘ current incidences. The transverse and yaw motion frequencies for 1 5 ∘ , 22 . 5 ∘ and 4 5 ∘ current incidences are higher than those for 0 ∘ current incidence. The energy levels of the yaw motion responses for 1 5 ∘ , 22 . 5 ∘ and 4 5 ∘ current incidences are about 10% of that for 0 ∘ current incidence. Moreover, parametric studies have been performed to examine the effect of submerged column height on VIM. It demonstrates that VIM in transverse direction grows significantly when submerged column height H / D is greater than 1.0. To be more specific, compared to the case with H / D = 1.44, VIM in transverse direction at H / D = 3.0 increase by around 120%.

In this paper, a boundary element programme MHydro, which is based on 3-D Rankine source method, was developed to investigate the ship-to-ship interaction problem. The method developed considers both stationary and forward speed cases. A new radiation condition, which takes Doppler shift into account, is imposed on the control surface. For the stationary case, the validations were carried out both in head sea and beam sea conditions. The experimental measurements, as well as the published numerical results, were used for the comparison. For the ship-to-ship with forward speed case, we validated our method through the experimental tests. Good agreement was achieved except the roll motion. The hydrodynamic coefficients and wave excitation forces were investigated and a very large sway force was predicted when the transverse distance between two ships equalled to the wave length. The wave elevations in the gap were also calculated. Discussions are highlighted on the shallow water effects.

The hydrodynamic interactions between the vessel motion and resonant wave response inside the three-dimensional moonpool are studied in the present paper. In particular, the shift of natural frequencies of a three-dimensional moonpool in the fixed and free-floating conditions is computed and examined systematically. Barges with different moonpool dimensions are analyzed to examine how the moonpool configurations affect the shift of resonant frequencies. Detailed analyses are performed for cases with and without recess. Numerical computations are performed for regular waves with wave headings θ = 0 ∘ , 45 ∘ and 90 ∘ . Free-surface elevations inside the moonpool with respect to the incident wave frequencies are computed for both fixed and free-floating conditions. It is observed that natural frequencies of the moonpool in the free-floating condition are higher than those in the fixed condition. For cases without recess, the shift of resonant frequency is more prominent in piston mode resonance, in particular for shallow or long moonpools. In contrast, for cases with a recess, the shift of resonant frequency in the first sloshing mode is relatively larger than that in piston mode. In addition, the effects of motion RAOs on the free-surface elevation inside the moonpool in the free-floating condition are also examined. It is found that the moonpool resonance can induce a local hump in heave and pitch responses. Moreover, through deriving a modified frozen mode approximation model, we reveal that, comparing to the fixed condition, the vessel motion in the free-floating condition yields a reduction in the added mass due to the fluid underneath the moonpool so that the piston mode resonant frequency increases.

The flows around two tandem circular cylinders in the subcritical and supercritical flow regimes are studied through three-dimensional numerical simulations. Different spacing ratios L / D (where L is the center-to-center distance between the two cylinders with D being the diameter of the cylinders) from 2.0 to 5.0 is considered. The instantaneous flow structures, pressure distributions and hydrodynamic forces on two tandem cylinders are analyzed at subcritical ( R e = 2.2 × 10 4 ) and supercritical ( R e = 3.0 × 10 6 ) Reynolds numbers. The present numerical solutions demonstrated that at the two R e , for L / D 3.5 , the shear layer from the upstream cylinder reattaches on the surface of the downstream cylinder and vortex street is only formed behind the downstream cylinder, while the shear layer from the upstream cylinder rolls up alternately for L / D ≥ 3.5 , and vortex shedding occurs from both upstream and downstream cylinders. Compared with the flow characteristics at R e = 2.2 × 10 4 , for R e = 3.0 × 10 6 , the flow separation positions on the upstream cylinder move backward and the width of the wake behind the upstream cylinder becomes narrower, which leads to forward movement of the reattachment position. At the supercritical Reynolds number, the mean drag and fluctuating lift coefficients of the upstream cylinder are nearly independent of L / D while the sharp increase of force coefficients of the downstream cylinder occurs for L / D = 3.5 . Comparing to the case at the subcritical Reynolds number, higher vortex-shedding frequencies from the upstream and downstream cylinders are identified. It is also revealed that the slope of the best fit line for φ (the phase lag of the fluctuating lift force between two cylinders) and L / D increases at the supercritical Reynolds number. In addition, for L / D = 2.5 , the mean drag coefficients of the upstream cylinder decrease sharply for R e > 1.0 × 10 5 , while those of the downstream cylinder rise slightly with the growing R e . Moreover, the fluctuating lift coefficients of the upstream cylinder are almost independent of R e , but those of downstream cylinder drop as R e increases.