The stress distribution and the deflection of blades of various screw propellers in steady flow are computed by the thick shell finite element (FE) method in combination with the surface panel method. A new general method is proposed to estimate the bicubic surface parameters of the deformed blade and to resolve the non-linear relation between the FE displacement distribution and the geometrical parameters of blade profiles on cylindrical sections. The characteristics of the surface flow on rigid and deflected blades are computed applying an approximation by Taylor's series expansion and minimization of the perturbation velocities. A modified hydroelastic interaction procedure is employed to obtain correspondence between the deformed blade form and the generated pressure distribution.
For a few propellers with different blade outline, the flow and the deformation patterns are shown and two main trends of opposite changes of the hydrodynamic characteristics are emphasized. In each case we investigated the influences of the variation of designed spanwise thickness distribution on the performance of the propeller.

View full abstract

The objective of this study is to develop a procedure of predicting the cavitation erosion intensity. To this end, it is necessary to estimate impulsive forces caused by collapse of cavity bubbles. In the first report the authors measured cumulative events of impulsive forces by collapse of cavity bubbles on a two-dimensional foil with a piezoelectric sensor. In this paper the authors measured the cumulative events again with three sensors of different kind to verify the results reported in the first report. The measurements by the three different sensors agreed well with each other. Among these sensors, the PVDF sensor was practical and suitable for the authors' purpose. In addition, collapse of cavitation bubble clusters was observed with a digital high-speed video camera synchronized with the signal of impulsive force sensors. Duration of collapse was the order of 100 microseconds, which was far slower than that of an impulsive force that was around 5 microseconds. According to theoretical calculations the impulsive pressure is generated when the whole cloud cavity collapses completely. But in this observation the peaks of impulsive force did not meet the instant of final collapse, but occurred some 10 to some 100 microseconds earlier than the final collapse. These results suggest that an impulsive force may not be caused by the global collapse of a bubble cluster but by the collapse of a part of cloud cavity.

View full abstract

The inclination to higher ship speed is advancing in recent years. Among others, expectation is great from high speed vessels supporting short distance transportation of highly value added products. While the studies on supercavitating propeller and others having been made in late years, their study objects were limited mainly to propeller's propulsive performance.
The root cavitation erosion, which is frequently sustained by high speed vessels, is the problem involved not only with deterioration of the propeller's durability or life but also with hindranceof the vessel's operational safety. Despite the fact that the study on its generation mechanism was made, and several countermeasures with some appendages against the phenomenon were devised, no development has been made yet on a method for an effective prevention of the phenomenon with any designing improvement applied to a propeller itself. If the root erosion generation peculiar to high speed vessels could be prevented, we understand that the relative method will surely comply with the need by the shipping industry.
Under such a background as above, this study aims at development of the new propeller effective to prevention of root erosion. Our concept of the new blade section for the root erosion prevention is described in the concrete, to be followed by the description about verification of its effectiveness by means of model test and actual vessel test, and successful development of the new propeller with preventive effectiveness against the root erosion for high speed vessels.

View full abstract

A new CFD simulation method has been developed for simulating unsteady motions of a vehicle with movable wings. The coordinate system used in this simulation method is fixed to the moving body and inertia forces such as Corioli's force and the centrifugal force are treated as body forces in the incompressible Navier-Stokes equation. A multi-block grid scheme is introduced for the body having a number of wings and the motion of movable wings is treated in a moving grid scheme. The method of simulation is explained and some results of the simulation applied to the keel of a racing yacht are presented.

View full abstract

Wing-In-Surface Effects are studied for two-dimensional wings of three profiles. The profiles are NACA0012 as a standard profile, NACA6409 as a high cambered profile and an S-shaped profile. Numerical simulation is carried out for steady flows past the wings at a high Reynolds number with turbulence by a finite volume method. Results show that for the high cambered profile increase of lift is significant but stability is very poor. The S-shaped profile has good stability and a moderate lift coefficient.

View full abstract

A numerical simulation method has been developed for predicting the maneuvering motion of blunt ships by solving the equations of motion with the Navier-Stokes equations. Since coordinate system is fixed to an arbitrarily moving ship, inertia forces must be incorporated into the NS equation as body forces. The hydrodynamic forces acting on the hull of the ship are obtained by solving the incompressible and time-dependent NS equations numerically. The hydrodynamic forces caused by rudder and propeller and their interactions with the hull are calculated by the mathematical model. This method was applied to the simulations of the 10-degree Z maneuvering motion of two VLCC models with the same principal dimensions and different aft-part frame lines, that is, one is so-called V-shaped and the other is U-shaped. The results of the simulations agreed well with those of model tests and revealed the typical difference in maneuverability between the two ships.

View full abstract

Since the interim standards for ship manoeuvring performance A.751 (18) was adopted in International Maritime Organization (IMO), it became very important to predict the ship manoeuvring performance with high accuracy at initial design stage.
In this paper, the theoretical prediction method for the hydrodynamic forces acting on ship hull which can be took into the effect of frame line configuration is proposed based on the slender body theory. Furthermore, it is found that we will be able to predict the course stability of ships with comparatively accuracy from the results of numerical calculation for the hydrodynamic forces acting on ship hull. Therefore, it is considered that the present method will be useful to evaluate the course stability of ships at initial design stage from the viewpoint of practical use.

View full abstract

Considering the simplified vortex model with ring, horse-shoe and 3D cross-flow vortex described in the authors' previous papers, the authors presented as component type mathematical model for hydrodynamic forces X, Y, N that act on a steered hull with an even keel. The mathematical model can estimate the hydrodynamic forces in turning motion by using the hydrodynamic characteristics analyzed even from oblique hydrodynamic forces through the model, and additionally describe the hydrodynamic forces in maneuvering motion with small to large yaw rate and drift angle.
This paper describes the mathematical model due to the simplified vortex system for the trim condition in general, such as in sea trials of cargo ships for IMO regulation and the conventional stage of fishing vessels with initial trim. Furthermore this paper presents a method for analyzing the characteristics of oblique hydrodynamic forces from the new viewpoint of measuring system with 2 support points at FP and AP.
This mathematical model can describe most characteristics of the hydrodynamic forces in oblique motion with drift angle 0 to 90 degree and turning motion with yaw rate |γ'|≤1, by using the hydrodynamic characteristics analyzed even from oblique hydrodynamic forces with drift angle covering 020 degree.

View full abstract

Many experiments had been carried out in the long experimental towing tank in Yokohama National University (YNU) with generating directional spectrum waves by segmented and independent wave makers and using side-wall reflections positively. It is important to make more accurate generation of the directional spectrum waves considering the increasing demands for experiments among such waves.
To accomplish these subjects, present paper investigates the characteristics of directional distribution by carrying out numerical generation of directional spectrum waves and its analysis for the case of without and with side-wall reflection. This calculation is based on the superposition of ring waves generated by the periodic line source distributions representing each wave maker unit. Mirror images are also introduced for considering reflections of side-wall including the ratio of reflection (reflection coefficient) because directional waves will not be reflected perfectly from real side-wall. In addition to this, directivity (D_d) and tapering effect for the distribution of source strength will be evaluated to generate more accurate directional spectrum waves. Especially, influence of spurious waves generated by the combined effect of finite width of segmented wave maker, wave length and main wave propagating direction will be investigated.
To analyze directional spectrum waves, we use MLM (Maximum Likelihood Method) from calculated time series of wave elevation and two component wave slopes at the arbitrary spatial point. This corresponds to the Laser Beam Type Surface Wave Probe developed in Y. N. U, and reliability of this method was examined already.

View full abstract

Present paper describes an image measurement system for the wave height distribution around ship hulls in regular wave. The wave profile on a laser light sheet is visualized by the difference of light scattering between air and water. Detecting the boundary of bright and dark area on the visualized image with an image processing technique, the instantaneous wave profile on the sheet illumination is obtained. The wave height distribution in calm water can be obtained by reconstructing the distribution from the measured time-series data on a computer.
The conditional sampling technique enables to obtain the wave height distribution in regular wave. The wave pattern around a ship in regular wave changes periodically, and its period is based on the encountering interval of wave and ship hull. The measured time-series wave profiles are sorted using their positions on the model fixed coordinate system and phases of encountering. It is the first try to obtain the time series wave pattern based on an experiment.

View full abstract

Since the tragedy of Estonia happened in Baltic Sea, the enhancement of safety of RO-RO passenger ship is the interested problem in the world. Therefore, many research works have been continued in many countries, including our country. There was a proposal by United Kingdom for International Maritime Organization, that is the requirement of the stability performance related to free water effect on deck. For investigating the proposal, a model ship was made as a typical RO-RO passenger ship in Japan and model experiments were carried out in damaged condition and in irregular beam seas with JONSWAP spectrum. The main conclusions from the experimental results are shown as follows,
(1) The ship does not capsize in the condition of no initial heel, but she capsizes with initial heel to damage side in higher significant wave height than the proposed one.
(2) The critical significant wave height is affected by the peak period of the wave spectrum.
(3) The positive area in GZ curve is related to the critical significant wave height.

View full abstract

A reasonable method used in prediction of ship motions leading to capsize in severe waves is developed on the basis of strip method. In this method the variation of metacentric height in waves is taken into account. Several simulations were conducted to predict the stability against capsizing of a container carrier 15000GT in severe waves due to parametric rolling. Finally the stable and unstable areas of the ship running in severe astern seas are computed

View full abstract

This report is concerned with statistical analysis on long-term measurement of wave induced load of naval ships. We have analyzed wave-induced vertical bending stresses at midship deck and wave conditions measured by 12 naval ships which are equipped with Ship Motion Analyzing Computer System (SMACS) over a period of 84 years. We have introduced contribution rates between wave period and wave height on long term distribution of wave induced vertical bending stress in each stress revels and contribution rates between wave period and heading angle between ship and wave direction on the long term distribution. We derive the following conclusions.
(1) When the range of the probability of exceedance in the long term distribution are from 10^-1 to 10^-2, the contribution rates between wave period and wave height and the rates between wave period and heading angle widely distribute over the wave conditions and heading angles.
(2) For the range of the probability of exceedance from 10^-4 to 10^-8, the contribution rate is centered at a wave condition and a heading angle. The condition equals to the broad sense of resonance when the peaks of both the response amplitude function of the wave bending stress in regular wave and wave spectra are superimposed.

View full abstract

A nonlinear calculation method based on the Mixed Eulerian Lagrangian method is presented for wave-induced motions of a 2-D floating body. Attention is placed on an effective calculation of the hydrodynamic force associated with the temporal derivative of the velocity potential in Bernoulli's pressure equation. Unlike other existing methods, the acceleration field can be computed simultaneously with the velocity field, which contributes greatly to reduction of the computation time. By use of Green's second identity, the new method is explained as an extension from the mode decomposition method, and close relations between the two methods are emphasized.
Computations are performed for a wall-sided model and a flared model, and numerical results of the waves at upwave and downwave positions and the body motions (sway, heave, and roll) are compared with corresponding experiments. The overall agreement is very good, confirming validity of the present method. Discussion is also made on the parametric oscillation in roll, observed for the flared model.

View full abstract

For the people interested in the behaviors of a floating body in waves, exact nonlinear (including viscous effects) calculations of a 3-D body of arbitrary geometry in arbitrary waves must be the final goal. As a first step for this goal, we proposed in the previous paper the calculation of the motion of a 2-D floating body in a perfect fluid using MAC method. In that paper, we compared the response amplitude operator of the body in small regular waves calculated by the presented method with the experimental results and with the ones due to the linear potential theory. We also compared the calculated results on the time history of the motion of a 2-D body in transient waves with the experimental results. In both cases, the calculated results agreed well with the experimental results and with the ones due to the linear theory as well.
Using MAC method, it is fairly easy to account for viscous effects in the calculation while it is quite difficult to do that in a calculation based on the potential theory. In this paper, as an extension of the previous paper, we first calculate the motion of a 2-D floating body in small regular waves in a viscous fluid and compare the response amplitude operater with experimental results and also with calculated results conducted in a perfect fluid. Next, we extend the ciculation to the motion of a 3-D floating body in a perfect fluid and compare the response amplitude operator of the body in small waves with the ones due to the linear potential theory. Finally, we calculate the hydrodynamic impact pressure on a 3-D body due to an entry into water and compare the results with existing results obtained by a perturbation theory.

View full abstract

It is known that a ship can be regarded as a type of wave measuring device, and under the assumption of linear superposition, wave spectra can thus be estimated from measured spectra of ship motions. Based on this idea, the evaluation of one-dimensional wave spectra, as well as directional wave spectra, has been extensively carried out in recent years. However, when a ship is traveling with an advance speed in following seas, there is a range of encounter frequencies ω_e for which cannot be related to the wave frequencies ω by a single-value conversion. Practically, this difficulty has been solved by keeping the lowest frequency solution, however resulting spectra may only a narrow range of wave frequency when a ship navigates with high speed in following seas.
In the previous paper, we proposed to determine the wave spectrum by using a nonlinear programming method, in which a standard wave spectrum, such as the ISSC, is selected, and a significant wave height and average wave period are used as design variables. The optimization was carried out in the wave frequency domain.
In this paper, we propose to determine the wave spectrum by using a method in which the ordinates of wave spectrum at an arbitrary wave frequency are used as design variables and the optimization has been carried out in the encounter frequency of the ship.
The capability of this approach has been confirmed by comparing the wave spectra estimated from the measured motions of container and ore-carrier models, with the ones obtained from the direct measurement of waves in the experimental tank.

View full abstract

The direct parameter estimation algorithm is applied to the simple linear maneuvering model (T-K model). The algorithm is composed of the IIR (Infinite Impulse Response) digital filters and RLS (Recursive Least Square) method. The IIR digital filters are introduced to avoid taking direct derivatives and to reduce the effects of measurement noise. Furthermore, the digital filters are free from the initial condition problem and the accumulation problem of integration error.
In order to discuss the quantitative accuracy, the authors carried out onboard tests. And, the direct parameter estimation algorithm is compared with an indirect algorithm with RLS method. There is satisfactory agreement of the ship maneuverability indices T-K estimated by the direct identification method and the ordinary graphical analysis. This concludes that the direct parameter estimation algorithm With IIR filters is a powerful tool for on-line identification of continuous systems.

View full abstract

A new energy utilization project developed by Tokai University was started in 1991. It was ordered to develop high-speed one-man water racer fully using solar power. In 1994, a new type watercraft was launched, that was full-submerged hydrofoil craft with 480W solar power. Her dynamical performance was tested during sea trials and improved year by year. This paper discusses the new maneuvering method of the hydrofoil craft with two struts and a tandem hydrofoil system. Her motion's control system was quite simple. Without any automatic control, roll motion stabilized by the driver's weight shift and steering like bicycle or motor cycle. This paper proposes a new equation of motion to describe her turning motion. The results of the numerical analysis show, which concerned the stability of the lateral motions and yow-roll-sway coupled turning motions.

View full abstract

A hybrid anti-rolling system was developed and equipped on board of the oceanographic research vessel “MIRAI”. The greatest advantage of the system is that it is capable of reducing a rolling motion by fifty percent at low speed, even while drifting. This enables to carry out oceanographic observations and samplings at rough seas at high latitudes. The hybrid anti rolling system has both active and passive mass movement by a pendulum mechanism which is driven by electric motors. It was proved by sea trials that the amplitude of rolling motion of the vessel was reduced to a half by the hybrid anti-rolling system in the Sea State less than 5.

View full abstract

The structure of the head of gravity currents is investigated in the experiment and the computation of lock-exchange flow. A finite volume of fluid was released instantaneously in another fluid of slightly different density in an open rectangular channel when a lock gate was opened. A box model was introduced to describe different regimes of a developing gravity current on the basis of energy argument.
Visualization experiments were made by the shadowgraph method and with dye to trace the head of gravity currents, the results of which validate the box model. The initial depth and volume of the released fluid affect the front speed. At the first stage the released fluid spreads at a constant speed, and then it slows down so that the frontspeed decreases with time to the power of -1/3 in self-similar regime. The current that has begun to be dominated by viscosity slows down further in the final stage during which the front speed decreases with time to the power of -4/5.
The computation was also made in the same condition as the experiments to study the structure of the head of currents in different regimes. The incompressible Navier-Stokes equation for an inhomogeneous fluid together with the transport equation for solute was solved by the finite volume method developed in the previous paper. The computation produces the transition between the development regimes so that the front speed obtained from the computed density field is in good agreement with the experimental result. The computational results indicate that the transition to the next regime is accompanied by remarkable changes in the geometry and the inner structure of the head. The head has maintained a sharp density interface even at low Reynolds number and the relatively large-scale entrainment of ambient fluid across the upper interface causes strong mixing behind the head. The models with the inviscid boundaries failed to catch the transition to the second stage, which suggests that the transition may require some stimulus associated with viscous motions. The result for the case of short lock-length shows the sudden decrease in the volume and in the density of the head brought about by the vortex shedding along the upper interface, which may be one of the mechanism to account for the strong mixing in the first and the second regimes.

View full abstract

Numerical simulation of ocean current is one of significant tools which are of practical use for environmental assessment of oceanophysical effects of large floating offshore structure, for example, on the surrounding sea water. In the previous paper, the so-called multilevel model of ocean current was adopted to simulate tidal current observed in the vicinity of an existing large floating offshore structure (LFOS), called Mega-Float model, moored off Oppama, Yokosuka. As a result of comparison between the observed current and simulation results, it was concluded that the main characteristic feature of tidal current in the close vicinity of the LFOS could be reproduced satisfactorily by numerical simulation. In the present paper, variation of water temperature and salinity, in addition to current velocity, are taken into consideration in order to enable numerical simulation to reproduce more detailed features of observed tidal current. The oceanographical observation carried out at the mooring site of Mega-Float model revealed that vertical structure of tidal current at low tide was complicated and changeable : on 29th August 1995, the upper current was southward while the lower current was northward, but one day after 29th August, both currents were northward. The numerical simulation presented in this paper partly succeeded in reproducing this complicated tidal current at low tide. In addition, the effects of Mega-Float model on tidal current, sea temperature and salinity were examined using the results of numerical simulation. As a result, it was found that Mega-Float model did not bring forth untrifling difference in sea temperature and salinity.

View full abstract

A numerical simulation method was developed for solving multi-phase flow with mass transfer through the interfaces. The interface is determined by solving the transport equation of marker density function. Surface tension is treated as a body force in the NS equation enjoying porosity concept. Through the interface, the mass of dispersed phase dissolves into continuous phase. As a result, the dispersed phase diminishes its volume. Three case studies were carried out to validate the method. Drop formation simulations demonstrate that the present treatment of the surface tension is accurate enough to simulate the oscillation of drop volume with the increase of flow rate. Simulated dissolution rate from a spherical droplet without flow is in good agreement with the analytical solution. Then the dissolution from a rising droplet is successfully simulated. It is visualized that horse-shoe type vortices are attached by the dissolved mass in the wake of the deformed rising droplet. The resultant Sherwood number agrees well with the solution of an empirical equation.

View full abstract

Recently, efficient numerical analysis code has been developed for pontoon-type VLFS with the capability to analyze hydroelastic responses in regular waves of relatively short wave length.
Using these response amplitude operator, time domain analysis can be performed for the responses in irregular waves by linear superposition principle.
By the same manner, the response to the impact due to dropping object or landing of aircraft on the VLFS can be estimated.
This paper deals with the numerical calculations of time series responses in irregular waves and the response to the impact or moving load. The numerical results were compared with model experiments showing good agreement.

View full abstract

In recent years, Very Large Floating Structures (VLFSs) such as floating bridges and Mega-floats have been planned for efficient utilization of ocean space. Accurate prediction of elastic behavior of VLFSs due to wave action is indispensable for their design owing to their flexibility.
The present study has been carried out under the Mega-float Project as one for developing an accurate general-purpose three-dimensional computer program for hyroelastic responses of VFLSs due to wave action in the water area with complicated geometry close to the real situation.
This paper presents a numerical method for predicting the hydroelasic behavior of a VLFS in the complicated water area sheltered partially by breakwaters and land. A standard modal approach is adapted for the fluid-structure interactions.
For the free-surface flow, an improved hybrid finite/infinite element formulation has been developed to reduce the computational size and cost for practical applications substantially. The flow field is divided by a fictitious boundary into an outer infinite domain and an inner irregular one, which are again divided into subdomains appropriately. The velocity potentials for the inner and outer subdomains are approximated by the corresponding vertical orthogonal eigenfunction expansions, and planar finite elements and hybrid infinite elements, respectively, and are matched each other on the corresponding interfaces in an averaged sense.
To very large rectangular plates, analytical Rayleigh-Ritz methods are applied.
Numerical results for a VLFS in the open sea have shown a satisfactory agreement with the existing numerical and measured ones. The response of a VLFS in a shelterd area and wave diffraction/refraction have also been examined.
Conclusions are that the present 3D hybrid element code is a powerful working tool for practical VLFS investigations.
Advanced simulations for 3D VLFS models will be reported in the subsequent papers.

View full abstract

A very large floating structure (VLFS) for various applications such as a floating airport is generally classified into a pontoon-type, a column-supported type or a semisub-type. It is expected that large elastic deformations of such structures are induced by the dynamic effects of waves, because the horizontal dimension is extraordinary large compared to the thickness of the structures or of their upper deck. However, there exist few researches as to how the structural type of a VLFS affects its elastic deformations in waves. Recently, we have developed a prediction method of hydroelastic behavior of VLFSs. In this paper, the characteristics of wave-induced elastic deformations of a pontoon-type VLFS and those of a column-supported VLFS are compared using the developed method. It is found that the characteristics can be categorized in terms of the area ratio of the water surface area of the corresponding VLFS to the plain projected area of the whole structure rather than be categorized in terms of the structural type. This, in turn, implies that the relevant parameter that determines the elastic deformations of a column-supported VLFS is not the number of columns but the area ratio

View full abstract

A B-spline type direct method is proposed in this study for predicting wave-induced hydroelastic motion of a very large floating structure. In the proposed method both the deflection of the structure and unknown pressure distribution are discretized with bi-cubic B-spline function. Hydrodynamic pressures due to the structural motion and incident waves are estimated by applying the pressure distribution method with Galerkin's scheme. The shape of the motion of each radiation freedom is given by the B-spline basis function, and the amplitudes of the radiation motions are determined by solving the Lagrange's equation of the structure that ensures the satisfaction of energyconservation principle. Comparisons between the calculations of the proposed method and experimental data have been carried out for the deflections and bending moment of a single structure, and it shows that the theoretical predictions and experimental results are in a very good agreement. In addition, the proposed method is applied to multiple elastic structures connected with beams. Again, the comparison of deflections of the structures and the bending moments of the beam connectors in various incident waves show good agreements between the calculations and experiments. The merits and demerits of the direct method and modal analysis method are also discussed.

View full abstract

In recent research, it has cleared that the motions of pontoon type very large floating structures (VLFS) are dominated by the dissipation wave that is induced by the floating structure and the fluid interaction. Incident wave excites the weather side of the structure, and then the deformation is propagated over all the structure. The deformation wave propagation to the main part of VLFS is not desirable, because the deformation requires the large area of high strength of the structures and the motion might affects the function of VLFS. In this study, the edge shape of VLFS that restrains the propagation of the deformation wave has been investigated. To evaluate the effect of the improved edge form, a program to calculate the response of VLFS that has arbitrary horizontal outline has been developed. Using the program, it has been confirmed that the notched edge reduces the deformation of the inner area of VLFS. In order to reduce the motion effectively, a relation between the edge form and the incident wavelength was found.

View full abstract

In the first report, the decay characteristics of waves incident to an array of cylinders were investigated. The cylinders are considered as a part of the supporting legs of a very large floating structure. Maniar & Newman have recently shown that certain resonant phenomena occur, at least numerically, in a long array of equally spaced cylinders fixed in head waves. Surface elevations between adjacent cylinders or the forces on each cylinder could reach as high as 20 times the corresponding quantities experienced by a single isolated cylinder. Numerical facts also suggest that the surface elevations or the forces decay from the head of the array toward the end of the array quite uniformly in incident waves of higher frequencies whereas they are enhanced rather than be decayed in incident waves of lower frequencies than the resonant frequencies. The purpose of this second report is to examine what happens around the resonant frequencies in real phenomena. The implications of the possible resonant phenomena on the hydroelastic behaviours in waves of a leg-supported VLFS are also examined.

View full abstract

The authors developed the algorithm which applies the quantitative risk analysis (QRA) to the rational safety assessment for a very large floating structure, considering a mooring system as an example. In the beginning they describe the rational or modern safety assessment based on QRA, and discuss about the general safety for a very large floating structure which is required to obtain the public acceptance. Then they consider the mooring system for the very large floating structure as an example to show the possibility of the application of QRA, and developed the time domain simulator for the estimation of mooring line forces which is used to collect the data base of the accident while the method to estimate accidental costs, that is to say, consequences is shown, and finally they discuss the algorithm of the safety assessment based on QRA.

View full abstract

The authors consider a pontoon type flexible floating airport as the very large floating structure. In order to design this kind of structure, we have to investigate several key points in advance. Among those key points, estimation of the wave drift forces is very important in order to design of the mooring system. In this paper, the authors consider the square type flexible flat plate with shallow draft, exactly speaking, zero draft as the floating airport. The authors develop the computer code of the slowly varying wave drift force for an elastic floating structure with shallow draft based on the so called Near Field Method which is validated by both the Far Field Method and the corresponding Experiments.
The authors conclude that even though the water depth is shallower, there is no big difference among the steady wave drift forces in each water depth, while the rigidity of the structure is smaller, more flexible, then, the steady wave drift forces or mooring forces are smaller.
Finally, they showed the flexibility of a floating structure decreases not only steady wave drift forces but also slowly varying wave drift forces under the numerical time domain simulation of a very large floating structure with the multiple dolphins mooring system.

View full abstract

The main purpose of the present paper is to investigate the effect of land boundary on low-frequency wave-induced motion of a large floating structure. Comparisons of the response function of the floating body motion to incident wave are performed between the cases where the body exists in an open sea and is surrounded by land boundary. The results indicate that land boundary influences the body motions and wave field significantly and the effect of the land boundary have the important role to evaluate motions of the large floating body.

View full abstract

When a very large floating structure (VLFS) is designed, it is quite important to consider the forces by tsunami especially in our country that is often attacked by big earthquakes. The forces on ocean structures by long waves such as tide, surge, and tsunami have been usually estimated by semi-empirical way, for example, using the drag coefficient due to the shape of the structure in constant flow. However, for VLFS like an ocean airport, currents may be different at the different position of the structure. Also, the hydrodynamic interaction between the structure and currents must be taken into account that can be ignored in cases of normal size structures. In this study, a new procedure to estimate the forces on VLFS by tsunami was proposed, in which distribution of currents and waves around the structure were estimated by multilevel model. The initial condition of surface water level of tsunami was given by the sea bed fault model with static parameters, and Reynolds equations using the hydrostatic approximation were used for calculating the propagation of tsunami. The deformation of VLFS and the field of currents and waves were calculated simultaneously, then the pressure force and frictional force on the structure were estimated by numerical way. The method was validated by comparison with the results of some other numerical calculations. The comparison with hydraulic model tests in experimental tank was also made and it is found that the present model can roughly estimate wave propagation and deformation of a solitary wave, motion responses of an elastic structure, and mooring forces on the floating body. Moreover, the forces on VLFS located in Tokyo Bay were estimated for some sea bed fault models with different characteristics.

View full abstract

Inherent damage zone model assumes that the stress at a point which is located at a small distance, γ₀ from the crack initiation position governs the fatigue characteristics regardless of the geometric configuration of the specimen. The distance γ₀ represents the size of effective damage zone and it is an inherent length of the material. Ishibashi proposed the model to explain the difference between fatigue notch factor and stress concentration factor of notched specimen. This paper shows that the inherent damage zone model can explain several fatigue phenomena near the fatigue limit and the crack growth threshold consistently without restriction of smooth specimen, notched specimen or cracked specimens with short and long crack length. A special feature of the paper is using the exact solution of stress distributions of notched and cracked specimens. Neuber's analytical solution for notched specimen and Westergaard's stress function for cracked specimen are employed for this purpose. Relationship between fatigue limit of smooth specimen and threshold stress of cracked specimen, relationship of fatigue strength between round-shaped flaws and crack, occurrence condition of non-propagating crack at the root of sharp notch, and fatigue notch factor are discussed quantitatively based on the proposed model.

View full abstract

Multiaxial fatigue behavior of boxing welded joints of a JIS SM400B steel (12 mm thick plate) was examined using a biaxial fatigue test system. For a specimen, two stiffeners were attached to a main plate by a CO₂ semi-automatic boxing welding. Residual stress measurement and finite element (FE) analyses were also done. Fatigue tests were performed under both uniaxial and biaxial (mainly out-of-phase) cyclic loads, and both results were compared and examined.
It was found that fatigue cracks in the biaxial test specimens were initiated at the boxing weld toes and propagated almost in the direction of lateral loads. And the relation between stress range near a weld toe, Δσ₅, and crack initiation life, N_c, was almost common in the biaxial and the uniaxial tests, proving that the increase of Δσ₅ due to the lateral loads is fully effective to the increase of cyclic fatigue damage because of the high tensile welding residual stress near the boxing weld toe. On the other hand, failure life, N_f, was slightly longer in the biaxial tests than in the uniaxial ones.
Finally a simple life estimation for the biaxial tests was made using the results of the FE analyses and the uniaxial tests, giving a little conservative results due to overestimation of stress range Δσ₅ and the difference in crack propagation life between the uniaxial and the biaxial tests.

View full abstract

Sacrificial specimen is a small sized fatigue specimen which is designed so that fatigue damage may be produced in the specimen sufficiently earlier than in a structural member when it is secured to the structural member. In the previous study, a sacrificial specimen which is secured onto the structural member by using adhesive agent is developed. And the fatigue property of the sacrificial specimen was investigated under constant stress amplitude. In this study, the sacrificial specimen is adhered to a smooth specimen or a welded joint specimen and the fatigue test is carried out under block loading with exponential distribution. It is found that the sacrificial specimen shows a stable fatigue property under block loading and has a high sensitivity for fatigue damage. Fatigue crack is produced in the sacrificial specimen sufficiently earlier than in the welded joint. Further, a method to predict fatigue life of structural member is developed based on the fatigue damage condition monitored in the sacrificial specimen.

View full abstract

In this paper, a stress sensor using piezoelectric element is developed to monitor the feature of stress history of structural member. The sensor is a rectangular box (90mm long, 60mm width and 25mm thick) made of thin aluminum plate, in which piezoelectric element, electric circuit including integrating circuit, button type battery, make and break switch, stress level meter using light emitting diodes (LED) and stress level cross counter are facilitated. The stress level meter can turn on and off different color diodes according to the change of stress level. The stress level cross counter can display the number of cycles which exceeds a specific stress level during a monitoring period. The sensor is bonded onto the structural member by using adhesive agent. The monitoring period can be determined by the attached switch.
In order to see the performance of the prototype sensor, experiments are carried out using a fatigue testing machine. The sensor is bonded onto a smooth specimen and multiple repeated stresses are applied to the specimen. The signal of the LED and the output of the counter are investigated by changing wave shape, stress amplitude and frequency. Followings are made clear from the experiments.
1) The sensor can operate under the stress range above 10 MPa, and the frequency from 0.05 Hz to 5 Hz.
2) LED of the stress level meter can turn on and off accurately under any stress amplitude and frequency.
3) When the stress frequency is 0.5 Hz or less, the level cross counter can accurately count the number of cycles which exceed the predetermined stress level.

View full abstract

A crystalline FEM theory which incorporates both non-Schmid effects and the Bausinger's effects is presented. The theory is applied to investigate the relationship between the strain hardening properties and the properties of plastic deformation localization and irreversible slip generation by calculating the deformation behavior of f. c. c. single crystal under cyclic loading conditions.
As a result, the followings are found :
1) The cause of plastic deformation localization and irreversible slip generation is the deviation of resolved shear stress and back stress. These phenomena are induced by the difference of stress distribution during tensile/compressive loading.
2) Plastic deformation localization and irreversible slip generation proceed slowly when the hardening rate is small and/or the latent hardening is weak.
3) The accumulation of irreversible slip saturates when the hardening rate is sufficiently small under the conditions calculated in this report. Plastic deformation localization and irreversible slip generation difficult to arise when the latent hardening is sufficiently weak.
4) The progress of plastic deformation localization and irreversible slip generation is accelerated when non-Schmid effect exists. The normal stress of slip plane changes the properties of these phenomena under the conditions calculated in this report.

View full abstract

Restraint of fatigue crack propagation by wedge effect of some fine particles was examined on a JIS A 5083P-O Al-Mg alloy.
For a fatigue test, a 5 mm thick plate specimen with a central EDM notch was used. As the fine particles to be wedged into fatigue cracks, three kinds of particles were prepared, namely, magnetic particles and two kinds of alumina particles of which the mean particle sizes were 47.3μm and 15.2μm. Before the fatigue tests, particles of each kind were suspended in an oil, and the suspension was spread on the specimen surface covering the notch part. Fatigue tests were performed by a servohydraulic type fatigue tester with a test frequency of 2 Hz and a load ratio of R=0.
First a control fatigue test was done by spreading only the oil on the specimen surface, and it was proved that the wedge effect of the oil itself is negligible. The alumina particles of large size were also not effective to restrain the crack propagation, because the suspension is difficult to be made due to the large particle size and the particles cannot be sent to the crack tip properly.
On the other hand, both of the magnetic particles and the alumina particles of small size had evident effects to restrain the crack propagation, especially in the case of the alumina particles the failure lives were extended by the range from 4×10⁵ to 7×10⁵ cycles. From some crack propagation plots, it was found that the restraint of crack propagation by the particles occurs mainly in the very early stage of crack propagation and the restraint effects become weak when the fatigue crack length exceeds approximately 3 mm.
After the fatigue tests, some macro- and microfractographic analyses were performed using a CCD microscope, a SEM and an EPMA, in order to examine the mechanism of fatigue crack restraint by the wedge effects of the fine particles. From those analyses, it was reasoned that the fine alumina particles wedged into a fatigue crack are subjected to cyclic pressures from the crack surfaces, crushed up into finer particles, and then form a kind of alumina coating (in the vicinity of the notch root) which prevents the free crack closure as a wedge.

View full abstract

A simple and efficient method of buckling/plastic collapse analysis of stiffened plates subjected to uni -axial thrust is presented. Plate part between stiffeners is divided into several plate elements in the stiffener direction. Thebuckling/plastic collapse behaviour of plate element is simulated using a simple dynamical model combining the results of elastic large deflection analysis and rigid plastic mechanism analysis. The influence of a torsional stiffness of stiffeners on the local buckling strength of plate part is taken into account. Stiffener part is modeled by beam-column elements which can simulate a flexural-torsional buckling behaviour of thin-walled beams. Coupling the plate and beamcolumn elements, the buckling/plastic collapse behaviour of stiffened plates accompanied by both local and overall buckling can be analysed. It is also possible to simulate a localisation of plastic deformation in a certain part of panel deflection and a resulting unloading in the remaining part, which have a large influence on the capacity beyond the ultimate strength.
Using the proposed method, a series of buckling/plastic collapse analysis of stiffened plates is performed, and the applicability of the proposed method is discussed through a comparison with the ordinary elastoplastic large deflection FEM analysis.

View full abstract

Corrosion at welds is pointed out as a main factor that causes damage to ships. Insufficient maintenance for ship causes serious accidents due to the deterioration of products. These accidents lead to serious problems such as environmental pollution from spilled oil, loss of life, etc. In order to prevent such accidents, we have to design the Life-Cycle of a product before product design, evaluate a product-life, and make maintenance plans by computer simulation. Additionally, we should think of realizing the best product design from the viewpoint of whole Life-Cycle management.
This paper presents the Welds Spring Model, which can be considered an effect of corrosion at welds of a ship structure, and shows the scenario for the collapse of a ship. The Welds Spring Model is defined by several springs on welds to consider welds as a structural element like parts. In addition, the simulation for evaluating the product deterioration by corrosion at welds is discussed. In order to support the Life-Cycle design with consideration of the corrosion, we have developed the Life-Cycle system that is based on a product model for the computer aided design and production.

View full abstract

Ship hull vibration is much complicated phenomenon because of the complexity of interaction between the vibration characteristic of ship hull, main engine, and shaft system (crank shaft through propeller) structure. To cope with this problem, we developed new method that is to analyze 1. Dynamic interaction between ship hull and main engine by means of substructure approach connecting each vibration characteristic of ship hull and main engine structure, 2. Vibratory response considering the influence of moving/rotating parts in main engine, and 3. Contribution of each exciting force to the vibratory response of the superstructure. By this means, we proved that the characteristics of the exciting force from the shaft is much influenced by the vibratory characteristic of the shaft, the axial force acting on the thrust block and the axial damper has much influence to the superstructure vibration. Also we showed that the reduction of axial force ispossible through the control of the vibratory characteristic of the shaft system.

View full abstract

This paper presents the numerical procedure for solving the fluid-structure interactive vibration of a three dimensional complicated structure such as the stiffened plate, deep girder, transverse ring and so on. So far, a classical boundary element method has been widely used for estimating the added mass effect. of the fluid surrounding the structure. However, when the plate thickness is relatively thin, the previous method may not. be applicable because the numerical error is not neglected to evaluate the fluid loading effect. on the plate. Thus, a new program code TPBEM is developed and applied to determine the distribution of the dynamic pressure on the thin circular plate and the web frame model. Some of the calculated results are compared with measured ones and the validity of the present. method is discussed.

View full abstract

Sloshing in the liquefied natural gas (LNG) carrier occur due to the large dimension of the tank and the characteristics of the liquefied cargo. The description of nonlinear fluid flow and random tank motion in irregular waves are required to predict the sloshing phenomena. A numerical code adopting SOLA-VOF scheme has been developed to analyse sloshing flow. For the ship motion analysis 3D panel method is applied and the results are compared with those of ABS class. Furthermore, to evaluate sloshing load in the directional waves, this study used ISSC directional wave spectrum with cosine square directional spreading function. Ship motion time histories are used as the input data of sloshing calculation. We performed sloshing analysis for No.2 tank of 138K LNG carrier with fully loaded condition. Higher sloshing responses are predicted on head sea where the sloshing peak pressures are inversely proportional to the zero crossing wave periods than the other seas in high filling and severe sea conditions. The calculated maximum peak pressures considering the cushioning effects of corrugated membrane in head sea can be used as the external forces for the structural analysis.

View full abstract

In order to evaluate the safety of a submerged structure subjected to UNDerwater EXplosion (UNDEX) loads, it is necessary to estimate the structural response accurately.
This paper describes the experimental results and numerical simulation results of UNDEX test using thin and thick cylindrical shells. From this study, fluid-structure interaction effects to peak value, waveform and impulse intensity of surface pressure are discussed. And, it is shown that surface pressure of thin cylindrical shell is strongly influenced by surface cavitation. Further, it is shown that the reasonably modified numerical simulation results have very good agreement with the experimental results, and the mutual comparison between experimental results and numerical simulation results is helpful to estimate the true peak pressure.

View full abstract

Fatigue tests on fillet welded joints were carried out using test specimens simulating reduced plate thickness and reduced fillet weld due to corrosion wastage to investigate the influence of corrosion wastage on the fatigue strength of fillet welded joints. The results of the tests and the related studies are summarized below.
It was observed that in cruciform fillet welded joints, the tensile fatigue strength of test specimens with reduced plate thickness decreased considerably compa'red to the tensile fatigue strength of sound specimens. In side fillet welded joints, the shearing fatigue strength of test specimens with reduced plate thickness decreased considerably compared to the shearing fatigue strength of sound specimens. Stress analysis of test specimens was carried out using solid model. The maximum stress arised in the root of the fillet weld at the central part of the width of the plate in cruciform fillet welded joints, and also in the root of the fillet weld at the end of center side of the test specimen in side fillet welded joints. The position where the maximum stress arised almost coincided with the position where the crack occurred in fatigue tests.
Rules related to thickness of web plate and leg length of fillet weld of hold frames in large bulk carriers were amended in view of the marine casualties to these ships. From the results of fatigue tests, it was concluded that considerable improvement may be anticipated, as a result of the amendments to rules, in the fatigue strength of fillet welded joints in which corrosion wastage has progressed.
From the results of studies on measured data of plate thickness, it was observed that the variation in initiation life and progress rate of corrosion wastage in structural members of hull was considerable in several cases. Accordingly, it is very important to carry out timely maintenance after carefully studying and evaluating the status of corrosion wastage in order to prevent hull damage due to corrosion wastage.

View full abstract

The objective of this study was to propose a design and maintenance integrated (DMI) approach for the corrosion protection (CP) system for large scale floating structures. In thisDMI approach, potential measurement and numerical simulating analysis is mainly performed so as to suggest reasonable maintenance means for the CP system to secure the floating structure.
It is expected that, with circulative operations of thisDMI approach for some floating structures with similar CP systems, understanding for ageing behavior of the CP system will be nearer to the truth and then numrerical simulating analysis gives more reliable prediction. As the beginning of such circulative operations of thisDMI approach, in this paper a chasing simulating analysis for the CP system of a floating oil stock tank was carried out. From the results of this chasing simulating analysis, trend of the CP system and some essential correlations were clarified. Therefore, a reasonable guideline for operating a protection potential monitoring system during a long time can be suggested.

View full abstract

Fracture toughness of high strength steels of 490 MPa strength class under dynamic loading has been evaluated on the basis of the Local Approach. The critical CTOD at brittle fracture initiation decreases generally with increasing the loading rate. This is due to the elevation of a local stress near the crack tip at high rates of loading. The test temperature also exerts an influence on the critical CTOD. By contrast, it has been shown that the brittle fracture resistance evaluated in terms of the Weibull stress, an integrated stress over a highly stressed region near the crack tip, would be a material property independent of the loading rate and test temperature. Using the Weibull stress, the dynamic fracture toughness can be predicted from static toughness results at a given temperature. Predicted results of dynamic toughness of high strength steels were almost consistent with experimental data. For the characterization of flow properties of materials under different loading rates and temperatures, the rate-temperature parameter proposed by Bennett and Sinclair was useful.

View full abstract

The aim of this study is analytically to clarify heat transfer between combustion flame and the steel plate, that is to obtain heat flux, during line-heating process with both experiment and analysis.
Two dimensional temperature distribution within combustion flame is measured by laser induced fluorescence technique. For simplicity, jet instead of combustion flame is used to calculated temperature and velocity distributions. Using standard k-ε two equations model and assuming logarithmic law as wall function, convective heat transfer for turbulent flow is analyzed.
When the highest temperature within flame is used as temperature of jet at the nozzle exit, calculated temperature distribution of gas flow in near-wall region is similar to measured one. Measured and calculated time histories of temperature in the steel plate are in close agreement with each other. Numerical method to obtain heat flux is proposed during line-heating.

View full abstract

The utilization of the welded bond joint process for aluminum ship structures gives the advantages of the light weight design, an improvement of the fatigue properties and the good toughness for a shock load.
In this paper, results from an experimental and theoretical study of the welded bond joint on tensile strength are reported.
In the experimental investigation, tensile loading tests were conducted on the cross type welded bond test pieces.
In the theoretical investigation, FEM analyses by use of the computer program MSC/DYTRAN for simulating the destruction of the weld bond joint were conducted. The nonlinear analysis accurately represented the behavior of the destruction mechanism of the cross type welded bond joint.

View full abstract

The outstanding obstacle to estimate vibration level is that the precise damping of fluid and structure can not be estimated because the mechanism of the damping has not been explicated. Recently, the damping matrix based on dissipation energy caused by fluid viscosity is obtained, and the damping matrix is expressed more rationally by introducing modulus of velocity distribution β_w as a parameter to estimate the influence of the water depth. We developed a method to experimentally obtain β_w in the shallow water. But the method is not applicable to the deep water because of approximation in the shallow water.
In this paper, we propose a method, which can be applied to the deep water, and obtain β_wby using the method. It is known that β_w in the deep water equals 1.0, by comparing experiment with theory of dissipation energy at peak frequency of 2-node vibratioe. But, in this paper, β_w is obtained by considering not only a peak but also whole area of the frequency and moreover by using curve fit algorithm through not only one peak but also the whole peak. Concretely, authors incorporate reduced added matrixes into transfer function by means of modal structural modification analysis, and formulate the partial differential iteration method by using the transfer function The formulation and the results are described

View full abstract