Showing papers in "International shipbuilding progress in 2005"

Improved time domain simulation of planing hulls in waves by correction of the near-transom lift

Abstract: Simulation of the planing hull in waves has been addressed during the last 25 years and basically been approached by strip methods. This work follows that tradition and describes a time-domain strip model for simulation of the planing hull in waves. The actual fluid mechanical problem is simplified through the strip approach. The load distribution acting on the hull is approximated by determining the section load at a number of hull sections, strips. The section-wise 2-dimensional calculations are expressed in terms of added mass coefficients and used in the formulations of both inertia and excitation forces in the equations of motions. The modeling approach starts from the hypothetic assumption that the transient conditions can be modeled based on those section-wise calculations. The equation of motion is solved in the time-domain. The equation is up-dated at each time step and every iteration step with respect to the momentary distribution of section draught and relative incident velocity between the hull and water and catches the characteristic non-linear behavior of the planing craft in waves. The model follows the principles of the pioneering work of E. E. Zarnick differing on model structure and in details such as the modeling of the lift in the transom area. A major part of the work is concerned with experiments and evaluation of simulations with respect to performed model tests and to published experiment data. Simulations of model tests have been performed and comparisons have been made between measured and simulated time series. The link between simulation and experiment is a wave model which is based on a wave height measurement signal. It is developed and evaluated in the thesis. The conclusions are in favor of the 2-dimensional approach to modeling the conditions for the planing hull in waves and among further studies is evaluation of simulated loads and motions to full-scale trial measurement data.

Parametrically excited roll in regular and irregular head seas

Abstract: A method is proposed to deal with dynamic stability of ships in head seas based on a time-domain, non-linear numerical model of ship's motions in 5 degrees-of-freedom (sway, heave, roll, pitch, and yaw). A quasi-static approach is adopted for heave, roll, and pitch motions in waves, in which the calculation of significant variations on restoring coefficients is made using a pressure integration technique over the instantaneous submerged hull. Dynamic and hydrodynamic effects in waves for a given encounter frequency are included in the response calculations, which are based on a strip theory. Comparisons between numerical and experimental results demonstrated the adequacy of the technique proposed to predict parametric roll in regular waves. The results demonstrated also that the method is capable of predicting wave-induced parametric resonance not only in regular but also in irregular waves. Finally, in the stochastic excitation a certain threshold value is assumed for parametric roll response and then the probabilities of exceedance are calculated as a function of the prevailing weather and ship's operational conditions.

A frequency domain system identification method for linear ship maneuvering

Abstract: A new frequency domain system identification method for estimation of hydrodynamic derivatives embedded in linear steering equations for ship maneuvering in calm seas is presented. The frequency domain multiple input-single output models developed for identification involves determination of constant, 'zero-frequency' hydrodynamic derivatives. The method is robust, non-iterative and computationally light and it does not require any starting estimates. In this method, the time domain operations are converted to linear operations in the frequency domain. The responses of the ship in a few standard maneuvers have been simulated in the numerical examples and the proposed method is applied to this data in order to estimate the hydrodynamic derivatives for all possible 'identifiable' combinations.

Comparison between experimental and numerical results of the nonlinear vertical ship motions and loads on a containership in regular waves

Abstract: The paper presents a comparison between experimental data and numerical results of the nonlinear wave induced heave and pitch motions and structural vertical shear forces and bending moments on a containership. The experimental data was obtained from tests with a model of a containership advancing in head regular waves with different steepness, ranging from small to large amplitude waves. Strong nonlinear effects were detected on the experimental data, identified by: the variation of the transfer functions with the wave amplitude, the existence of higher harmonics and the asymmetry of the response records with respect to the positive and negative peaks. The experimental data is compared with numerical results from a nonlinear time domain strip method. The method assumes that the radiation and diffraction hydrodynamic forces are linear and the nonlinear contributions arise from the hydrostatics and Froude-Krilov forces and the effects of green water on deck. The effects of vertical viscous forces due to flow separation are also investigated. The associated forces are represented by a semi-empirical formulation and depend of cross flow drag and lilt coefficients. Nonlinear vertical motions and loads on ships advancing in waves have been investigated before, both experimentally and numerically however, the present paper focuses on several new aspects of the nonlinear vertical structural loads in regular waves. The new aspects include the identification and quantification of third harmonic amplitudes, the influence of the regular wave amplitude on the amplitudes of the first three harmonics and on the sagging and hogging peaks, and the influence of the steady structural loads, that exist also with the ship advancing in calm water, on the asymmetry of the vertical loads in waves. Finally, the influence of viscous forces associated with the large amplitude motions on the wave induced motions and structural loads are also investigated.

Impact Pressure Distribution Reconstruction from Discrete Point Measurements

Abstract: The paper presents a method to reconstruct the momentary pressure distribution in hull-water impacts. Measurements of the propagating pressure pulse in one position of the hull at a particular time instant, are associated with measurements in other positions at other instants, by a set of assumptions and interpolation techniques. Hereby the complete pressure distribution can be reconstructed with the use of only a limited number of transducers. The method is evaluated with data from full-scale trials with a planing craft in waves. Reconstructed pressure distributions are applied as loads in a finite element model of the hull. The resulting structural responses are calculated and compared with corresponding structural measurements with good accuracy. This confirms that the reconstructed distributions well resembles the real distributions. The method offers a powerful tool to better handle impact pressure measurements, and enables more detailed experimental analysis of the phenomena. Areas of application are for example in full-scale design evaluations and in evaluations of time-domain simulations of loads and motions for planing craft in waves.

An implementation of the method of auxiliary state variables for solving seakeeping problems

Abstract: An analysis of the problem of the conversion of frequency-domain model for linear seakeeping hydrodynamic forces into the corresponding time-domain representation is carried out After a short consideration of the memory-functions method, the study is focused on the so-called Method of Auxiliary State Variables whose advantage is that the original mechanical system consisting of the ship and the surrounding fluid and having distributed parameters is approximated with a simpler lumped parameters system described by a relatively simple set of linear ordinary differential equations An algorithm for determination of the coefficients of those equations based on the fractional approximation of the added masses and damping coefficients curves is proposed and tested

Application of double skin breakwater with perforation for reducing green water loading on high speed container vessels

Abstract: This paper briefly reviews the phenomenon of green water faced by container ships and a time-domain method based on strip theory for predicting the occurrence of green water Verification of this method with experimental data in regular waves is also carried out showing good agreement A simple CFD hydrodynamic model is developed to represent the green water flow on deck and the parameters of this model are based on experimental data In order to test the appropriateness of this model, comparison in terms of horizontal green water load (acting on a vertical structure) and vertical deck pressure is carried out with test data Having achieved good agreement from this comparison, the research looks into the application of breakwaters in reducing the loading effects of green water Double-skin breakwaters with and without perforation are simulated facing similar green water condition Since protected structures can withstand a certain level of load without suffering detrimental effects, perforation is introduced to lessen the green water load that breakwater is to bear Three systematically varied sizes of perforated holes are simulated and effects are evaluated The research finally concludes on the performance of double-skin breakwater with/without perforation in reducing green water load and the overall effects of the perforation in balancing this load onto the structure and also the breakwater itself

New prediction method on the stopping ability of diesel ships with fixied pitch propeller

Abstract: A method is suggested to predict the stopping ability of the diesel ships with fixed pitch propeller. Based on sea trial measurements, the procedure of full astern stopping test is analyzed and modeled as the first order surge differential equation. Coasting with propeller windmilling state is taken into account for the reliable prediction of stopping ability. Duration of coasting state is determined from the propeller pitch ratio and the RPM at which braking-air is supplied. Resistance and astern thrust are estimated from the speed and power at the maximum continuous rating condition. Calculated stopping distance and time are compared with sea trial measurements, and it can be confirmed that our method can be a useful tool to predict the stopping ability of diesel ships.

On the Procedure for the Determination of the Probability of Collision Damage

Abstract: The present SOLAS regulations on probabilistic damage stability have shown anomalies for irregular or complex compartment arrangements. In this paper it is motivated that the source of those problems can be found in the use of a limited set of crisp, sharp, damage boundaries, which arose from the analytical integration of the underlying Probability Density Functions. A possible remedy would be to postpone the integration process until an actual compartment or set of compartments is evaluated. For practical reasons numerical integration is the most applicable method for that task. This proposal is elaborated and implemented in an experimental computer program, which is used to test the approach on a number of example cases. These tests have shown that there is a numerical deviation between conventional SOLAS and the proposed approach, which comes as no surprise. For practical reasons it is investigated whether new density functions, derived with the conventional method, in combination with numerical integration could lead to numerical compatibility. Finally, the effects of the current work on revision of the SOLAS rules are discussed, recommendations are made and subjects for further research are identified.

Inviscid duct-rotor interaction elements for a decelerating ducted propulsor

Abstract: Two aspects of modeling the impact of a characteristic thick boundary layer developing along the inner duct surface with inviscid elements are explored. First, it is argued that the trajectory of shed-vortex filaments adjacent to the walls depends on the boundary-layer speed reduction and the duct contraction selected to achieve decelerating through-flow control. Self induction associated with the forced contraction carries the tip-vortex filaments toward the duct surface, producing locally large surface speeds. This interaction calculation is combined with other components of a simple inviscid model of ducted-propulsor performance: axisymmetric duct and hub boundary-element approximations coupled with a lifting-line representation of each blade row. Parameters of the overall model are selected to bring about correlation with the circumferential-mean pressure distribution on the duct and component loads acting on the blades of the ERG Pumpjet, a strongly decelerating ducted propulsor. A second impact of the developing axial flow-speed reduction near the passage walls is a local mismatch between the flow angle and rotor surface slope. Parametric numerical estimates using an inviscid lifting-surface design code illustrate the degree of change in the rotor blade-tip geometry needed to achieve flow alignment. The rotor for the ERG Pumpjet has a large-diameter hub with significant cross-section area. An adjustment in shaft thrust to compensate for the increased pressure acting in a hub gap aft of the rotor produces a corrected axial force level more in line with other decelerating ducted propulsors.

Some considerations on the probability distributions for the damage length and damage penetration based on a re-analysis of recorded ship collisions data

Abstract: A re-analysis of the Collision Database used in the development of harmonised subdivision and damage stability rules recently proposed by SLF47 has been undertaken, aimed at clarifying the grey area identified in the zoneof large ships and in particular the large passenger cruise vessels. The results of the analysis confirm indeed the validity of HARDER for ships up to a medium size, say 175 m, above which some ship typology appears heavily penalised. It appears that, also in view of the characteristic ratio between the main ship dimensions, maintaining present nondimensional probability density functions for the damage length and penetration up to that length and then taking the parameters of the pdf as constant in dimensional form for larger vessels could remove the problem without modifying the general structure of the probabilistic harmonised approach.

Investigation of the panel-free method for potential flow computations

Abstract: The panel-free method has been investigated for its validation in the first- and second-order wave force problems, where the geometry of a body surface is described by NURBS mathematically and the integration for the induced velocity and the potential have been performed using Gaussian quadrature. There are no assumptions on approximation for the body geometry and the source distribution, which have been always made in the conventional panel method. Accuracy and convergence of the panel-free method have been demonstrated through the numerical examples for a moving submerged body and a Wigley hull.

The capacitance method of measuring the wetted surface area of a ship model

Abstract: The High Speed Marine Vehicle Committee of the International Towing Tank Conference has recommended that in tests on models of high speed marine vehicles the actual wetted surface area should be used to calculate the frictional resistance component in the scaling procedure rather than the static wetted surface area used for low speed ships. This recommendation has been made since it has been shown that the wetted surface area can vary significantly from the static wetted area, particularly at higher speeds and therefore affect the predicted power requirement. The Committee has also recommended that, since there is no universally agreed method of measuring this wetted surface area in model tests, the method of measuring the area should also be given. A new method of measuring the actual wetted surface area has been developed. The method is based on capacitance where the model hull has been given a metallic coating and then an insulating coating so that it effectively becomes one plate of a capacitor with the water of the towing tank or water channel becoming the other plate. The experiments have shown it to be a viable technique. It has the advantage that it can be undertaken at the same time as the standard resistance tests while not adding to the manpower requirement such as that necessary with visual observation of the wetted surface area.

On the use of a wind-wave model in the prediction of operational performance of marine structures

Abstract: The quantification of the performance of ships and marine structures is multidisciplinary in character. It requires input from the areas of ship hydrodynamics (for the resistance, the inherent motion characteristics and mean environmental forces), the area of ship-operations (for an operational scenario that describes the master's reactions on the anticipated and encountered behaviour with related criteria for tolerable behaviour) and oceanography (for a description of the interacting wind and waves). The most complete description of the operational performance up to now is based on a statistical approach, using long-term wind and wave statistics for the operational area or route. In this approach it is not possible to account for spatial variations of the wave characteristics along coastal routes, the coherence of wind speed, wind direction and their persistence on one hand and the characteristics of the generated waves (height, period, spectral characteristics) on the other. Regarding the predicted performance statistical methods have difficulties in accounting for complex operational scenarios and the effects of wind and wave persistence on the performance at the mission level. A method that is currently under development is based on time domain "scenario simulations". The wind and wave input for these simulations (several years of hourly characteristics of wind and waves) is obtained through a hindcast for the operational area. This hindcast accounts for the natural evolution of the spectral shape (thus accounting for separate wind-driven and swell-wave components), the directional spread and coastal effects (such as current and sheltering) on wave growth and decay. The performance is established by evaluating the progress on a mission in time; simulating a large number of missions yields the performance statistics at a mission level. The present paper describes the techniques underlying this method. The paper focuses on the wave input for the scenario simulations. Based on results of sample simulations it is concluded that wind-wave models provide a practical basis for a performance analysis. It is also shown that the variable spectral shape as accounted for in these simulations can have a considerable influence on the performance prediction.

Stress concentration in cylindrical cargo tanks of liquefied gas carriers

Abstract: A short introduction to liquefied gases and types of cargo tanks is given Attention is paid to the independent tanks, especially those of C type operating under pressure up to 20 bar, for which pressure vessel criteria are relevant Strength analysis for such a cylindrical tank with dished heads, design in accordance with the Classification Rules, is performed The analytical and numerical approach, based on the theory of shells of revolution and the finite element method respectively, are employed The stress concentration in the cylindrical shell at the stiffening ring and in the sphere edge at its connection with the toroidal shell is analysed In the former case the strength criteria are satisfied but in the latter they are not The AD Merkblatt B3 tolerates such situation assuming that the FEM results are not realistic enough Since the analytical solution indicates that the numerically determined stress level is overestimated only by 5%, the reinforcement of the tank head by a thicker spherical segment is recommended The maximum stress is reduced to the allowable level

Limit state design of ship grillages under local loads

Abstract: Determination of limit loads based on the limit analysis is discussed in the paper for locally loaded grillages of ship structures, using ice loads and side grillages as examples. Both simple and complicated grillage arrangements with web frames and stringers are considered. The solution is derived using a combined version of the kinematic method allowing one to reduce the solution to successive examination of the limit loads for grillage members as isolated beams. Shear deformations are accounted for in all cases. The results are presented as closed-form formulations.