The hydrofoil interaction of hydrofoil catamarans (HC) is studied for HC200A and HC2000. It is revealed that the optimal angle of attack must be determined with careful consideration of interactions including the effect of free-surface deformation by the forward hydrofoil. A design for a 4000 displacement-ton type is presented and its hydrodynamical properties are studied. It is demonstrated that the resistance and motion properties are superior to HC2000 notwithstanding the larger number of hydrofoils.

View full abstract

In the present study, Navier-Stokes equation is numerically solved by use of a finite difference method to obtain the 3-dimensional flow field around a prolate spheroid at small incidence. The computed flow field shows the same characteristics to that of the observation by a flow visualization technique which was reported in the author's previous report. The existence of a bubble-type separation at the rear end and the nodal point in the surface shear-stress pattern are recognised in the computed results.
The 3-dimensional vorticity field is derived from the result of the computation. The vorticity structure is analysed by dissolving the vorticity vector into two components, i. e. the one is the component along the flow direction (locally streamwise vorticity) and the other component is in the locally transverse plane. The two components of the vorticity show different characteristics, i. e. the locally transverse vorticity vector shows a maximum value near the fore end and decreases in the downstream as the boundary layer grows. On the other hand, the streamwise vorticity grows in the after region and keeps its value in the wake region, although the magnitude of the streamwise vorticity is quite small compare to that of the transverse vorticity.

View full abstract

A finite-volume, conservative upwind scheme has been developed, based on the flux-difference splitting method. Pseudo-compressibility is introduced to the continuity equation. The cell-centered is adopted, i. e., nodes for flow variables were placed at the center of each grid cell. With this combination of the scheme and the node-cell layout, the global conservation property has been derived in a straightforward manner.
The scheme was applied to two types of flows. First the flow past a circular cylinder was computed using the O-grid at the Reynolds number R_e =40. The integrated momentum and mass fluxes at inner and outer boundaries agreed up to more than 9 significant figures after 1, 000 time steps. Thus the global conservation property was confirmed. The computed drag coefficient value agreed well with other computed values. The same flow was computed using the H-grid. The drag coefficient value thus obtained differed very little from the O-grid value.
The flow past a flat plate with a point of mapping singularity was computed at an attack angle of 30 degrees. It was confirmed that the global conservation property of the present scheme is not affected by the presence of mapping singularities.

View full abstract

Large Eddy Simulation (LES) with a Finite Difference Method is presented for a fully developed turbulent flow in a square duct. Furthermore the present method is applied to investigation into a microbubble injected flow.
The 3rd order up-wind finite difference is employed to dissipate numerical errors which arise from the convection term. Other partial differential operations in space are approximated with the 4th order central finite differences. A regular grid system is adopted and all velocity components and pressure are defined at the same grid points.
Turbulence-driven secondary motion is reasonably simulated without a special anisotropic model.
The microbubble and water mixture is treated as a homogeneous fluid with spatially varying properties. The local density and molecular viscosity of the mixture are reprensented as a function of the rate of bubble concentration in boundary layer. The turbulent structure depending upon the bubble injection in the boundary layer is studied.

View full abstract

This paper presents a new concept to reduce turbulent frictional drag by injecting high-viscosity fluid into boundary layer. When the turbulent region of boundary layer is filled with high-viscosity fluid and the viscosity of the viscous sublayer is kept low, the velocity profile in the boundary layer should change substantially. The Reynolds stress in turbulent region becomes less which requests more velocity gradient there to keep the momentum transfer same. It results in a reduction of velocity gradient at the viscous sublayer which gives the reduction of shear stress at the wall.
Such a boundary layer structure could be realized by injecting two different fluids from double slits on a wall. Sugar sirup was used as the high-viscosity fluid at the experiment. The sugar sirup and water were injected into turbulent boundary layer on a flat plate through up-stream and down-stream slits respectively. A steady injection was realized by pushing out the fluid in a tank using compressed air. The shear stress was directly measured by shear stress pick-ups mounted flush on the wall.
The experimental result showed a substantial reduction of shear stress by injecting sugar sirup and water from the double slits. The maximum reduction rate was more than 50%. A series of water/ water injection experiment was also made to know the effect of injection itself.

View full abstract

The present paper deals with the interaction problem between the hull and the propeller of a ship advancing at a constant speed on the still water surface. The Rankine source method has been employed in order to take into consideration the effect of the free surface on the hull and the propeller. The vortex lattice lifting surface method has been applied to the propeller. The interaction problem between the hull and the propeller has been solved numerically by an iterative procedure, starting from the Hess & Smith solution for the source distribution on the hull surface of the double model approximated ship as the basic flow. The effect of the interaction of the propeller has appeared on the closest hull surface as the pressure reduction and on the upper free surface as the increase of the wave just above the propeller. The calculated thrust deduction factor has showed a fairly good agreement with measured ones for 4 m and 6 m model ships.

View full abstract

This paper describes the first development of sophisticated pressure measurement technique for a full scale propeller and the success of the measurement. First of all, the special pressure pick-ups with Helmholtz chamber were designed. The full scale measurements were carried out on the training ship “SEIUN-MARU”. Six pressure pick-ups were equipped at each of four propeller blades. The pressure signals were transmitted from the pressure pick-ups on the propeller blades through slip-rings to FM receivers and analyzed using static pressure calibration data obtained in the dock before the voyage.
This ship was operated so carefully that the working conditions at each propeller revolution rate were kept constant, that is, K_T = 0.210. Then, the measured non-dimensionalized pressure distributions were similar at each condition, except cavitation regions. Except the lowest revolution rate, sheet cavitation or tip vortex cavitation was observed and the influence of cavitation of other blades on the pressure measurement was found out. The present measurement techniques have the accuracy of ±0.03 kg/cm². By the pressure coefficient, it amounts to ±0.3 at 70 RPM and ± 0.07 at 149 RPM, respectively.
The measured pressure distributions were compared with the theoretical ones obtained by the existing lifting surface theory. In this calculation, the estimated nominal wake distribution was employed, including the tangential wake based on the measurement in a towing tank. Excellent agreements with theory were found at most of the measurement points, especially the fore part of the blades. These results clearly demonstrate that the use of this estimated wake and the lifting surface theory with the concept of the equivalent two-dimensional profile is quite reasonable for a conventional propeller principally. Near the angular position of the top, some discrepancies between the measurements and theory were observed probably due to the deformation of the nominal wake and the leading edge separation. The present full scale measurements indicated that there still exist some problems on the lifting-surface theory and the use of nominal wake.
These measurements also confirmed that the measured pressure in the sheet cavitation region was nearly equivalent to the vapor pressure at each working condition.
The present study has provided a number of invaluable standard data to validate the numerical computational techniques on marine propellers at high Reynolds number.

View full abstract

In recent years, the techniques of numerical analysis of the propeller performance have been developed remarkably, by the advances of the studies on the trailing vortex geometry.
Although many numerical methods give us the results that agree pretty well with the measured data of the open-water characteristics of propeller at design advance speed, the agreements are sometimes poor at low advance speed, especially in case of the propellers with high pitch ratio and large blade area ratio.
This paper proposes a new mathematical model to present the procedure of the flow separation from the tips, and tries to show that the separation leads to increase the thrust, torque coefficients at low advance speed.
The results of our numerical study show that the influence of the separated tip vortex appears on the trailing vortex geometry, as well as on the loads and pressure distributions over the blade surface, especially at low advance speed.
The present method is advantageous, this author believes, enabling us to calculate the performance of propellers of various types accurately, over the wide range of advance speed.

View full abstract

A correlation of noise radiated from sheet or bubble cavitation occurring on propeller blade surface is well established. On the other hand, the correlation is not clear on noise radiated from vortex cavitation occurring at a distance from propeller blade. In the present paper, cavitation patterns and noise characteristics of the propeller accompanying only the vortex cavitation were investigated by model tests in a cavitation tunnel and full-scale measurements in order to know the correlation of noise radiated from vortex cavitation of propeller. Comparisons of full-scale cavitation pattern with model scale showed that vortex cavitation depends on not only cavitation number but also Reynolds number. Predicted noise level from model tests using scaling formulas of cavitation noise agreed well with full -scale measurements in the case where vortex cavitation patterns in model and full scale are similar. Cavitation number may vary at Reynolds number to the 0.15 power to keep similarity of pattern of vortex cavitation.

View full abstract

Characteristics of cavitating flow around a finite span hydrofoil are studied by a finite difference computation using Bubble Two-phase Flow model. The flow with Reynolds number of 103 is calculated at the angle of attack of 10 deg. on a NACA 0012 hydrofoil with elliptic planform of aspect ratio of 3.
The mechanisms of the cavitating flow, particularly the relationships between the vortex system around the hydrofoil and the cavitation, are investigated with the use of the high spatial resolution in the numerical calculation although the Reynolds number is low because of the computation time and memory capacity problem. The discussions on the calculated results give the following results :
(1) Since the planform of the hydrofoil is elliptic, the ratio of cavity length to chord length becomes almost constant along the span when the cavity is small. But as the cavity grows, the tip region begins to generate larger cavity. The reason for this is that the tip vortex expands the separating flow region on the suction side of the hydrofoil near the tip.
(2) The larger cavity weakens the tip vortex because of the reduction of the lift force and/or the diffusion of flow due to cavity. The diffusion due to cavity also increases the width of its wake.
(3) The above calculated results agrees qualitatively with the existing experimental results. It can be said that even the present calculation with low Reynolds number has clarified many of the mechanisms of cavitating flow around a finite span hydrofoil.

View full abstract

This paper describes the propeller noise reduction of the newly built supporting ship “Yokosuka” for the 6500 m deep submergence research vehicle. Propeller noise is one of the main underwater radiated noise sources and it was very important technical item to reduce the propeller noise below the required level drastically. So both improvement of the propeller shape and optimization of the CPP blade angle-revolution number combination were investigated widely from a viewpoint of minimizing cavitation occurrence at various operating conditions. Tip unloaded skew CPPs were finaly designed, making noise measurements of model propellers in the cavitation tunnel and theoretical calculations by the propeller lifting surface program. The full scale propeller noise was measured at her sea trial and compared with the predicted noise. It was found that noise levels of the full scale propller were very low and sufficiently satisfied the required noise level and our method was very effective in designing a low noise propeller.

View full abstract

Stopping characteristics of a ship due to the reverse rotation of its propellers in a shallow water area are investigated. Extensive captive model tests are carried out in order to identify the characteristics of the hydrodynamic forces acting on the ship in its stopping manoeuvres. Experiments on the stopping manoeuvres using a free running ship model are also conducted. The time history of the behaviours of the ship as well as the macroscopic stopping characteristics such as the distance & the time consumed for the stopping are examined through the comparison with the corresponding data obtained in deep water. Finally numerical simulations of the stopping manoeuvres are carried out while using the experimentally determined hydrodynamic forces to see how accurate the prediction can be achieved. The possible causes for the discrepancies (although it is not very large) between the numerical simulation results and the experimental results are discussed.

View full abstract

A panel method to calculate unsteady ship hydrodynamic forces is introduced. First, formulations were made for both steady and unsteady free-surface conditions under the assumption of small perturbation over the double body flow around a ship. A newly derived free-surface condition for unsteady motion makes a theoretical pair with Dawson's steady linearized free-surface condition. Next, a Rankine panel method was applied to solve the equations based on the present unsteady free-surface condition. For radiation condition of waves a numerical damping was introduced into the free-surface condition. Calculations were made of the unsteady hydrodynamic forces such as added mass, damping and wave exciting forces for a two-dimensional submerged cylinder and an ellipsoidal ship hull form. By comparing with other calculations and experiments it was shown that the present numerical method is useful for better understanding of the unsteady free-surface flow problems.

View full abstract

On the prediction of ship manoeuvring performance at the initial design stage, it is considered that a loading condition is one of the important parameters caused for the manoeuvring characteristics. For the prediction of ship manoeuvrability with high accuracy, it will be required to estimate the hydrodynamic forces acting on ship accurately in any loading conditions.
In this paper, the approximate formulae for estimating the hydrodynamic forces acting on ship in any loading conditions such as half loaded, ballast and trim by stern conditions are proposed. These approximate formulae were derived from the results of model test. The model ships used for obtaining the hydrodynamic forces are 13 ships consisting of general cargo, car carrier and RORO ships. And the model test was carried on 13 ships for fully loaded condition, on 11 ships for ballast condition and 5 ships for half loaded condition.
By comparing with the measured results of free running model test, the prediction results agree well with model test results. Therefore, this method will be useful for practical prediction of manoeuvrability for conventional ship at the initial design stage. However since those approximate formulae have been investigated on model ship, there still remain some problems to be solved such as a correlation, scale effect and so on, to predict the manoeuvring performance of full scale ship.

View full abstract

In this paper, the authors present a new method for calculating vertical motions of large-sized high-speed ships with hydrofoils. It is very important for this calculation to predict correctly the dynamic lift of the submerged hydrofoils which travel close to the free surface under the waves. Measurements were made of the dynamic lift of the hydrofoils moving at constant speed in regular waves. Using these experimental results, we propose a method for estimating the dynamic lift of the hydrofoils travelling beneath the waves. The ship motion is predicted by the strip method taking account of the effects of the nonlinear hydrodynamic forces and dynamic lift of the hydrofoils.
To verify the present method, the authors compare the computational results with the experimental ones of HC2000. HC2000 is a new type of hydrofoil catamaran having a displacement of 2000 tons and composed of twin V-hulls and four rectangular hydrofoils. The computational results in still water and head seas agree very well with the experimental ones of the 1/50 model of HC2000.

View full abstract

A 3-D source method is applied for computing wave loads on the coss structure of a Semi-submerged Catamaran, SSC (or Small Waterplane-Area Twin Hull, SWATH) running in regular waves. The 3-D Green function with forward speed, which has been prohibitively time consuming to apply to a SSC problem, is used with the aid of the steepest descent method.
Among wave loads, a side force, which squeezes and splits two demihulls of a SSC, is crucially important in the strength design phase because of the SSC's distinctive configuration. It is shown by both theoretical computations and model experiments that the side force drastically decreases when the SSC is running. This is explained by that hydrodynamic interactions between the two demihulls, which are due to the action of free waves from one demihull to the other, decrease in the existence of the forward speed because some of the free waves flow downstream without reaching the other demihull.
Fluctuating pressure distributions on SSC surface, which contribute to wave loads, are computed and discussed. Some of the computed results are successfully compared with experimental ones.

View full abstract

The required minimum output of main engine for large fat ship from the view point of safety navigation is evaluated.
The manoeuverable sea conditions are evaluated by estimation of ship speeds, drifting angles and check helms in various seas. In this estimation, influences of added resistance, steady lateral force and steady yaw moment in weves are accounted. The characteristics of main engine is also accounted. The evaluated manoeuverable sea conditions are in good agreement with ones by log data.
Assuming the several cases of output of main engine, the relation between the manoeuverable sea condition and the output of main engine is evaluated. Using this result, the relation between the unmanoeuverable probability and the required minimum output of main engine is estimated. The probability that the ship meets unmanoeuverable condition will be reduced by the increasing of main engine output, however, it shall be limitted up to mechanical availability of main engine. Thus the criteria for the required minimum output of main engine shall be set up.

View full abstract

Wave focusing has been attracting ocean engineers as one of the most promising techniques to control ocean waves. It creates a calm sea area and helps efficient utilization of wave energy. In the present work, a hydrodynamic singularity distribution which expresses a wave focusing lens is derived by the method of matched asymptotic expansion, assuming slenderness of the lens and high frequency of incident waves. The singularity distribution gives the following necessary conditions for scattered waves in each section of the lens : there is no reflection from the lens and the transmitted waves suffer a phase shift in passing the lens. The phase shift is given by the wavenumber and the distance between the section and the focus. From these conditions, we examine a sectional shape of the lens and determine the whole geometry.
It is shown by experiments and numerical computations using the two dimensional doublet distribution method that a submerged chevron shape plate, which is suitably folded, scatters a wave system which satisfies the above conditions at a certain wave frequency, but not in wide band of wave frequencies because of dispersion of water waves.
Then it is shown by experiments that a certain number of submerged circular cylinders, which are horizontally arranged at intervals just like a raft, transmits waves which have enough phase shift to focus waves but reflects almost no waves in wide band of wave frequencies.
Finally, we examine performances of three types of lens, namely, submerged flat plate, submerged chevron shape plate, and submerged circular cylinders, in both regular and irregular waves. It is shown by numerical computations that the wave focusing efficiency of the lens consisting of circular cylinders is about twice that of the flat plate type lens and that the drift force acting on the former is less than half of that on the latter in irregular waves.

View full abstract

The objective of present paper is to clarify the charactaristics of separated flow around a submerged horizontal cylinder with rectangular cross section in regular waves and steady currents.
The hydrodynamic forces on the submerged cylinder were calculated by the discrete vortex method. The numerical results were compared with those obtained by other numerical method so that the accuracy of the present method was confirmed.
The experiments in currents and in waves-currents were carried out and the effects of free surface and of wave-current interaction on the hydrodynamic forces and the strauhal number were discussed.
Moreover, the numerical results obtained by the present method were compared with the experimental results so that its applicability was clarified.
The results of the visualization experiments were compared with the flow pattern computed by the present method and the properties of flow field in the vicinity of the cylinder were discussed.

View full abstract

This paper presents a numerical method of analyzing a feedback control system of rolling motion of a ship in waves by a U-section type activated antirolling tank. The antirolling tank system is activated by impellers installed on the channel's centre of tank. The presented method consists of a simulation technique of liquid motion in the tank with impellers and of a design procedure of a feed back control system. The simulation technique is remade for the liquid motion in a U-section tank on the basis of the Marker-and-Cell method and its computer program (sloshing cord). The design procedure is based upon a theory of control system (the Root locus method).
A series of numerical computation are executed for three activated antirolling tanks by using proposed method in any conditions, and several problems for analyzing a feedback control system are discussed.

View full abstract

In the previous paper, since a new mode of capsizing accompanied with the period doubling bifurcation, which was regarded as a precursor of the chaos and crisis in the dynamical system theory, was observed in the model tests, the authors pointed out that an investigation into the relation between capsize and chaos should be carried out.
In the present paper, a simple capsize equation with nonlinear cubic term in the restoring force is solved by the numerical time simulation. A safe basin, which is defined as a non-capsizing region in the phase plane of rolling angle and rolling velocity, is obtained for the fixed control parameters such as forcing frequency, forcing amplitude and damping coefficient. In order to draw a complete safe basin in some square area of the phase plane, 301 × 301= 90, 601 grid points are used as initial conditions for the time simulation. By the results of simulation until the prescribed number of forcing cycles, e. g. 50 cycles, each grid point is distinguished in accordance with capsize or non-capsize. The boundary of safe basin obtained by such a method is metamorphosed complicatedly and in a fractal way as the control parameter is varied. This means the sensitivity and complexity of dependence of capsize on initial conditions. Melnikov analysis is also carried out in order to find the critical condition of homoclinic tangency, which means the beginning of the fractal metamorphoses in the basin boundary.
The area of the safe basin is also decreased in a fractal way like a devil' staircase as the forcing amplitude is increased. This signifies that the basin boundary has a feature of the Cantor set.
In the control parameter plane of the forcing amplitude and forcing frequency, the capsizing or non-capsizing region can be separated for some fixed initial condition as well as fixed damping, by using the similar numerical method. In this case, more than 1, 350, 000 grid points are used. The boundary between both regions is again fractal and complicated. Analytical expressions for the boundary are farely good approximations to the numerical results, which suggests that the simple analytical expressions may be used as criteria for capsizing.
An investigation should be continued to cover the wide range of control parameters and extended to asymmetrical capsize equation of a biased ship and a Mathieu type capsize equation.

View full abstract

A numerical method for the two dimensional full nonlinear Radiation-Diffraction problem is developed to simulate the motions of arbitrary shaped 2-D floating body in waves. The mixed Eulerian-Lagrangian method is utilized for the free surface. To calculate the accurate hydrodynamic pressure on the body surface, that is to calculate the accurate acceleration of body, both the velocity field and the acceleration field combined with the equations of body motions are solved. This method enables us to keep the dynamic equilibrium of forces between fluid and floating body on every time step. The Boundary Element Method (BEM) is used to solve the field equations.
For illustration, two computational examples are presented in comparison with the experimental data : (i) a transient motion of midship section body in an incident wave ; (ii) a growth of parametric oscillation of roll motion of flare shaped body in a regular wave. These examples show that the computational results are in excellent agreement with measured body motions.

View full abstract

A new theory is developed for predicting the side-wall effects upon the radiation and diffraction forces acting on a ship in waves, by applying the unified slender-ship theory devised by Newman to the case where vertical and parallel side walls are present. In the present theory, only a solution of the radiation problem suffices for the calculation of the wave-exciting force and thus of the ship motions in waves. Needless to say, both zero- and forward-speed problems can be handled with the same calculation scheme.
Numerical results are compared for the zero-speed case with independent results by a 3-D integral-equation method, and compared for the forward-speed case with corresponding experiments which have been obtained by the authors with a floating spheroid of beam-length ratio 1/5 and a Lewis-form ship model of beam-length ratio 1/6, both running at a Froude number 0.1. In all cases excellent agreement exists, and thus it can be concluded that the usefulness of the theory has been confirmed.
As an application of the present work, a number of computations based on the present theory have been performed over a practical range of Froude number, oscillation frequency, and tank width. Then diagrams are newly prepared for predicting not only whether the tank-wall effect is expected or not, but also the region in which the effect is less than 10 percent of the open-sea value.

View full abstract

A simplified method described in the previous report, has been extended to calculate steady and unsteady lift of arbitrary three-dimensional body and wing configurations, in which the velocity potential is expressed by the doublet distribution over body, wing and the wake. The velocity potential on a thick body is simply obtained by subtracting the interior velocity potential defined inside the body from the solution of the doublet distribution for the translating oscillation. This method is applied to study hydrodynamic interactions between body and wing configurations in steady and unsteady lifting problems.
Experiments for a body of revolution with and without a pair of low aspect ratio wing are carried out for steady lift with angle of attack and unsteady lift in heaving oscillation. Those results are compared graphically with the present theory, together with approximated calculations by equivalent wings.

View full abstract

The at-sea experiment of the moored floating structure “POSEIDON” has been carried out at the Japan sea since 1986 for four years.
This paper deals with the characteristics of directional wave spectra measured at the test field. Directional wave measurement in the field is important because of the increasing intrest in relation to maritime design procedures, and because of the growing number of model basins with multi-directional wave capabilities.
In this experiment, three ultra-sonic type wave probes were installed as a line array at the sea bottom of test area, and the measured three time series of water surface elevations were used to analyze directional wave spectra by Maximum Likelihood Method (MLM).
The directional wave spectra can be expressed as S (f) ·G (f, θ), where S (f) is frequency spectrum and G (f, θ) is directional function. Using measured data, the characteristics of S (f) and G (f, θ) were investigated intensively. The main results are as follows ;
1) The shape of the high frequency side of frequency spectrum S (f) is not proportional to f^-5, contrary to expectations, but it is proportional to f^-4. This was confirmed by applying Toba's 3/2 power law to measured data. The peak of frequency spectra become sharp when peak frequency shift to low frequency range.
2) Estimated Directional function G (f, θ) can be approximated by the so-called Mitsuyasu Type function. The spreading parameter S becomes maximum near the peak frequency of S (f) , and it spreads arround 1025.
The Wave direction estimatied from directional function, doesn't always accord with the wind direction by the refraction phenomena due to the shallow effect.

View full abstract

Estimation of directional frequency transfer functions of floating offshore structures is important to estimate their behaviours in realistic short crested irregular waves (such waves are called as directional spectrum waves)
Of course, there are some theoretical calculation methods for such directional response functions, but it seems that there are not so many experimental results confirming such theoretical ones. For that purpose, square basins are needed.
On the other hand, recently the authors developed a wave maker which can generate directional spectrum waves in a long towing tank, and many experiments of offshore structures are conducted in such waves.
So we tried to make inverse estimation of directional transfer functions of a floating offshore structure using both one dimensional response spectrum measumred in such waves, and measured two dimensional directional wave spectrum.
Estimated results by present method on heave, pitch, and roll response functions are reported here, but this method will be applicable to other situations like motions of floating structures in actual seas by measuring its motions and incoming waves.

View full abstract

In the previous studies, basic methods to estimate motion or attitude of the vehicle (camera), i. e. ego motion, from video image data obtained continuously by a VTR are proposed, and their performances and applicability to the actual images are investigated. By applying an iterative estimating algorithm based on the Extended Kalman Filter to continuous pictures of the ground surface, it is assured that the motion can be estimated with enough stability.
In this study, continuous underwater images are obtained by a ROV, and its motion is estimated by the same algorithm to investigate applicability of the methods for actual use in the water. The accuracy of the estimation is evaluated for various conditions, and that of 0.25 edgree (maximum error) in yaw angle is assured for ordinary images with appropriate landmarks. The necessity of calibration for the edge distortion is pointed out in case very few landmarks exist in the pictures.

View full abstract

A self-organizing neural-net-controller system (SONCS) is proposed as an adaptive system to handle unknown dynamics of a controlled object and unpredictable environmental conditions.
SONCS consists of a controller, forward models, and evaluation and adaptation mechanism. The forward models are neural networks which express the forward dynamics of the controlled object. The controller is a neural network which provides control inputs to the object and the forward models. The evaluation and adaptation mechanism is a modification tool which calculates the difference between actual and desired motion of the object and modifies the system to get a specific function. The basic idea of this system is to adapt the controller network with this evaluation and adaptation mechanism observing the object's motion with the forward models. SONCS is also equipped with an initiation tool called, premature controller, which can be easily constructed with just qualitative information on the object's dynamics.
Efficiency of the developed SONCS is demonstrated in the application to 'PTEROA60', a test-bed for underwater vehicles. A suitable neural-net-controller, which can let the vehicle swim stably in desired state, is organized in this system through free-swimming in a tank.
It is concluded that an adaptive system, which can generate automatically a desired controller to keep the controlled object in target state, can be constructed using neural network technology. The proposed SONCS has a great possibility to realize highly sophisticated control system which can deal with complicated control problems.

View full abstract

This paper treats the automatic position and the attitude control for the underwater vehicle staying still or moving at dead slow speed.
The motion control of the underwater vehicle cruising at high speed is rather easy and now in practical use successfully, because the hydrodynamic characteristics and the dynamics of the vehicle can be assumed to be linear, which allows application of linear control system using control surfaces.
But the position and the attitude keeping control of the underwater vehicle staying still or moving at dead slow speed becomes non-linear system, because the speed of all degrees of freedom becomes comparable and the acceleration or the deceleration motion must be applied using thrusters.
This paper describes the linear approximation of this non-linear system, and presents a linear design method.

View full abstract

ROVs with computer aided trajectory control and acoustic feedback sensors will be needed for lifting and transportation of objects related to the underwater activities such as inspection, installation and repair of offshore structures. Especially in deep seas, environments are hazardous for human and ROVs become very useful tools. These ROVs are also demanded in the activities related to huge offshore platforms for space utilization. In the repair and inspection of such structures, the amount of works are expected to be so large that without these ROVs the works can not be accomplished.
These ROVs are thought to have relatively complicated configurations due to large thruster, and the dynamics will be highly nonlinear and have uncertainties in their dynamics compared to the streamlined vehicles. The nonlinearlities depend on velocity, and hydrodynamic coefficients are poorly known. When such system are controlled, control method must be able to handle the nonlinear dynamics and uncertainties in parameters which depend on the motion of ROV. Gain-scheduled technique which uses a set of linear controllers for different combinations of speeds and directions of motions, can handle these problems. Although theoretically possible, the number of linear controllers become so large that this method is not practical. Another technique which is able to handle the problems and adopted in this paper is sliding mode technique. If the bounds of uncertainties in parameters which determine the dynamics of ROV are known, the technique can incorporate them in the formulation and allows uncertainties in the system. Dynamics change in ROV caused by lifting objects can also be treated as a uncertainty in parameter and control is expected to be robust.
The trajectory control of ROVs is examined experimentally by model basin test using a small ROV with four thrusters. Surge and Yaw are controlled by two thrusters in horizontal plane. Heave and Sway are controlled by thrusters arranged diagonally in vertical plane. Ultrasonic positioning system was developed for the basin test. The ROV used in the test was relatively small, and high accuracy of one millimeter order was required for the positioning system and realized by using ultrasound of high frequency.

View full abstract

The author has developed two types of cable laying simulation programs based on three dimensional dynamic analysis method in order to analyze dynamic behavior of cables, repeaters and branching unit during laying and recovery.
One is a computer simulation program for laying and recovery of submarine cables with repeaters. This program has following two new functions is addition to conventional cable laying simulation.
(1) Slack control laying function. The cable pay-out speed is automatically calculated by internal feed back loop so as to make bottom cable slacks be constant. Using this function, the optimum cable pay -out speed history can be obtained by giving the bottom profile and cable ship speed profile.
(2) Tension control laying function. The cable pay-out speed is automatically calculated by internal feed back loop so as to make shipboard tension be constant. The speed error of cable ship measured by 'taut wire' is discussed using this program. This error is exactly analyzed using this tension control laying simulation.
The other is a computer program for laying and recovery submarine cables with a cable branching unit. The cable branching unit (BU) is newly installed to the optical submarine cable systems. This newly developed program can analyze dynamic behavior of BU and cables around it, and help to study how to lay BU with sufficient cable slacks and recover it safety.
The sea trial was carried out in order to measure hydrodynamic constant of the submarine cable and verify this computer simulation method at the depth of 6000m by attaching sensors to the submarine cable at some intervals. According to the sea trial, the normal drag coefficient Cd was measured to 2.5, andd the cable configuration measured by this experiment was in good agreement with the result of the computer simulation.

View full abstract

The adsorbent bed system proposed in the 1st report consists of large number of units which are the thin disk cartridges enclosed in nets, and the each unit is arranged below the sea surface in horizontal layers by suspending ropes hung from a buoy.
In this paper, the possibility of the following oprating system is discussed. The units filled with fiber ball type-adsorbent are controlled in inclined position aganist sea current by the winches installed on a buoy so that sea current may permeate the units. When the units are collected on the buoy and also rearranged underwater, the units are controlled in horizontal position so as to maintain the smooth operations of system by reducing the horizontal displacement of the units through decreasing the drag forces of units due to sea current.
The uranium adsorption test using the model unit reduced to a scale of 1/10 was carried out in the test channel of uniform flow of sea water through 21 days. On the other hand, the flow distribution in the unit was calculated and the uranium adsorption of the adsorbent enclosed in the uuit was simulated numerically. A good agreement can be seen between the measured and simulated uranium adsorptions.

View full abstract

A new method to identify the characteristic matrices from the response of the vibration system (transfer function) was proposed in the 1st report, and the accuracy of the proposed method was confirmed by the calculation using the transfer function without any experimental error.
In this report, the characteristic matrices of a cantilever beam and a free-free beam are identified from the measured transfer function having some experimental error. When the transfer function containing sufficient information about the vibration system is used for identification, the characteristic matrices are accurately identified from the measured results containing some error, and the accuracy of the proposed method is confirmed again.
Next the characteristic matrices of whole structure is derived by combining the experimentally or theoretically identified characteristic matrices of two sub-structures. The transfer function calculated using combined characteristic matrices of whole structure is a little different from the measured results. Because this difference is caused by the error in the identified characteristic matrices of the substructures, for the practical application of this estimating procedure the more accurate approach based on the proposed identification method must be developed.

View full abstract

Dynamic buckling behaviour of a T-section beam subjected to impact loads is consecutively investigated. In the previous reports it is found that even in the case of higher impact loads the lateral buckling did not always occure. In this report conditions of occurrence of the buckling is investigated by series tests. Results of experiments show that the momentum of a impact weight is a main factor for the occurrence of the buckling. The occurrence of the buckling depends on the momentum. This result is discussed with a simple analysis.

View full abstract

This paper discuss the behavior of ship structural panels subjected to acoustic excitation. In order to provide a clear understanding of the phenomenon of course of transmission of sound by a flexible panel, two dimensional sound-elastic interaction vibration problem has been examined. We have assumed two vibration models on a panel in an infinity large rigid baffle. The one is a Timoshenko model, the other is a two dimensional elastic model. We have introduced the exact solution of Timoshenko model and have formulated the coupled vibration of sound-elastic interaction model using the boundary element method. Comparing the sound transmission loss between the two models, we have shown that the principal element of the sound transmission is caused by out of plane vibration of panel and the transmission by in plane vibration can be neglected.

View full abstract

This paper concerns with the collapse analysis of reinforced concrete shells using the Rigid BodiesSpring Model (RBSM). Reinforced concrete shells are expected to be widely applicable in offshore engineering structures, but it is thought fairly difficult to estimate the ultimate strength of the structures due to their complicated material behaviours in crack formations, plastic deformations and crushing of concrete. The RBSM method, which deals with yield lines and crack lines explicitly, is suitable to the collapse analysis of reinforced concrete shells. The flat rigid plate element developed by Toi and Kawai for the general thin-walled shells is extended to the collapse analysis of reinforced concrete shells. Numerical studies on simply supported square plates, cylindrical tanks and cylindrical shell roofs are conducted in order to examine the validity of the algorithm of the present method.

View full abstract

The present study is concerned with the physical explanations of the linear and the cubic finite elements for beams and axisymmetric shells through the comparisons of their strain energy approximations with those of the Rigid Bodies-Spring Models which are the discrete elements suitable to the plastic collapse analysis using the concepts of plastic hinges and hinge lines. The established conditions for the equivalence between these two modelings, which are given as the relations between the locations of the numerical integration points and those of the occurrece of plastic hinges, can be conveniently used in the economical plastic collapse analysis of framed structures and axisymmetric shells where the locations of plastic hinge formations are controlled by the movement of numerical integration points. Some numerical results are shown in order to prove numerically the obtained relations and to verify the validity of the proposed shifting technique of numerical integration points, which is identified as 'the shifted integration technique' in the present paper.

View full abstract

Three-dimensional nonlinear analysis method of underwater line structure is presented and its validity is discussed by direct comparison with experiments. This analysis method is formulated by finite element method and dynamic behavior of the structures are solved in time domain. The motion equation is integrated directly and the method can handle all nonlinearities involved in the dynamic motion of the line structures such as hydrodynamic nonlinearity, geometrical nonlinearity, nonlinear boundary condition and so on. This method is a combination of Newmark's time integration scheme, one of the most reliable scheme, and Newton's iteration scheme for solution of nonlinear equations.
Newmark's time integration scheme is an implicit scheme and its relatively complicated form produces a system of nonlinear algebraic equations which must be solved at each time step by iteration. Newton's iteration scheme is effective in solving the nonlinear equations and the analysis method is expected to be more robust against divergence. But direct application of the method requires large matrix of derivatives and time consuming calculation causes inefficiency. In the present method, the matrix of derivatives is decomposed into small band matrices and the nonlinear equations can be solved minimizing the inefficiency in the calculation. Thus the analysis method become robust and efficient by combining the Newmark's and Newton's schemes.
So far a number of methods have been published related to the dynamic analysis of underwater line structures. Because of the nonlinearities handled in the analysis methods, these methods should have been tested by comparing the solution with experiments, but the validity of the methods were not tested except some cases. Primary reason is that general purpose instruments to measure the dynamic behaviors of underwater line structures in towing tank model tests were not available or limited. In this paper, development of a general purpose ultrasonic sensor for towing tank tests is described and the validity of the analysis method is discussed by direct comparison with expemimental results obtained from towing tank test. In the ultrasonic instrumentation system, high accuracy is required due to the small size of models and high frequency is employed, and coordinates of maximum 16 points can be determined at accuracy of 1 mm.

View full abstract

Recently detailed three-dimesional finite element analyses have commonly been performed for the structural design of the middle part, or sometimes of the whole part of a ship. The computer aided design of this kind is effective for the rational determination of scantlings of established ship structural systems. It should be contrasted for newly developed ships that difficulties may arise because of the various constraints accompanying to the present CAE-system as to examine the feasibility of the wide variations of possible structural systems at the initial stage of the structural design.
It has been a controversial issue whether the rapid development of computer hardwares and structural analysis softwares will soon accomplish the level which will enable the structural synthesis including the selection of an optimum structural system. Since we have not yet had the definite answer to the question whether a machine could think design strategies, it seems appropriate to pursuit not only complex but accurate CAE-system, but also simple but approximate methods in order to assist skilled structural designers at the very begining of ship design.
In the present paper a simplified analysis is proposed for the proper selection of structural systems of a car carrier, whose transverse structural system has wide variations such as partial bulkhead system and strong web frame system depending on the relevant design philosophies. Introducing a simple beam model, which can be solved by a symbolic manipulation procedure, one can observe that the racking deformation of transverse section is mainly influenced by the non-dimensional shearing rigidity defined in the present analysis. Comparing the partial bulkhead and strong web frame systems, the shearing rigidities at the bulkhead sections having the latter system is considerably small, and correspondingly its racking deformation obtained by the simplified analysis is several times larger than that of the former system.
In order to predict the racking deformation of actual ships having various transverse rigidities, an effective method is proposed, where the simple beam analysis using symbolic calculations is combined with the numerical data obtained by precise three-dimensional structural analyses of ships having similar principal dimensions. The validity of the proposed method is confirmed by comparing the predicted values with the precise numerical results.

View full abstract

An attempt is made to investigate into a practical assessment procedure to evaluate the safety and reliability of ships on the basis of the ultimate longitudinal strength of ship's hull girder.
The external bending moment acting on the hull girder is assumed by using an existing rule of classification society. The ultimate longitudinal strength is analyzed by idealized structural unit method, in which a structural model for plate element subjected to biaxial load is formulated by idealizing the actual nonlinear behaviour of the element taking account of the influence of initial imperfections as well as the interaction effect between the local and overall buckling of the structure. The central safety factor of the hull girder is evaluated by dividing the external bending moment into the ultimate bending moment. The reliability assessment is made by using Cornel's second-moment method.
The proposed safety and reliability assessment procedure is then applied as an example to evaluate the central safety factor and the safety index of a double skin tanker. The influence of initial deflection, welding residual stress and the yield stress on the safety and reliability of the hull girder under sagging and hogging moment is investigated. The stiffening effect of the longitudinal girder and side stringer is also discussed.

View full abstract

When an offshore structure is damaged through a collision with a ship or an iceberg, certain influence from the fluid force acting on the colliding body is expected. However, reports on this subject are limited in number and the provided information is rather fragmentary. The characteristic of the phenomenon changes with the relative speed of collision. In other words, it depends on the relative duration time of collision. In general, the duration time is determined not only by the mechanical property of the collided structure, such as its stiffness and strength, but also the shape and the mass of the colliding body.
To obtain general view of the phenomena for both when the behaviors of the structure are elastic and elastic-plastic, the authors studied the effects of the fluid force under collision through experiments and numerical analyses. It is shown that the degree of the effect of fluid force can be classified by parameters which are related to the relative duration time of the collision. Further, the degree of the effect of the fluid force is shown to be predicted using the proposed parameters.

View full abstract

In this paper a procedure for the optimization of ship structures is developed based on plastic design of ship structures consisting of the transverse and longitudinal members using the Sequential Unconstrained Minimization Technique (SUMT). Using the developed method, the optimization of a segregated ballast tanker is carried out considering the weight of one hold length as the objective function. Midship section scantlings are chosen as the design variables. Constraints are considered with respect to plastic collapse of hull girder under longitudinal bending, of structural members subjected to axial force, shear force and bending moment, and of platings subjected to lateral pressure. Constraints are also set in buckling strength of platings under inplane load and in design variables such as minimum plate thickness. The variations of maximum bending moments and shear forces due to changes in design variables are calculated based on Taylor series expansion method so as to reduce the computation time of iterative calculation for function minimization in the SUMT procedure. The tanker structure got optimized and the results show that the proposed method based on plastic design is an effective tool in the optimization of ship structures. A comparison of the proposed method with the optimization based on elastic design is also made and the results are presented.

View full abstract

General instability is one of the most important subjects of cylindrical shells reinforced by ring stiffeners under uniform pressure. Recently, based on the application of nonlinear analysis using the finite element method, and more reliable collapse tests using models manufactured with high fabrication technique and of uniform material characteristics, the study to review the design procedure, which were based on the experimental works of collapse tests using a number of scale models, has been carried out.
Since pressure hulls of deep diving submersibles are usually welded structures, the welding distortion is inevitable, which might have a large influence on collapse strength. The authors have presented the elasto-plastic analysis using three dimensional finite element in the first report to estimate collapse pressure reduction of general instability by initial imperfection.
To investigate the infuence of initial imperfection of complex mode and that on the overall structure, however, more effective mothod is required instead of the three dimensional analysis.
In the present paper, the authors propose the elasto-plastic analysis using the quasiaxisymmetric finite element, which represents the deformation of stiffened shells by Fourier series. The analytical results compared with those obtained from the current full three-dimensional analysis show a good agreement. Futhermore, the authors investigate the influence of initial imperfection on the collapse strength of scale models, which are machined or welded, by comparing the results of the theoretical analysis and those of experimental test.

View full abstract

The buckling evaluation method of the cylindrical shell with stiffener rings and a branch nozzle subjected to external pressure is investigated by the experiments on the model cylinder and nonlinear finite element method. Taking into account the effects of initial geometrical imperfections and prebuckling biaxial stress states, the buckling pressures predicted by the simplified finite element method are fairly in accordance with experimental buckling pressure.

View full abstract

To assess the load carrying capacity of a ship's hull under extreme longitudinal bending, it is very important to know the behaviour of individual structural members. In this connection, the behaviour not only under monotonously increasing load but also under cyclic load as well as during unloading and re-loading after buckling/plastic collapse has to be investigated into.
From this point of view, a series of elasto-plastic large deflection analysis is performed by the Finite Element Method on plates subjected to cyclic inplane load. The main conclusions obtained are :
(1) The inplane rigidity during unloading after buckling/plastic collapse converges upon a certain value with the increase in applied strain.
(2) The residual deflection, when unloaded after the compressive ultimate strength has been attained, increases with the increase in applied compressive strain, whereas the recovered deflection during unloading tends to become a constant.
(3) When extreme cyclic load is applied under constant strain amplitude, the stress-strain and stress-deflection relationships tend to converge to the respective certain loops after several cycles, but not completely converge.

View full abstract

An object-oriented expert system, aiding in the structural design of semi-submersibles, particularly those constructed with ring-stiffened shells subjected to external pressure, is studied in this paper. A ring-stiffened shell must have enough structural strength to avoid collapse. On the other hand, there are some requirements such as minimal weight and minimal construction cost. Since the two abovementioned requirements are mutually opposing, an optimal design technique must be considered. In this study, a sensitivity analysis is implemented inside of the expert system to achieve an optimal design.
The developed expert system is basically written in Uti-LISP and implemented on an EWS-A ring-stiffened shell is designed by this system using the same requirements that were used in the design of an actual submersible. Total constructon cost and weight of the submersible designed by this expert system and that of the real submersible are compared here.

View full abstract

10 kinds of metallic materials including newly developed copper alloys were investigated at first by the immersion test in natural sea water. After that, 4 kinds of anti-fouling metallic materials were investigated by the actual ship test conducted in the underwater ship hull of a car bulk carrier running in the Pacific Ocean for 16 months in order to make clear the corrosion resistance, anti-fouling properties and surface roughness of the materials.
The following results were mainly obtained by these tests. (1) The general corrosion was found on the almost all specimens attached to the underwater ship hull of car bulk carrier. There was no erosion found on any specimen of tested materials, though the speed of the carrier (7.8 m/sec.) exceeded the critical flow velocities (1-3 m/sec.) for copper and copper alloys in the case of heat exchangers using sea water.
(2) The anti-fouling properties of copper and copper alloys were able to be maintained with the dissolution of copper, that is the corrosion of copper and copper alloys.
(3) There was the slime attached to only a specimen of 90/10 Cu-Ni alloy mounted on the underwater ship hull. It is considered in the case of 90/10 Cu-Ni alloy that there was a possibility that the necessary antifouling properties were not obtained for the underwater ship hull.
(4) The surface roughness of all specimens attached to the underwater ship hull was in the range from 40μm to 70μm, which was smaller than that of ordinary anti-fouling paints.

View full abstract

11 kinds of anti-fouling metallic materials including 9 kinds of newly developed Cu-10%Mn-Al-Fe alloys were investigated at first by the immersion test in sea water for 7.5 months in order to make clear the relationship of corrosion rates and anti-fouling properties with the chemical composition of alloys and the immersion period.
After that, 8 kinds of anti-fouling metallic materials including 6 kinds of newly developed Cu-10% Mn-Al-Fe-Ni alloys were investigated by the actual ship test conducted in the small ferryboat running in the Nagasaki Port for 6.5 months in order to know the corrosion rates and anti-fouling properties of these materials exposed to the actual underwater ship hull.
The following results were mainly obtained by these tests.
(1) The corrosion rates of copper, 90/10 Cu-Ni alloy and Cu-10%Mn-Al alloys decreased with the exposure period, and it became difficult to keep the corrosion rates of these materials more than the critical corrosion rate necessary to maintain the anti-fouling ability for the under water ship hulls.
(2) The impingement attack was found at the edges of copper sheets adhered to the underwater ship hull of small ferryboat.
(3) The attachment of marine bio-fouling to underwater ship hull was much easier in the small ferryboat than in the large car bulk carrier.
It is considered that the anti-fouling properties and the resistance against the impingement attack of copper and the copper alloys were not enough for those of anti-fouling metallic materials covered on the underwater ship hulls of small boats. Moreover, it is considered in the case of large ships that it may be adequate to conduct the detailed investigations about the behaviors of anti-fouling metallic materials exposed to the various kinds of environments after the underwater ship hulls are covered.

View full abstract

The structural expressions for the distributions of the probability of crack detection (POD) and the probability of false crack indication (POF) in the visual inspection method are presented, where the scatters in the distributions are caused by many factors related to the applied inspection method and the defect conditions in the structure. In order to estimate the POD and the POF from the results of the field inspections carried out for a structure by an inspection team under a specified condition, multiple visual inspection method combined with the successive precise inspections which are carried out for limited locations is developed. The crack classification method is introduced to improve the accuracy of the capability evaluation. The applicability of the method is numerically examined for a structural model which has crack-form defects. The advantage of the method is that the mean values of POD and POF as well as their distribution properties can be estimated without using any information other than the results of field inspections.

View full abstract

Characterization of interlaminar fracture toughness will play an important role in the establishment of a damage-threshold/fail-safe approach to composite materials and structures. Double cantilever beam (DCB) and end notched flexure (ENF) tests are reviewed and revised methods are proposed for accurate and easy estimation of mode I and II interlaminar fracture toughness. Thickness normalized crack length versus cube root of compliance plots are used for data reduction of DCB test. CSD measurement technique is introduced to ENF test and stabilized ENF test is proposed. Mode II interlaminar fracture toughness can be estimated directly from load versus CSD diagram without measuring crack length. Applicability of proposed protocols are well confirmed by experiments of four kinds of carbon fiber composite systems, T300/2500, T800H/3631, APC-2 and Celion6000/PMR-15.

View full abstract