A new computation method that can cope with three dimensional breaking waves in the framework of curvilinear body-fitted coordinate system is developed. The free-surface treatment in this study is based on the marker-density technique developed in the TUMMAC-VIII method, and the algorithm of solving the discretized N-S equation is that of the WISDAM-V method. By simulations with three hull forms, it is demonstrated that this new method can simulate three dimensional breaking waves around the body of arbitrary shape with certain degree of accuracy.

View full abstract

Grid dependence in Navier-Stokes solution for free surface flows around a ship hull is investigated through numerical experiments. A Navier-Stokes solver used is a multigrid, finite-volume code with the artificial compressibility approach. Convection terms are evaluated by the third order upwind scheme based on the flux difference splitting. Time integration is carried out by the Runge-Kutta scheme. Nonlinear free surface conditions are implemented and a computational grid is re-generated to follow the free surface deformation. Free surface viscous flows around a Series 60 (Cb=0.6) hull are computed by the solver using three grids with different grid density. Solutions are compared with each other and with experimental data and grid dependence of solutions is discussed.

View full abstract

A 3D finite-volume simulation method is developed for unsteady hydrofoils beneath the free-surface. A curvilinear grid system is fitted both to the body surface and the free-surface, and the bounday condition on the moving aileron is rigorously implemented. The unsteady characteristics of the flows about 2D and 3D hydrofoils with a moving aileron is well simulated.

View full abstract

Numerical simulation is made of the 3D breaking motion about a part of a semisubmergible containing a circular cylinder mounted vertically onto a lower hull in regular periodic gravity waves. The TUMMAC-VIII finite-difference method is newly developed for such a problem. By use of marker-density method the 3D breaking motion is approximately implemented in the framework of rectangular grid system. The generation of breaking waves by the interaction of incident waves with the structure is well simulated and interesting features of breaking waves are revealed with certain degree of quantitative and qualitative accuracy.

View full abstract

This paper describes the experimental and numerical results of a 45 deg. sweptback hydrofoil of NACA 0012 section and aspect ratio of 3, comparing them with those of an elliptical planform hydrofoil described in previous papers.
In the experiment, lift and drag coefficients, pressure distribution of the hydrofoil, velocity distribution around tip vortex and wake distribution were measured at some typical conditions. Flow field around the foil was also investigated by the numerical computation.
Following conclusions were obtained;
(1) Pressure gradient from root to tip appears on the back side of the sweptback foil because of its planform configuration. This causes an out-flow in the boundary layer.
(2) The sweptback foil is less prone to stall, compared to the elliptical foil.
(3) The tip vortex cavity of the sweptback foil is weaker than that of the elliptical foil. The sheet cavity on the sweptback foil is divided into several parts which rotated at their terminations in the same direction as the tip vortex. These results can be elucidated by the occurence of longitudinal vorticity due to the out-flow.

View full abstract

We develop a method to culculate the characteristics of supercavitating propeller making use of 2-D theories for partially cavitating and supercavitating wings. At first, we obtain the equivalent 2-D wing sections for the blade of a propeller operating in the noncavitating condition by using the quasivortex-lattice method (QCM). Then we apply 2-D cavitating wing theories to the equivalent 2-D wing sections and calculate the characteristics of supercavitating propellers by the concept of the wing element theory. Through comparisons between calculated results and experiments, we confirm the usefulness of this method.

View full abstract

A new technique for wall friction measurement, non contact and instantaneous measurement, is developed. The method is a kind of the so-called image measurement using oil flow on wall and image analysis technique. Oil flow owing to wall friction is very small and also strongly depends on way of painting. The possibility of the measurement is examined in a flat plate flow.
It is possible to measure the wall friction, if small displacement of oil can be measured. The spatio-temporal derivative method of image analysis is effective for this measurement. The ordinary oil film method for circulating water channel, based machine oil and liquid paraffin with carbon black powder, is employed here from the practical view point.
The present method was applied to a juncture flow around a circular cylinder on the flat plate, the so-called horseshoe vortex, in which it is impossible to use any probe for wall friction measurement. The limiting streamlines on stern hull are also measured. The result is agree with the visualized one with milk painting method.

View full abstract

When the water exchange or substance cycle in inner bays is considered, the permanent current flowing in one direction for longer time, which is composed of tidal residual current, density current, and wind driven current, is more important than the oscillatory tidal current whose magnitude is usually larger. In this paper, a multi-level model was applied to the calculation of permanent current and density variation in Tokyo bay and the change of the state of stratification and the accompanying current field from summer to autumn was simulated. In the numerical simulation, the observed field data such as wind, atmospheric temperature, etc. were used for input of the calculation and the results were compared with the observation values of currents, salinity and temperature. The simulation could describe the reactions of current and density field against the changing wind under the stratified condition pretty well. The long term variations of the vertical structure of salinity and temperature from summer to autumn were also predicted well qualitatively. Additionally, the effects of boundary conditions on the results of numerical simulations were also examined, It is clarified that the simulation results are much affected by the boundary conditions such as river discharge and the vertical structure of salinity at the outer ocean.

View full abstract

Kansai international airport was newly constructed and opened off Senshu city in the bay of Osaka. This airport was built on an artificial island newly reclaimed from the sea. Although other similar airport projects are also being discussed in Japan, application of reclamation system in the deeper sea is very difficult to realize because of the necessity of the massive sand for the filling of the land. Long period of construction and environmental impacts are also obvious defects of this system.
On the contrary, an airport with floating system clearly has advantages in some points. Especially, it is considered to give less damage to the environment than reclaimed one.
The authors wish to show numerical simulations by means of multi-level model to discuss the effects of artificial islands on ocean current. The boundary conditions of floating islands for numerical model are newly treated in this paper. By means of the multilevel model, differences of current pattern around floating and reclaimed islands are discussed in cases of simplified rectangular bay, Osaka Bay, and Tokyo Bay. It is clarified that a floating airport changes the current pattern less than reclaimed one.

View full abstract

A new numerical approach based on the 3-D panel method is presented for the estimation of the wave forces acting on a blunt ship advancing in oblique short waves. Unnatural numerical solutions had been obtained when we applied the 3-D panel method to the wave-making problem of the ship advancing in short waves, and reported comparing with experiments in the 1st report. The reason why the 3-D panel method gives such solutions was not yet made clear, even though some investigations about that problem have been carried out in the 2nd report.
In this paper, a new numerical approach has been developed which is free from the aforesaid unnatural solutions. The Spline approximation is effectively applied to the expression not only of the geometry of the ship surface but also of the velocity potential distribution on it, and the integral equation is solved directly with respect to the velocity potential itself instead of the source distribution. Employment of the Spline approximation, moreover, makes it possible to take into account the line integral term and the effect of the steady distrubance on the realistic ship surface.
After the numerical test of the present method by comparing with the other results based on the isoparametric approximation or the analytical method, the present method is applied to our wavemaking problem, and it is confirmed that the unnatural solutions can be eliminated perfectly by employing the present method.

View full abstract

Based on Bayesian modeling procedure, a stochastic procedure is introduced. Using this procedure, directional wave spectra can be estimated by using the ship motion data which were recorded during the ship underway. The prior distribution in this procedure consists of two parts. The one is a stochastic constraint so called Gaussian smoothness prior distribution. And the other one is adopted to avoid excessive estimation for the directional wave spectra. Furthermore, the encounter frequencies are transformed into the true wave frequencies even in following seas.
In order to examine the reliability of the proposed procedure, the onboard tests were carried out. And, the possibility of this procedure can be shown and problems encountered on applying are reported.

View full abstract

A new method to estimate the directional wave spectra from motions of a ship running in the directional seas is developed in this paper. The method is constructed based on the Parametric Method assuming the linearity between incident waves and motions of a running ship. The accuracy of the present method is investigatd by numerical simulations and is shown to resolve different directional seas consistently from motions of a running ship, particularly in the following sea condition. The method is also used to analyze experimental data obtained from a free-running test of a VLCC model and is shown to be a powerful tool for the directional wave estimation from motion of a running ship.

View full abstract

For the design of ships or offshore structures, the long term wave statistics is very important. Usually, ship data namely the visually observed data on board are used for such purpose. But, the density of ship data are not uniform over the sea area and there is another problem on the accuracy of visual observation.
Recently, a new wave statistics of North Pacific Ocean was published by I. WATANABE et al. (1992), and this statistics includes hind-casted long term wave data other than the conventional ship data and measured data by buoys.
There are differences between the characteristics of ship and hind-casted data, but the effects on long term ship responses are not clear.
So, this paper investigate the influences of the differences of wave statistics on long term ship motion and long term ship operability.

View full abstract

Interim Standards for Ship Manoeuvrability was adopted at 18th IMO Assembly in Nov. 1993, and ship manoeuvrability becomes important from the viewpoint of human life safety and ocean environmental protection. In relation to these conditions, it is necessary to examine manoeuvring performance under the external force of waves.
In this report, Fujii-Takahashi's semi-empirical equation which can calculate the added resistance in short wavelength region is extended to calculate the surge and sway forces and yaw monent caused by directional spectrum waves having short wavelength. By introducing these forces to M. M. G. model, simulation for zigzag test in short wavelength waves is carried out and compared with the free running model experiment in the towing tank of Yokohama National University, in which directional spectrum waves can be generated. Further, manoeuvrability indices are obtained by K-T analysis of simulation results using 1st order approximate equation.

View full abstract

An actively controlled anti-rolling system has been developed with a view of reducing the rolling motion of a ship by the movement of the mass controlled by actuator. It is of hybrid type combining the pendulum-base passive type with active type driven by a relatively small electric motor. The hybrid type has merits of (a) providing almost the same damping performance as active type (depending solely on actuators for obtaining controlled movement of damper mass), while requiring only a much smaller control force for moving the damper mass, and (b) continuing to function solely on its passive mechanism in the event of power failure. This system consists of a sliding mass on the rail shaped in a circular arc, and the compact, passive pendulum mechanism is realized that does not require a suspension structure such as a simple pendulum or spring mechanism. The driving force to control the movement of the damper mass is imparted from the electric motor through reduction gearing connected to a gear and pinion mechanism. The LQ control theory has been adopted for controlling the damper mass. At-sea experiments were performed with a ship (weight approx. 190t) to verify damping effects of the hybrid anti-rolling system. The rollings were reduced to about 1/3 in beam seas under the condition that the ship was stationary. The performance was further compared between hybrid type and passive type. Hybrid type promised better damping performance by 15% than passive type. The hybrid type also attenuated rollings with a forward speed of ship to about 1/2 in following seas, in the case of which passive type presented difficulty of providing damping effects markedly. Good agreement was obtained between measurement and calculation.

View full abstract

When a heavy facility or equipment is set on or picked up from the sea bed in deep sea area, it seems to be difficult to keep safety anytime throughout the long operation period, because of the risk of encountering rough sea conditions.
Although conventional heave compensator using air spring mechanism has been already applied in such an operation, the compensation performance of such a passive system would not be sufficient against large heave motions in rough sea conditions, and dynamic hook load would come to dangerous level, especially in case of resonance condition of the air spring due to long period heave motions.
In this research, it has been studied to stabilize the position of suspended heavy mass in sea water and reduce the dynamic hook load by adding active control force to the passive air spring mechanism.
Simplified model tests and corresponding theoretical simulations have been performed and the results are shown in this paper.
Tested model consists of vessel part, suspended mass part and heave compensation mechanism part. The responses of those were measured in various wave conditions. Heave compensation mechanism was simplified using AC servo motor controlled so as to simulate the performance of the ideal air spring and also to simulate hybrid mode together with active control force. The command of the active control force is derived from the signals of vertical accelerometer sensing vessel heave motion, and pulse generator sensing relative motion between the vessel and the suspended mass.
The following conclusions have been derived by the model tests and corresponding theoretical simulations.
(1) Variable hook load in hybrid mode (with active control force) has been reduced to about 1/10 compared with the case of fixed mode, and about 1/3 compared with the case of passive mode (using air spring only).
(2) Theoretical simulations have shown good agreement with the tested model results.

View full abstract

Motion of offshore structure in damaged condition is important from the view point of safety. For the response estimation of damaged condition, some research works are reported, but most works treat only heeled or trimmed condition. The combined heeled and trimmed condition seems not to be treated.
One reason is that there are some considerations that even in damaged condition, the frequency response functions will not be changed largely and another reseason will be that this condition is very complicated in space and not only calculation but also experiments are very difficult.
But by experiments in damaged condition, it is found out that some responses are changed largely within the range of ocean wave frequency.
Here, we challenged this problem and a computer program was developed in order to estimate linear (infinitesimal) motion responses of floating structures especially in combined largely heeled and trimmed condition in waves and compared with experiments. Calculated results showed relatively good coincidents with experiments.

View full abstract

One of the serious problems of the Tension Leg Platform (TLP) is the springing phenomenon in the mooring system due to super-harmonic resonance of heaving motion in high frequency zone. Second-order sum-frequency vertical wave exciting forces are closely related with the springing phenomenon. When the second-order vertical wave exciting forces are small at the resonance frequency, the springing response will be reduced.
In the present paper, column footing type floating bodies whose first and second order vertical wave exciting forces are very small at the specific frequency, are applied to the TLP. The properties of the springing responses on column footing type floating body are computed and these results were compared with experimental results, so that the effects of springing responses are made clear in the tether of tension.

View full abstract

It is known that wave-induced forces on horizontal cylinders are significantly decreased due to a circulating flow generated around the cylinder at low Keulegan-Carpenter numbers. An estimation method of wave forces on a two-dimensional horizontal circular cylinder using a simple vortex shedding model has been proposed by the authors. For estimating the wave forces on horizontal members of offshore structures like lowerhulls of a semisubmersible, however, it is necessary to clarify three-dimensional effects, such as the end effect or the juncture flow effect.
In this paper, the end effect and the effect of columns on wave forces on a lowerhull submerged in regular waves with its axis parallel to the wave crests were experimentally investigated. A two-dimensional lowerhull cylinder and finite-length lowerhull models with and without columns are used in force measurements, velocity measurements and flow visualizations. The results of the experiments showed that the wave force reduction decreases in the case of the finite-length lowerhull model with columns due to the disappearance of the circulating flow at the end and the junctions with columns. The results also demonstrated that the vertical wave forces on the finite-length lowerhull model with columns are asymmetric because of the flow accelerated between the columns.

View full abstract

A part of authors have made some studies on wave exciting forces on large floating structures composed of multiple body elements for ocean space utilaization such as floating airport, taking hydrodynamic interaction into accout. In the previous paper, some characteristics of interaction effects from computed results were reported, including the followings:
1) Peaks of response function of wave exciting force of inner body elements are larger than of outer ones.
2) There is a kind of singular frequency where wave exciting force of body elements in head sea condition reach six times as high as that of a single body as a result of interaction effects.
Model experimental results are shown and two points above are confirmed in this paper. It is proved from both experimental and computed results that at the singular frequency there appears standing wave. It is also discussed that the singular frequencies appear only in head sea condition.

View full abstract

The hydroelastic response of a large and long floating offshore structure moored in short-crested waves by tensioned tether buoys is studied in this paper. The applied model is a psuedo-elastic body by connecting each rigid unit part at deck with elastic members. The main concern of this paper is the distribution of vertical displacement in short-crested waves.
Firstly, the deflections of deck part among regular waves are calculated based on the method of combination of modal analysis and hydrodynamic force analysis and the results are compared with measured values at multi-points. The deflections in short-crested waves are discussed through experimental results and theoretical preditions. The response spectra and distributions of variance of vertical displacement along centerline of structure are shown for both long and short-crested waves and the influence of flexibility on deflection has been discussed.
In addition, motion of TL-buoy and tether tention in case of single buoy and the buoy moored to structure are investigated. Finally, the interactions between buoy and structure are discussed from the view point of realization of mooring system for large and long floating offshore structure.

View full abstract

Mooring motions works on oceanographycal measurements such as acoustical position fixing, acoustical tomography, and so on. Therefore, acoustical measurements of mooring motions are required to obtain good accuracy on acoustical oceanographycal measurements. In this paper, a drag coefficient of chained glass balls to get a buoyancy force of a mooring system are determined on the basis of experimental data in the sea. Mooring motions are investigated with lumped mass method using its drag coefficient. A taut mooring motion draws a circle in a rotating current, on the other hand it draws a track like a numeral of 8 in an elliptically rotating current. Displacements of a taut mooring system can be estimated from current velocities based upon two kinds of simple dynamical equivalent models. This makes easier to measure mooring motions than acoustical method.

View full abstract

The hull structure of conventional tension leg platforms (TLPs), operating in the Gulf of Mexico and the North Sea, consists of pontoons and four or six columns supporting a wide deck structure on which topside facilities needed for offshore petroleum exploration are installed.
In this paper, authors propose a simple TLP having one column structure which can be used, for example, as a minimum size satellite platform for a marginal oil field. Its hydrodynamic behaviors are discussed on the basis of results of theoretical analysis and measured data obtained through one year at-sea measurement of a prototype TLP (water depth: abt 61.0m) performed by JOIA (Japan Ocean Industries Association) according to the committment from the Ministry of International Trade and Industry.
The main results of the sea test are as follows; 1) tension fluctuation of each tendon of one-column type TLP are almost same except short wave period range, 2) the relation between the maximum wave height and significant wave height approximately follows the Rayleigh distribution, 3) measured motion responses and tendon tension fluctuations coincide well with theoretical calculation, 4) the assumed design wave height approximately agrees with the maximum wave height estimated from the measured wave data, 5) the minimum tension of each tendon can be kept positive for given return period.

View full abstract

This paper presents a research on how to keep the riser angle of iclination within the admissible extent in the deep sea drilling operation, from the viewpoint of the positioning of the drill ship and the effects of using actuators attached along the riser.
In the operation of deep sea drilling where the water depth is deeper than 3000m, it is very important to keep the top and the bottom angles of the riser inclination smaller than 4 degrees to prevent the drill strings inside the riser from touching and breaking the riser. To satisfy this criteria and to reduce the payload, the riser is tensioned at the top with a tensioner, the weight in water is reduced by attaching the buoyancy module, and the ship position is controlled by using the DPS against the disturbance resulted from current or wave. However, because of the huge mass and the very low rigidity, the dynamic response of the riser is generally so weak that the riser cannot follow the ship motion well, and that it resulted in decreases of availability of the system.
Based on the above circumstances, the equations of coupled motion of the ship and the 4000m riser are formulated. With these equations, the optimal control is formulated. Then simulation results by Newmark-β method are shown for a set of parameters of 750tf, 850tf top tensions and 1kt, 2kt currents. From this study it becomes clear that an actuator with about 270kgf thrust can effecitively improve the response of the riser bottom and top inclination angles by the ship motion. And the inclination angles of the riser can be manipulated by the ship positioning within the permissible range. Furthermore, it is clarified that the two actuators with maximum thrust of 500kgf can effectively control the dynamic responses of the riser, and that positioning the ship with 850tf riser top tension, the angles of inclination at the riser top and bottom can be kept in the desirable range of less than 2 degrees even when the current speed is 2kt.

View full abstract

In this paper, a computing method for water impact of arbitrary shaped bodies will be proposed. We use body-fitted coordinate system to model the arbitrary profile of the rigid body surface and the transient deformation of the free surface is solved by the fractional volume of fluid method. With this combination of the numerical techniques the rigid surface and the free surface boundary conditions are naturally treated at the intersection of both surfaces that improves remarkably the stability and accuracy of the numerical simulation.
For the purpose of verifying the present computing method, classical problems of water impact of wedges and circular cylinders are studied. Moreover, water impact of a ship bow section is simulated to illustrate the applicability of the present computing method to practical ship design problems. In these basic studies, we considered the cases in which the body enters an initially calm water surface with a constant velocity. The results of the simulations are examined against the existing analytical theories and published experimental data.

View full abstract

Accurate estimation of ship maneuvering ability is required since the IMO adopted the standard for ship maneuvering ability. And the accuracy of estimated hydrodynamic forces acting on a ship is most important to the estimation of ship maneuvering ability. A numerical solution method for the Navier-Stokes equation based on the finite-volume approach is employed in this study and viscous flow field around two types of full ship, with aft part differing from each other, under oblique towing condition is computed. Model experiments are also conducted to validate numerical results. In consequence of comparison between computation and experiments, it is found that hydrodynamic forces, pressure distribution on hull surface, and flow field in the neighborhood of the ship stern agree satisfactorily well with each other.

View full abstract

From the view point of marine safety, it is of importance to evaluate the ship manoeuvring motions, such as the berthing in a harbour, the drifting behaviour of a disabled ship. For the prediction of ship manoeuvrability with high accuracy by numerical simulation, it is required to estimate the accurate hydrodynamic forces acting on ship not only in relatively high speed motions with small drift angle but also in low speed motions with large drift angle. At larger drift angles the flow around the ship's hull will be complicated due to vortex shedding and separation, but it is considered that the manoeuvring characteristics of a ship will be significantly influenced by hydrodynamic forces, initiated by the cross flow underneath a ship from the inflow to the leeside, and cross flow effects become dominant as the ship's drift velocity becomes relatively larger compared to the ahead speed.
In this paper, special attention is given to the cross flow around each cross section of a ship during oblique motions with large drift angle, from lateral shifting motion to ahead motion. A distribution of bound vortices and free vortices is used to represent 2-dimensional separated flow around ship's section. The separation points are defined from velocity distributions for non-separated flow. The representation of the latest free vortex shed from a separation point at any incidence is discussed. Numerical results for a container ship hull in drifting motions are presented. By comparing with the measured results of oblique towing model tests at increasing forward speed, the calculated lateral forces and yaw moments acting on a ship agree well with experimental ones. Therefore, this approach will be useful for more accurate prediction of ship's manoeuvrability at low speed.

View full abstract

Recent maritime disaster of a large tanker often causes a serious oil pollution. To prevent such disaster, some kind of manoeuvring standard has been demanded over the world. International Maritime Organization (IMO) has been working out to settle the standard in these several years. As the result, the interim standards for ship manoeuvrability (A. 751) was adopted in November 1993. It is applied to the ships of 100 m in length and over, and every chemical tankers and gas carriers, which are constructed on or after 1 July 1994. This interim standards cover turning ability, initial turning ability, stopping ability and course-keeping/changing ability. The criterion of stopping ability is defined that the track reach in the full astern stopping test should not exceed 15 ship lengths. Ship designers should evaluate this stopping distance in early design stage.
In this paper simple mathematical models for predicting stopping distance and time are presented, and these models are confirmed by a number of full-scale trial results. Then, it has been clarified that the non-dimensional stopping distance and time are almost proportional to (displacement of ship/main engine horse power) and the square of initial ship's speed.
Through this research, it is also pointed out that the stopping distances of many VLCCs have already exceeded the above limit. This criterion becomes very severe particularly for such large ships. Since the application of the standard for stopping distance regulates the minimu required horse power simultaneously, the author would like to emphasis for the criteria to be re-examined and finalized under the world-wide agreements.

View full abstract

A general methodology for the assessment of the inherent manoeuvring characteristics of ships, is presented. The methodology, which takes into account the non-linear nature of manoeuvring motion, is built on a specific set of manoeuvring requirements, translated into suitable criteria and measures of performance. These are later integrated in a hierarchical procedure which allows for combining effectively conflicting characteristics and producing a picture of overall manoeuvring capability. The problem is established as a typical decision problem in the presence of multiple objectives and conflict, with the first part of the paper devoted to the theoretical foundation of the general assessment model. Application is undertaken for the case of ferries, with consideration of a wide range of design parameters. As a result of this, areas of good or poor manoeuvring performance are identified. A clear distinction is made between the need to turn in limited space and the need to respond quickly to rudder commands. The methodology can be used in early ship design as well as in developing/optimising vessel type-specific manoeuvrability standards.

View full abstract

In ship structures, diminution and deterioration of structural members are serious problem. And the diminution of structural members have an effect on the deterioration of structural performance of ship hull. Hence it is necessary to investigate the relation between the diminution of structural members and the deterioration of structural performance of ship hull.
This paper describes followings.
(1) Statistical characters of annual corrosion rate of each member.
(2) Deteriorating rate of section modulus of hull due to diminution of structural members.
(3) Buckling strength of stiffened panel considering the effect of structural diminution and deterioration.

View full abstract

The multiplier method is applied to structural optimization of double hull tanker in this paper. The hull construction cost of one tank hold length in the midship segment of double hull tanker is taken as an objective function. Since there are standard values for plate thickness, the plate thickness can not be changed continuously. In order to solve this problem, two optimization processes are introduced to deal with those discrete design variables. In the first optimization process, it is assumed that all design variables vary continuously. After the first optimization process, the plate thickness is set to the nearest standard value and the rest of design variables is decided in the second optimization process. In this study, the effect of standard plate thickness on optimum design and the optimum values varied with fillet welding price rate as well as the influence of limiting value for longitudinal spaces on optimum design are investigated numerically.

View full abstract

To know the force and displacement in the case of head on collision is quite important in designing safe ship. In this paper, the Yang & Caldwell's technique of deriving simplified equation of mean collapse force based on the Wierzbicki's collapse mechanism is improved and applied to the ship bow structures. The ship bow structure is divided into simple elements of L shape, T shape and + shape. The absorbed plastic energy in crushing of the these elements is derived. The method is extended to the case in which plates of several different thickness exist. The collapse modes are chosen so that the collapse force becomes minimum, while maintaining the consistency between elements. The absorbed energy of each elements are summed up to be the total absorbed energy in the crushing of the bow structure. The number of waves between frames is determined to make the collapse force minimum. The crushing force is computed as the absorbed energy divided by the crushing length. Also extension are made to the non-prismatic structures.
The finite element analysis using MSC/DYTRAN is applied to verify the validity of this approach. The comparisons are made with experiments by Nagasawa et al. The good agreements of this simplified equation with numerical analysis and experiments are observed.

View full abstract

It is well known that a thin flat plate undergoes secondary buckling after initial buckling took place and the deflection of the initial buckling mode has been developed. The plates in actual structures are usually accompanied by initial deflection due to welding. So, the influences of initial deflection have to be considered when the secondary buckling strength of plate members in actual structures is evaluated. In this connection, series of elastic large deflection analyses are performed applying the analytical method and the finite element method. In the analyses, symmetrical and non-symmetrical initial deflections are assumed based on the measured results. It has been found that:
(1) Although the secondary buckling of a thin rectangular plate with initial deflection due to welding generally takes place in a symmetrical mode, the non-symmetrical components of deflections cannot be ignored in the analysis, since there is an interaction between symmetrical and non-symmetrical components.
(2) The initial deflection reduces the secondary buckling strength when it takes place far after the stable deflection mode has been developed. The reduction becomes significant with the increase in aspect ratio of the plate
(3) The lower bound of secondary buckling strength increases with the increase in the magnitude of initial deflection, because some components of initial deflection resist the initial buckling deformation to take place.
(4) The inplane rigidity of the plate is significantly reduced by the secondary buckling.
(5) The collapse mode of plates under compression is influenced by the secondary buckling, particularly near the aspect ratios at which the mode of primary buckling changes.

View full abstract

The general governing equations of equilibrium and dynamic stability for thin shells subjected to follower loads and undergoing large deformations have already been detailed in the previous reports for the general shell defined in a monoclinically convected coordinate system. Various numerical results were obtained to show the feasibility and applicability of the present approach. The existance of a threshold point of dynamic stability for a particular shell was illustrated to prove the effect of disturbances on the behaviour of shells near the point of critical static stability.
The present report expands on this approach elaborately for a wide curvature range from shallow to deep shells and obtained the characteristic curves of dynamic stability threshold for partial cylindrical shells and partial spherical shells. It is found that while the spherical shells generally fail dynamically before the static critical limit, cylindrical shells retain their dynamically stable behaviour a little more beyond. The backgrounds of geometrical and strength aspects during the deformation process are also analyzed. The existance of a post-critical stability phase for shallow shells, which had been hypothesized in the previous report is investigated here and conclusive numerical evidence to this effect is obtained. The practical aspects of a shell design for the elastically feasible range of ultralarge deformations can be visualized here, for which there may be a need arising in the futuristic constructions in the sea with dynamically deflecting outer shells that may be allowed to flex safely like the tarpaulin tent in the wind.

View full abstract

Offshore structures should have a high degree of structural safety even under extreme enviromental loadings. The authors have studied a composite steel-concrete structure of sandwich system for the strength members of huge offshore structures, where concrete is placed between steel plates.
In the previous papers, the authors carried out both experimental and theoretical investigations into the strength of the composite structure. It was clarified that the composite structure has various excelent properties: The ultimate load-bearing capacity of the composite structure is very high and it can absorb a great deal of energy until failure under any type of loading conditions.
In the present paper, the authors carried out both experimental and theoretical investigations into the dynamic toughness of the sandwich composite structures. Experiments were carried out using the two dimensional models of composite structure under high speed loading. A nonlinear analysis was developed to predict the toughness of sandwich beam under dynamic load, where the material non linearities of both concrete and steel with strain rate were taken into account. The nonlinear analysis accurately represented the behavior and toughness of the sandwich beam structure.

View full abstract

In this research, a 3-D thermal-elastic-plastic FEM is developed to simulate the out-of-plane deformation caused by one-sided automatic submerged arc welding. Efforts have been made to improve the reliability and accuracy of the FEM calculation. To prevent the phenomenon of transverse-shear-locking, which is observed as overestimation of the stiffness of a plate for a bending problem when solid elements are used, reduced integration method is employed and proven to be effective. A special type of spring-element is utilized to deal with the contact problem between the plate and the working table.
Using the 3-D FEM, the effects of various factors, such as gravity, the method of supporting, magnetic constraint and force from the backing near the welding line have been investigated. The results from the numerical simulation can reasonably explain the phenomena observed from the real situation in shipyards. In the case of FCB automatic submerged arc welding which has the feature of large heat input, out-of-plane welding deformation is mainly affected by constraint condition. The gravity of the plate and the backing force have great influences on the shape of angular distortion. Magnetic constraint can reduce the welding deformation effectively.

View full abstract

In dealing with structures of complex configurations such as ship structures, generation of finite element model is one of a major bottleneck. To overcome this problem, an object oriented finite element modeling system, MODIFY, has been developed by authors^1)2)3)4). In ship structural design, many kind of finite element analyses are carried out for different geometric models. For example, whole ship analyses with shell elements are carried out in early design stages, and analyses of welding part with solid elements are done for detailed design stages.
In this paper, several improvements of the finite element modeling system are introduced. First, a solid-object is defined as a geometry-object to deal with solid models. second, a two-manifold B-Rep model is used for a geometrical representation of all geometry-objects and analytical-condition-objects. Third, a 3 D solid mesh generator based on the front method is developed. Finally, by using an object oriented database for construction of the system, it becomes possible to deal with a large number of object oriented data in the system.
Several examples of a finite element model generated by the system are shown.

View full abstract

Structural analysis of large scale structures like ship is usually carried out on the models of several different scale level, varying from whole ship, cargo hold parts and to the detail structures. Currently, the finite element models for different analysis level is generated from different geometric model, mainly because the unit geometry of mesh generation is different in each level of the analysis. In this study, a prototype system is developed that automatically generates finite element models for different analysis level from single ship geometry data.
For this purpose, two steps are taken. First is the construction of the database that includes the degree of detail for each product part and hierarchical structures of product parts. Second is the computation of intersections to produce the mesh generation part automatically from the product part.
As an example for demonstration of the present prototype system, input data formulations for different analysis scales are carried out on a 1/2 hold model of a tanker.

View full abstract

This is the second report on dynamic characteristics of an underwater large rotational shell. This paper focuses on the seismic response of the shell to horizontal ground motion. The natural modes which were obtained in the previous paper are used to estimate the seismic responses, which is based on the theory of power spectrum. The response of the shell in water is very different from that in air because of the effect of large added mass and the seismic input spectrum, and so this kind of numerical analysis of fluid-structure interaction problem has to be analyzed numerically in advance in the design process. And the features of dynamic characteristics of the half drop shaped shell that authors proposed are discussed herein as well.

View full abstract

There has been hot discussions on the characteristics of wave-induced pressure fluctuation on the side shell of VLCCs, since considerable number of fatigue cracks were found on side-longitudinals of second generation VLCCs. In Japan, it was mainly dealt with by SR-216 Panel, where a full-scale measurement on a typical second generation VLCC, tank experiments on a scaled model and theoretical investigations were carried out. The main themes of the discussion seem to be whether characteristics of pressure fluctuation is predictable even for small values of wave-length to ship-length ratio like 0.1-0.3 and how peak-to-trough value can be obtained of wave-induced pressure fluctuation on the side shell close to the load-water-line.
Tank experiments on a scaled-model ship were carried out at Nagasaki Experimental Tank of Mitsubishi Heavy Industries to answer the questions, and it is shown that measured results agree with calculated results by strip method after a simple correction on non-linearity effect caused by the crossing of free-surface. Results of long-term prediction with the same correction also show good agreements with the results of the full-scale measurements.

View full abstract

This paper deals with numerical analysis of sloshing problems for the middle-sized double hull tanker. A numerical simulation method for non-linear liquid sloshing in the cargo oil tank is proposed by making up a new sloshing cord based upon the SOLA-SURF scheme and a simple two-dimensional sloshing cord. The new sloshing cord has many functions for analyzing sloshing liquid violence motions in the octagon-shaped tank fitted with centre girder, deck centre girder, deck side girder, horizontal girder etc. A series of numerical simulations are executed for cargo oil tanks of a middle-sized double hull tanker and of a single hull tanker sailing at forward speed in regular waves, and numerical calculations are executed for four model tanks in order to compare with experimental results too. By numerical simulations, it is confirmed that the sloshing cord is useful for analyzing non-linear sloshing and for estimating sloshing pressure in liquid cargo tanks of a double hull tanker.

View full abstract

Side-to-side tanks are the most economical for an arrangement of cargo tanks of middle-sized tankers with double-hull structures. Liquid sloshing is a matter of some concern for these tankers. Experimental study is carried out on sloshing loads on the middle-sized tankers with double hull structures. A ship motion simulator is developed for the sloshing experiment, which gives a tank model 4-degrees of freedom motion (quasi-3D motion), rolling, heaving, swaying and pitching.
Tank models are of two kinds, one is two-dimensional and the other is three-dimensional. Basic models of 2D and 3D are rectangular shape with no inner structural members and are modified to investigate effects of bilge hoppers and shoulder tanks, effects of center-line girders, and that of horizontal girders on a transverse bulkhead.
Sloshing impact pressures are measured at inner surfaces of the models and on the inner structural members. Model tests are carried out for regular rolling, swaying motion and irregular rolling motion. In each case, fill depth and period of motions are changed. Period of each measurement of the model tests are 300 seconds. Peak values of impact pressures are detected, processed statistically and evaluated by 1/10 highest mean values.
As a result, basic aspects of sloshing loads are shown through the experiments: effects of fill rates on resonance frequencies, effects of inner structural members on sloshing impact pressures, and effects of swash bulkhead to reduce sloshing. Measured sloshing loads in irregular waves by model tests are compared with the estimates computed by a linear probabilistic theory. They agree comparatively well, and the discrepancy between them seems mainly due to the nonlinearity of sloshing phenomena. The relation between impact pressure and loads still remains a subject to be studied.

View full abstract

In these days, direct strength calculation methods has been increasingly and effectively used in ship design for preventing structural members from buckling and yielding. However, regarding to fatigue strength assessment, it is generally known that it is not clear how FEM stress output ought to be used and that how FEM input ought to be prepared for fatigue strength estimation sake.
In order to attain direct strength calculation, in this report, it is emphasized that an adaptation step is needed from a SN curve as a material strength characteristics to equivalently converted SN curve in which the vertical axis value is expressed by a local stress range and is made compatible to stress calculation tool.
This is because that SN curve is the essence of fatigue life and that the vertical index of a conventional SN diagram, that is nominal stress range, is difficult to calculate of an actual ship structural details due to the complexity of stress distribution and the non-flat shape of the actual optimized details.
And also in the study a consideration was made on the limit of FEM modeling, and eventually it is made clear that the plate element FEM, which is not able to allow the plate thickness to change during calculation, may underestimate the stress concentration factor when it is accompanied with large sectional deformation of plate and weld bead. An example of the case was verified by the series fatigue test of T-shaped fillet weld joint by adopting the three point bend loading with small distance of supporting ends compared with the plate thickness.

View full abstract

In the previous paper by authors, fatigue limit and the stress at the transition point, the point where the deformation mechanism of the material changed from visco elastic into visco elastic-plastic behavior, of mild steel were found to be almost identical under constant-amplitude loading conditions, and the visco elastic-plastic constitutive equation which represent cyclic stress-strain response of mild steel was derived.
In this paper, complete compression-to-tension load-controlled axial loading test programs-constant amplitude and random loading with an exponential distribution of stress amplitude- were performed for mild steel plane specimens. From the test results, coefficients in the visco elastic-plastic constitutive equation derived in the previous paper are given as a function of both the value of applied stress and their cycles. Then, the correlation between fatigue limit and the stress at the transition point is discussed.
As a result, following four main points are obtained.
1) There is a good correlation between fatigue limit and the stress at the transition point under variable-amplitude loading conditions.
2) The transition point stress decreases when stresses larger than fatigue limit (corresponding to the transition point stress) are applied, and then converges to a certain value, depending on the value of the larger applied stresses and their cycles.
3) Given the condition of variable amplitude loading, the transition point stress can be calculated using the derived constitutive equation.
4) Using the above results, a method for determining fatigue limit under variable-amplitude loading conditions is proposed.

View full abstract

Newman showed crack opening/closing model during fatigue crack propagation under uniform stress field which was based on the Dugdale model, but modified to leave plastically deformed materials along the crack surface as the crack advances. In his model, however, he neglected the change of stress distribution ahead of a crack during loading process in the calculation of crack opening load. Moreover he also neglected the elastic deformation for a materials along the crack surface. Therefore he introduced plastic constrainted factor of 3.5 which was larger than the appropriate value of 1.21 obtained from comparison between COD's by elasto plastic FEM and those by Dugdale model for center notched specimens for the purpose of fitting experimental crack opening load.
In this paper, Newman's model is modified to solve his neglected problems mentioned above and is expanded to arbitrary stress distribution field for the purpose of obtaining re-tensile plastic zone's generated load (RPG load). Moreover we consider plastic shrinkage at newly generated fracture surface during unloading process which corresponds to a part of deformation due to emancipation of acting force with a magnification of yield stress at an immediate growth of a crack in the region of fictitious crack in Dugdale model.
Some calculation results are shown for understanding the effect of the plastic shrinkage at newly generated surfaces and the change of loading. Simulated results suggest that the effect of stress ratio and the effect of delayed reterdation due to decreasing maximum load become small as the plastic shrinkage becomes large.

View full abstract

Fatigue crack propagation tests with various stress ratios and stepwise decreasing maximum load while keeping minimum load are carried out for center notched tension specimens. In these tests, RPG load (Re-tensile Plastic zone's Generated load), crack length and number of cycles are measured by using our newly developed measuring system in the 1st report.
We also calculate RPG load by using our developed simulation model of crack opening/closing phenomena based upon Dugdale model in the 5th report with various shrinkage coefficients α of plastic deformation due to emancipate internal force in newly generated crack surface at an immediate crack growth. Then we obtain α value for the steel used with comparing between simulated RPG load and measured one for the testing condition of R=0.05. By using the obtained α value, RPG loads for other stress ratios (R=0.3 and 0.5) are calculated by using the simulation model. As a result, simulated RPG loads are in very good agreement with measured RPG loads for R=0.3 and R=0.5.
Then a crack growth curve for each testing condition is calculated by using ΔK_RP from simulated RPG load and also shows in very good agreement with measured crack growth curve for each stress ratio.
RPG load increases just after decreasing maximum load and reaches maximum and then decreases gradually as a crack advances in the cases of decreasing load with small magnification. Then delayed deterdation effects appear and delayed life increases with larger decrement of maximum load. It is also comfirmed that simulated RPG load using above α value is in good agreement with measured RPG load for these cases. It becomes also clear that crack growth curve in case of decreasing maximum load can
be assessed quantitatively by using ΔK_RP based upon the simulated RPG load.
Moreover it is confirmed that larger decrement of maximum load in stepwise loading leads to stopping a crack even when final maximum load with large decrement of maximum load is larger than RPG load. This is because RPG load increases rapidly just after decreasing maximum load and approaches to final maximum load in case of large decrement of maximum load both in experiments and simulations. Simulated crack growth curve by ΔK_RP using calculated RPG load in this case is also in good agreement with experimental one.
Therefore it can be expressed that the condition of stopping a crack is ΔK_RP_??_0 for a long crack. In other word, it becomes clear that ΔK_th or (ΔK_eff)_th is not a material constant.

View full abstract

In order to investigate the fatigue crack growth behavior under various biaxial loading conditions, twelve pieces of welded fatigue test specimen are designed in such a manner that the repeated nominal biaxial stress range ratios vary from 0.2 to 2.0 at the orthogonal joint of the specimen. It is of great interest to understand the sharp crack turning under certain biaxial loading conditions, where a crack changes its propagating direction approximately perpendicular to the initial direction. Various morphological modes of fatigue failure are actually obtained by experiments in the present work.
It is also important to estimate an accurate crack path by numerical simulation for the proper understandings of fatigue crack propagation behavior. A step-by-step finite element approach has been proposed for fatigue crack path prediction, in which a two-dimensional cracked domain is remeshed by the modified quadtree mesh generation algorithm, and the stress field ahead of the current crack tip is analyzed by taking account of the higher order stress field parameters. A curved crack increment can be determined by using the first order perturbation solution together with the local symmetry criterion. The crack tip extends for a certain increment size, and the procedure is repeated for the next step. The present method is applied to the simulations of the fatigue test specimens. It is found that the simulated crack paths are in fairly good agreement with the experimental results and that the curved paths of fatigue cracks could be greatly influenced by the biaxial stress ratios, the loadings, and the geometry of the stress concentration corner.
Although fairly good agreements are obtained between the simulated crack growth lives and the experimentally obtained ones for simple three point bending specimens in a previous paper, significant discrepancies are sometimes observed in the present study. We have not had the definite answer to the question, but one of the sources of the disagreement could be the compressive welding residual stress in the test specimens, which could significantly reduce the effective stress intensity ranges of some specimens.

View full abstract

The influence of elastic modulus of weld matal on fatigue strength of welded joints was studied for boxing weld joints under tensile stress. Boxing weld joints with weld metals of carbon steel (elastic modulus: 206GPa), austenitic stainless steel (193GPa) and Invar (150GPa), an Fe-36% Ni alloy, were tested. Furthermore, elasto-plastic finite element analyses (EP-FEA) were carried out on fillet weld joints with weld metals of different elastic modulus to study the strain distributions in the vicinity of the weld toe.
The results are summarized as follows:
(1) Fatigue strength of welded joints with weld metal of Invar is higher than those of carbon steel and austenitic stainless steel. Enhancement of fatigue strength of joints with Invar weld is especially large at high stress ranges for crack initiation life.
(2) The lower elastic modulus of weld metal of fillet weld joints is, the smaller the maximum principal strain in the vicinity of fillet weld toe by EP-FEA is. This phenomenon is explained by the yielding in stress concentration region of weld metal due to the displacement-constant condition in the vicinity of weld toe.
(3) The result of EP-FEA shows that lowering the elastic modulus of weld metal to decrease strain in the vicinity of weld toe is more effective in higher region of nominal stress applied to fillet weld joints. This tendency corresponds to the experimental results of boxing weld joints of Invar.
(4) According to the EP-FEA, mechanical behaviors in the weld toe of joints with weld metal of low elastic modulus qualitatively corresponds to that of the model in which bonded material of weld metal and base metal (and HAZ) in series with a notch is loaded under constant displacement.

View full abstract

Fatigue cracks often initiate at the fillet weld toe of structual discontinuity parts in welded struc-tures. Large tensile residual stress usually acts in these parts in as-weld condition. As the tensile residual stress forces fatigue strength to decrease, we intended that tensile residual stress turns to compressive residual stress for the purpose of enphancement of fatigue strength.
Simplified method obtaining flame line heating condition for maximum reduce of residual stress is investigated from the previous study of flame line heating and welded residual stress.
As a result, it is confirmed that residual stress with the order of yield stress (σ_??_) in the as-weld condition at boxing weld toe changes to compressive residual stress with the order of -0.5σ_??_ by flame line heating for HT-50 steel.
Then fatigue tests of boxing welded joint using HT-50 steel were carried out both for as-welded joint and flame heated joint. Drastic improvement of fatigue strength is observed after flame line heating.

View full abstract

In this paper, the fatigue strength of welded joints of 6N01 aluminium alloy extrusions is studied. Less copper content (C_u_??_0.02%) alloys were chosen, taking corrosion resistance to sea water into consideration. Two series of specimens were prepared. One is non-load carrying transverse fillet welded cruciform joint (T-TYPE), and the other has two longitudinal stiffeners welded on both sides of the main plate (L-TYPE). For L-TYPE, three types of specimens, changed in details of end of longitudinal stiffeners from the conventional corner boxing, were arranged. For some specimens, weld toes were ground with a pencil grinder. Using the specimens, fatigue tests under axial tensile loading (R=0) were carried out.
In the fatigue test results, evident differences existed. The fatigue strength of ground T-TYPE specimen was highest, and that of as welded L-TYPE ones was lowest. The fatigue strength of as welded T-TYPE specimens was higher than that of ground L-TYPE ones. These SN data were consolidated and yielded to one SN curve, using the equivalent stress, taking the maximum stress into account. The equivalent stress S_??_ was expressed as follows; where S_max=maximum stress, R=stress ratio, m=constant, S=stress range, K=stress concentration factor, S_n=nominal stress range, and S_R=residual stress.
In the last part of the paper, some techniques for improving the fatigue strength of corner boxing welds are illustrated.

View full abstract