A grid generation method called “geometrical method” was used to generate three-dimensional grid around a Series 60 (Cb =O.7) ship hull.
The surface grid of the ship hull was generated by representing the hull surface using spline curves which connect the given offset points. The outer boundary was defined as a cylinder of a given radius. Intermediate points of the initial grid were given on a straight line connecting inner and outer boundary points.
The initial grid was modified iteratively to satisfy four requirement criteria, i. e., orthogonality, smoothing, clustering, and minimum spacing. They were quantified based on geometrical configurations of the grid. In the procedure, the grid point locations were changed gradually to satisfy those requirements, or balance in case they conflicted. Finally, the grid was reclustered in a single sweep to be applicable to high Reynolds-number flow computation. Therefore, it is possible to generate coarse grid initially with small number of grid points and then increase the number of points in a simple manner afterwards, thus minimizing CPU time and man-hour for generating the grid.

View full abstract

The finite difference methods for the 2-dimensional Navier-Stokes and the Euler equations with the nonlinear free surface conditions are developed to study the basic characteristics of the numerical scheme. The numerical experiments are made for the free surface flow around the hydrofoil of the NACA 0012 wing section. The effect of the difference of the numerical implementation of the boundary conditions for the Euler solver on the final solutions are discussed. The Euler solution with the proper boundary condition is compared with the Navier-Stokes solution together with the experiment or other inviscid calculations.

View full abstract

The theoretical calculation has been carried out on variation of inflow velocity induced by the rudder and on fluctuation of thrust and torque of a blade of a propeller due to the effect of the rudder by the method employing the vortex lattice model to the propeller and the rudder. The propeller is five-bladed and with no rake angle, and the rudder is a hanging type with the streamline section.
The fluctuation of inflow velocity on the propeller blade induced by the rudder has been about 6% of the advance speed when the angle of the rudder is zero and has increased up to 12% of the advance speed when the angle of the rudder is ten degrees. The mean value of the induced velocity by the rudder on the propeller blade has always been 3% forward with any angle of the rudder. This fact means that the effect of the thickness of the rudder is constant and independent on the rudder angle. The fluctuation of thrust of a blade of a propeller has been 11% of the thrust without the rudder when the rudder angle is zero and 22% of it when the rudder angle is ten degrees. The mean value of thrust due to the effect of the rudder has also been 3% increase. Torque has fluctuated 9% with zero angle of the rudder and 17% with 10 degrees of the rudder angle, and has shown 2% increase as the mean value.
The amounts of such fluctuations are not small, so that they should be taken into consideration in analysis of unsteady characteristics of a propeller.
The efficiency of the propeller itself has increased 2-3% due to the effect of the rudder. But the drag of the rudder has reduced the total propulsive efficiency of propeller and rudder. The loss is 2-3% with zero of the rudder angle but becomes 19% at J=0.55 with 10 degrees of the rudder angle.

View full abstract

This paper deals with pressure distribution which becomes effective when the propeller works in a severe condition resulting in the generation of harmful cloud cavitation. In such a case, it is considered that a way to stabilize sheet cavity and suppress cloud cavitation is more promising than the usual way to reduce the amount of cavity. Since pressure fluctuations induced by a cavitating propeller are mainly due to time variation of cavity volume, such an approach has a probability of not only suppressing erosion but also reducing pressure fluctuations.
Using a propeller design method developed in previous reports, two propellers were designed and tested at a condition where the original MAU type propeller generated cloud cavitation and erosion. One propeller had flat pressure distribution which was proved effective at a condition of no cloud cavitation in the previous reports. The other had triangular pressure distribution with a negative peak at the leading edge and a gentle pressure recovery towards the trailing edge to stabilize sheet cavitation and reduce its volume variation. The experimental results showed that the triangular pressure distribution propeller succeeded in suppressing the cloud cavitation completely while the flat one failed. However, neither cavity volume variation nor pressure fluctuations were lessened by the triangular pressure distribution although almost the same levels were kept in both the pressure fluctuations and noise.
Thus two-dimensional and quasi-steady study was performed to obtain a prospect for the pressure distribution which stabilizes sheet cavitation and also reduces its volume variation due to the change of the angle of attack. Two foil sections with a negative pressure peak at the leading edge and a region of constant pressure which is lower than the vapor pressure were designed with theoretical predictions by boundary layer and nonlinear cavity flow calculations. Comparative experiments to an existing MAU type section result showed that the new foil sections much lessen the cavity volume variation due to the angle of attack. They have a possibility of reducing the fluctuating pressure amplitude to less than half of that of the MAU section in addition to suppressing cloud cavitation.

View full abstract

This paper deals with the prediction of propeller cavitation noise both from model experiments in a cavitation tunnel and from a theoretical calculation along with its comparison with full scale measurements for two ships, a cargo liner and a training ship. It is proposed to estimate separately blade rate tones and broadband noise both in model experiment and theoretical calculation considering the different noise-generation mechanisms. In model experiment, the acoustical-mechanical reciprocity technique is applied for the estimate of blade rate tones and a theoretical estimate formula supports the experimental prediction of low-frequency broadband noise. Full scale measurements of far-field noise are carefully planned to get reliable data so as for example to get rid of the effects of sound reflection at the sea surface. The comparisons between predictions and full scale measurements show reasonable agreements.

View full abstract

This paper describes the following studies carried out from a viewpoint of the propeller noise reduction of the supporting ship for the deep submergence research vehicle, using controllable pitch propellers (CPP) of the previously built supporting ship “Natsushima” :
1. Noise measurements of full scale propeller
2. Noise measurements of model propeller in cavitation tunnel
3. Theoretical analysis of CPP operating condition giving rise to minimum propeller noise
From propeller noise levels in various navigating conditions measured in the full scale test, it is found that there exist optimum combinations of propeller blade angle and revolution number which bring minimum noise.
This fact is very important to reduce the propeller noise of the supporting ship which is wide in the propeller operation range.
The model test also shows the existence of similar optimum combinations bringing low propeller noise and that these phenomena are strongly related to the occurrence of the propeller cavitation.
Theoretical analysis of the propeller cavitation occurrence conditions by use of the propeller lifting surface program proves that the above optimum combinations are in good agreement with those giving the shock free entry conditions of the propeller.

View full abstract

This study follows to the author's former theoretical study on method for estimation of bubble density distribution by acoustic attenuation.
The remaining problems are discussed for the realization of real time indication system of bubble distribution, such as the maximum acoustic pressure for linear vibration and building-up time of vibration.
Under these conditions, experiments to measure bubble density distribution functions both photographically and acoustically were carried out using bubble generator. Results were compared with each other and it was found that coincidence of these results were very good. Thus, the acoustic attenuation method was verified to be good for bubble distribution measurement.

View full abstract

A semi-submerged displacement-type high-speed ship with wings is studied on the possibility as a medium transport-distance vehicle. The study is mainly concerned as to resistance characteristics and submergence capability.
The vehicle, called HSV (High-speed Submergible Vehicle with wings) here, is consisting of a main hull, main wings and a strut to support super-structure. The HSV floats with a certain draft like a convensional ship while it is still, and it submerges due to a downward lifting force by the main wings when it moves.
The lift force by the main wings and the submergence depth were measured to be confirmed that HSV could submerge as expected. The wave-making resistance is remarkably reduced by the submergence. The wings contribute to reduce the wave-making resistance also through its interference with the waves generated by the hull.
The estimation of the effective horsepower required about 40000 PS which was not actually acceptable. One half of the estimated horsepower was frictional resistance component. It can be expected that the wave-making resistance and the viscous pressure resistance can be still reduced by a modification of the hull and wing forms. However an innovative invention must be necessary for the frictional resistance to be reduced.

View full abstract

Hydrodynamical configuration design is made for a hydrofoil catamaran equipped with a flap system. In spite of the additional elements and the larger hydrofoil area the resistance derived from towing test shows same order of magnitude with the simpler models described in the first report. It is also demonstrated that the use of the flap system extends the speed range of low resistance with high elevation of the hull. The ship motion tests in regular incident waves show that the heave and pitch transfer functions are on the much lower level than the other designs currently used.

View full abstract

The validity of a linear solution is investigated for an oscillating and translating problem of a surface-piercing body. In order to avoid numerical inaccuracy, analyses are concerned only with the two-dimensional radiation problem.
In the first half of the present paper, studies are made of a solution satisfying the classical free-surface condition with only the effect of uniform stream taken into account. The method of integral equation with the so-called line-integral term is applied to obtain a numerical solution. This solution has the property that the fluid velocity is bounded at intersections between the body and free surfaces. The same solution is also constructed by the method of multipole expansion, which is newly developed in the present study. It is found that additional contributions to the rate of energy flux arise from the intersections in the case of surface-piercing problem with forward velocity.
In the second half, the hybrid method is developed to obtain a solution which satisfies the free-surface condition with the effect of steady perturbations included. Contrary to the classical solution, this modified solution gives no contributions to the rate of energy flux at the intersections. Computations are performed for the added-mass and damping coefficients of a half-immersed circular cylinder. The integrated pressure forces are not much influenced by the forward speed and by the steady perturbations. However these effects are not necessarily small upon the distribution of velocity potential on the body and free surfaces.

View full abstract

The displacement effect in the theory of body motion in waves is discussed from the view-point of perturbation theory.
First of all, in order to clarify the physical meaning of the displacement effect, a body motion in an unbounded fluid is discussed. Then the first order problem of a body motion in a uniform flow with free surface and the second order problem of a body motion in a still water with free surface are discussed.
It becomes clear that the displacement effect appears in the velocity potential by making use of the space coordinate, and then it cancels out the displacement effect which appears in the pressure equation, when the pressure on the body is calculated.

View full abstract

This paper describes the influence of shipform characteristics on resistance increase and speed loss in wind and waves. Resistance increase and speed loss are estimated by calculation. The estimations are carried out for 4 kinds of ships, which are general cargo ships, bulk carriers, pure car carriers and container ships. The particulars of these ships are standardized by auther's former research based on the data of many actual ships.
The relations between ship's dimensions and enlargement of resistance increase on each kind of ships are analyzed. And the estimated results of resistance increase of the standardized ships are showed in the form of residual resistance.
Furthermore, ship's performance and required power in still water of these standardized ships are supposed by using the latest data of actual ships. Speed loss in wind and waves is predicted by using those performance data. And finally the relations between ship's dimensions and speed loss on each kind of ships are analyzed.
The result of these analyses shows that huge ships with low service speed have a trend of causing a great deal of speed loss in wind and waves.

View full abstract

An observation of bottom impact phenomena on a flat bottomed ship has been made by means of a high speed video recording system and the transparent model of a tanker. The model was towed in regular waves while making pitching and heaving motions freely. Two video cameras were used in the experiments. One camera was set to take a picture of bottom impact through the transparent hull and the other camera was used to record the impact pressure in an oscilloscope. Both pictures were superimposed for synchronism between pressure fluctuation and the sequence of events. Three kinds of impact phenomena were observed during the experiments. One was oblique impact, another was trapped air impact and the other one was normal impact. The video shows that the high peaked pressure in the wide area appears in the case of trapped air impact while the pressure gauge is in the trapped air and the so called vibration of compressed air follows. The normal impact which is characterized by earlier touching of the forepart of the bottom were examined to show that its high pressure are related to the rapid expansion of the wetted area. Discussion are also made on the possibility of coexistence of trapped air and the normal impacts, based on the experimental results. The correspondence of the events to the pressure fluctuation were examined to show that the oblique impact could be explained by the estimation by one of authors. In every cases it is seen that the three-dimensionality plays important role in the bottom impact phenomena.

View full abstract

This paper is concerned with statistical estimation of total second order responses including the slow drift oscillation of a moored floating offshore structure in random seas. Stochastic interference between a wave frequency response and the slow drift oscillation affecting on the probability distribution is discussed.
Assuming that the responses are represented in the form of two term Volterra functional series, an extended theory of probability density functions of an instantaneous response and its maxima are developed from the previous paper. It is applied to sway of moored floating semi-circular and rectangular cylinders, and confirmed its accuracy comparing with the exact solution in case of a pure second order motion at first. Next, non-normality of the total second order motion is investigated. It is shown that it is significant especially in case of heavy damping of the system and totally different from non-normality of the pure second order motion. The interference between the first and second order responses plays important roles on the probability density and the extreme value of the total response.

View full abstract

The previous paper showed results of slow motion forced oscillation tests in regular waves concerning slow drift oscillation of a floating body moored in random seas. Systematic combinations of waveheight, wave frequency, motion amplitude and motion frequency were chosen as test parameters. In equation of hydrodynamic forces which is used for analysis of the measure forces, wave orbital velocity is taken into account in relative velocity to the model in Morison's drag term in this paper, although it was neglected in the previous one. Obtained results of wave drift damping are in much better agreement each others than the previous ones, although waveheight, wave frequency, motion amplitude and motion frequency are different for them.
In addition to a box-shape barge which is tested in the previous paper, tanker and semi-submersible models are newly tested. This paper shows the results of wave drift dampings for these three models as well as drag coefficient increases for them due to waves. Drag coefficient increases due to waves for a tanker and a box-shape barge are less important than those for a semi-submersible.

View full abstract

The objectives of present paper are to formulate the nonlinear diffraction problem for the long waves induced by a short wave groups using the multiple scales perturbation method and to find the effect of 2 nd-order long waves on the ocean structures.
The governing equation for the 1 st-order zeroth harmonic potential is obtained by using the solvability condition of the boundary value problem for 3 rd-order zeroth harmonic potential. The formulas for the surface elevation of the 2 nd-order long waves induced by modulated short wave groups are derived analytically.
Numerical results are presented for the case where the envelope of incident waves-train is a sine function and the influences of the long waves around a huge circular cylinder are considered.

View full abstract

To have an understanding of why and how slow drift damping forces of floating structures increase in waves, an experimental study is carried out using forced oscillation techniques. The object of this study is set on the viscous damping force. Two types of ocean structure are selected as the model, Tension Leg Platform and Semisubmersible. In case of TLP model the viscous drag force acting on columns (vertical cylinders) occupies most of the drag force on the total model, on the other hand, the drag force on lower hulls (horizontal cylinder) of semisubmersible is given a relative importance. These models are selected because the viscous drag forces acting on the models are dominant in the slow drift damping force and the differences in the characteristics of the drag forces acting on the vertical and horizontal cylinders should be made clear.
With the view of investigating the damping forces for the slow drift oscillation, the viscous drag forces acting on models forced to surge in combined two harmonic motions in low and high frequencies are measured and those on the models forced to surge with large amplitude and low frequency in regular waves are also measured. In the experiments on the TLP model, forces acting on the columns and lower hulls are measured separately. The drag coefficients obtained from these experiments are compared with those obtained from the simple harmonic forced surging tests. Experiments show that the drag coefficients of models performing the slow drift oscillation in waves depend on the reduced velocity Ur and in the large Ur region they are higher than those obtained from simple harmonic oscillation tests in still water. The rate of increase in the drag coefficient of the horizontal cylinder is much higher than that of vertical cylinder. By this reason the rate of increase in the slow drift damping force acting on the semisubmersible is greater than that for TLP model.

View full abstract

Generally the critical motions and relative water surface of a moored semisubmersible are very complicate under combined external disturbances ; namely steady wind, dynamic variable winds and irregular waves. This paper deals with the steady and variable parts using frequency domain approach. The analysis is done for wave frequency and very low frequency. Theoretical predictions are compared to experimental results and a good agreement is obtained.
Also the effect of the damping coefficient on the response spectrum estimation of very low frequency is discussed, and the cause for scattered statistical values estimated from response spectrum is investigated. Finally, a practical method for total prediction for motion and relative motion including the very slowly varying phenomena and steady drift parts is introduced.

View full abstract

The paper discusses the dynamic responses of a semi-submersible platform under compound external forces, namely irregular wave, irregular wind force and steady current. A time domain numerical simulation technique is described and it is shown to give fairly good agreement with experimental results correspond to survival condition. It is pointed out that the lift force acted on the lower hull cannot be neglected and is important to estimate the lower frequencies component of heave response. Furthermore, it is shown that the frequency depending damping force coefficient can be represented by that at natural frequency of motion.

View full abstract

In the previous papers by the present authors, it was shown that a particular shape of the thin submerged plate (a “Crescent type”) would be the best one for concentrating ocean wave energy at a certain focal zone. This was demonstrated by performing some foundamental hydraulic tests as well as some numerical simulation.
However, in order to proceed further to next practical step, it might be thought that the wave force distribution around the structure should be clarified in detail, and at the same time, an appropriate design procedure should be implemented. Therefore, in the present paper, the wave force distribution will be studied by both experimental and theoretical methods in the first place. And subsequently a necessary design procedure for the plate, including its actual construction method, will be described. Some design proposals will be reported for the site conditions of representative coastal areas in Japan.

View full abstract

A method for estimating the linear hydrodynamic forces acting on unmanned untethered submersibles is investigated. Two models are selected as examples, one has a slender body with a purpose to lower drag as in the case of a laminar flow aerofoil section, and the other has a body of an aerofoil section suited for gliding performance.
The estimated values of stability derivatives are examined and compared with the measured values by PMM test, and are proved to be satisfactorily accurate. To analyze dynamic motion characteristics, simulations of stability and control response are carried out. The advantage of the proposed method of estimation are demonstrated for the stage of initial design.

View full abstract

It is well known that a single anchored ship causes often the slewing motion by disturbances such as current, wind and wave. This slewing motion will be undesirable factor on anchoring ship, and it will be of importance to restraint this slewing motion for safety of navigation, operation and traffic control at the restricted sea.
In this paper, two types of control system are proposed for restraint the slewing motion of a single anchored ship, and the characteristics of each type of control system are discussed by simulation of the slewing motion. One of these types is named “WIDE CONTROL”. This aims that the ship is always in equilibrium condition under external disturbances such as current, wind and wave. The other type is “NARROW CONTROL” which aims that the heading direction of ship coincides with the direction of current. The former control will be useful for anchorage being unrestricted sea, the latter control for restricted sea.
From the results of simulation on restraint of slewing motion, it is shown that the slewing motion of a single anchored ship will be restrained by using the two types of control system proposed in this paper properly corresponding to the external disturbances.

View full abstract

Authors have been studying a guidance and control system for an autonomus underwater vehicle of a streamlined cruising type, which is appropriate for deep sea investigation in a wide area for its high efficiency of energy consumption, and it does not need safety equipment for human operators. However, in the unmanned control, real time utilization of visual information for control and other decision making, which is a major advantage of human operation, is difficult at present.
In the study, basic methods to estimate motion or attitude of the vehicle (camera) from video data which is obtained continuously by a video recorder and/or a camera are investigated, and several alternative methods to estimate the vehicle motion are evaluated by actual sample pictures. Finally, possibility of the motion estimation from visual information is investigated by applying those methods to actual underwater video images.
From the study, it is assured that estimation of camera motion (ego motion) from usual landscape pictures is possible by simpler and quicker methods. on the other hand, it is not so easy at the sea bottom, for there are few landmarks and visual pattern does not change clearly. However, by using full information and selecting appropriate estimating points, possibility of the motion estimation at the sea bottom is assured for the case human operator can recognize the motion visually.

View full abstract

In the present report, attempts are done to analyze problems of information processing and decision-making of human pilots in berthing maneuvers of large ships and to specify supporting procedures for them from the view-point of human engineering.
At first, the procedures to supply the information on ship situations to a pilot are proposed for both maneuvers of approaching and berthing. It is pointed out that a pilot meets with a difficulty in grasping the situations of control devices because of too much information from various routes and a kind of one board display system is recommended for showing compactly the situations of control devices. The problems in the communication among a pilot, captains of tugs, operators on a berth and assistant members in a bridge are also discussed in detail. Lastly, several problems are discussed for improving the system performance as a whole.

View full abstract

In this research, the marine traffic control is distributed into each ship in the water area, and through the distribution of the traffic control we propose a new maneuvering style, called cooperative maneuvering. This paper presents the proposal and the development of a problem solving approach to realize the new maneuvering style.
Cooperative maneuvering represents both the distributed traffic control strategy and a new maneuvering style. Its most important feature is the cooperation between ships by using inter-ship communication. The distribution of control entails a series of problems like how ships should talk with each other, how a ship should make her decisions of action which do not interfer with actions of other ships, etc. To solve these problems we developed a distributed problem solving approach by addressing the nature of cooperations in cooperative maneuvering.

View full abstract

To build a computer aided design system, it is necessary to develope a suitable model for representing design objects like a ship and design processes that determines the mechanism of redesigning strategy. In this research, design object is described as a combination of a data structure and a set of constraints which restricts the behavior of some related design para-meters. And design process is represented as if-then rules which are expression of a skillfull engineer's knowhow of design.
The object-oriented representation form is employed to develope a computer aided ship basic designing system. In this system, design object, design parameter, constraint and designer are represented as “object”. An object is a data structure with argorithms called method describing its own behavior or characteristic. Object-oriented representation is suitable for constructing design systems because : (1) Complicated data structures and their behavior can be easily represented, (2) Using the function of inheritance one can put many similar objects in order efficiently, (3) Message passing paradigm makes it easy to organize the whole system, (4) An object is independant and self-supporting so the system operating on the object-oriented environment has excellence in maintenance. We implemented constraint-riented system by defining constraints as objects.
In using this system, we can easily add or delete design parameters, constraints and if-then rules, so we are able to build or modify the model structure and knowledge-base flexibly. Finally, some application results of basic design of ships are presented to show the effectiveness of the system.

View full abstract

Integrated CAD/CAM is required in many industrial fields. Building integrated CAD/CAM for ship-building, we have to consider of characteristic problems such as how to input many detailed information of parts and how to measure against frequent design changes. In this paper we propose a description method of hull structures by relation and explain geometrical reasoning to fix figures from this description.
We also propose a parts decomposition method applying the simplifying and integrating rules. Our experimental program is able to deal with objects consist of planes only, but improved rules are needed to put this program into practical use with curved surfaces.

View full abstract

Since many design variables affect each other for ship structural design, it is difficult to decide them all at one time. Generally, a large number of trials in updating those variables is necessary to complete the design in order to satisfy the requirements by a customer and some other conditions. Hence, constructing expert systems for helping ship structural design have been desired these days.
In this paper, the expert system for aiding constructing a bulk-carrier is studied with emphasis upon the structural design of the midship section. The expert system is written in LISP which makes it possible to have more consultative and interactive systems. A knowledge base is called for deciding design variables, which are checked in a rule base and are updated until all conditions are satisfied. After assuming the initial design variables at the midship section with satisfying NK-rules locally, the longitudinal strength of the section is computed. An optimization technique is developed for obtaining the required longitudinal strength at the midship section in which the minimum weight is chosen to be an objective with handling the design variables as discrete numbers.
For an illustration of the system, a bulk-carrier is designed. The procedure of deciding the midship section is demonstrated in this paper which started from the two requirements such as “dead weight” and “speed of ship”. In order to check the efficiency of the created system, the midship section is compared with one having the same principal requirements.

View full abstract

In order to investigate the underwater noise radiated from ships, as the first step, the study on the radiated sound from a rectangular panel is carried out.
The vibration of a panel is represented by using the monopole and the dipole, and the sound pressure fluctuation radiated from the panel is calculated. The calculated sound pressure have a good agreement with the measured results. The sound pressure radiated from a panel vibrating with free surface around it is smaller than that covered by the baffle plate. And the radiation efficiency also becomes small when the panel vibrates with free surface around it. Therefore, the measured radiation efficiency of a rectangular panel vibrating with the free surface around it cannot be applied for the shell plate of actual ship.
The natural frequency of the rectangular panel in the water is estimated by Kito's method, and the radiation efficiency of underwater sound of a panel is calculated by applying Maidanik's method. These results correspond with the mean value of the results which are calculated by using the monopole for a panel vibrating in a single mode. Then, it is considered that the radiation efficiency estimated by these procedure can be applied for a actual ship.

View full abstract

Following the previous report which proposed a new calculation model for crank shaftings of the long stroke marine diesel engine and an evaluation method for the stiffness of crank throws, a calculation method for the coupled axial-torsional forced vibration is presented in this report. In the method, conversion of the crank pin force into the equivalent axial force is newly introduced. Moreover, accurate axial-torsional vibration measurements were carried out during both shop and sea trials. Minute analysis on the test results revealed the mechanism of the vibration.
Calculation results by the new method show fairly good agreement with the measured ones only except for higher harmonics oscillation at the resonance of 1-node torsional vibration, which is recognized to be caused by non-linearity of the thrust block stiffness.

View full abstract

In this paper numerical solution method for spatial elastic linkage mechanism with clearances and friction is presented. Deformation of each linkage member is considered to be infinitesimal, and therefore each member is analyzed by conventional finite element method. And deformation vector is separated from displacement one at each time step by using displacement function of a rigid body. Contact, friction, and viscous force vectors are calculated in the same way as Rigid Bodies-Spring Model. Finally the present method is applied to a slider-and-crank mechanism and obtained results are compared with the experimental ones.

View full abstract

Sea trials were conducted featuring measurement of wave impact loads and hull response for a patrol boat with displacement about 1, 300 ton, while the boat ran in severe waves with angle of wave encounters and speeds changed.
Some results of the analysis of the trial data were reported in the first¹⁵⁾ paper, in which bow flare slamming, impact pressure on the bow flare and very high logarithmic decrement of whipping vibration were discussed.
This paper presents some results of the further analysis of the data, as follows :
1. Bow flare slamming induced very high bending stresses on the deck, whose distribution on it was given, and relation between local stress and deck stress are also discussed.
2. Bow flare slamming induced whipping vibration. Careful observation of measured data led to understanding how whipping vibration was initiated by bow flare slamming, which also implied the mechanism of high decay of whipping vibration at its initial stage.
3. Whipping stresses have so high peak values and high frequencies compared with ordinary wave bending stresses that they may cause severe fatigue problems. Based on the measured data, fatigue strength of the hull was diseussed.

View full abstract

118 cases of ship collision accidents were surveyed in the records of Japan marine courts, for the purpose of assessing the damage state of the struck ships. Hayashi's quantification theory type I was applied to the damage assessment. The damage grades were expressed as 1 to 5 points in accordance with the descriptive words of the damage. Predictive damage grades as criterion variables were obtained in five classes of struck ships which were classified in terms of the gross tonnage. Here the ratio of mass of a striking ship to that of a struck ship and the striking speed were selected as the main items of the cause to damage grade.
The results obtained are as follows :
(1) The present damage assessment is verified to lie on a significant level by an analysis of variance, for the collision cases of struck ships below M class (below 699 G. T.).
(2) The category scores of the items used in the quantification theory indicate that a striking speed has a great influence on actual collision damage, as is predicted by an equation of the kinetic energy lost in the collision.
(3) Fuzzy reasoning approach can be useful for linguistic evaluations pertaining to structural damage under special conditions of collision.

View full abstract

The strength of ship structure in collision is one of the important matters in the term of saving crew lives and preventing a heavy disaster from the economic point of view, in the case of which the collided ship are earring dangerous cargoes such as liquified natural gas and so on.
It must be nessesary to estimate the time history of dynami ccollision force and if necessarily, the penetration of the collided ship in order to design ship structure.
In this paper, the time history of dynamic collision force was calculated by introducing mass, spring and dashpots system in the substructures which were localy striked and by considering motions and hull vibration of lateraly collided ship. And extending this method, the penetration of the collided ship was approximately estimated by using static destructive test result.
From these result, the following conclusions were obtained.
(1) When the ratio of damping coefficient to spring coefficient was not very small, impulsive force acted on the localy striked substructure at the first stage of collision.
(2) The smaller ratio is, the larger penetration is.
(3) In order to avoid over-estimation of the dynamic collision force and penetration, ship motions and hull vibration must be considered.

View full abstract

When offshore structures are collided with supply boats, the damages or the plastic deformations are limited within the collided membrers in most cases. The supporting structures which support the collided member only oscillate elastically. Though no damage is caused in themselves, the dynamic characteristics of the supporting structures are very important to study the motion of the collided members and to predict the degree of the damages.
To clarify the dynamic.characteristics of the supporting structures, the authors employed the modal synthesis. The types of the dynamic characteristics are classified using the number of the dominant vibration modes as an index. Further, it is shown that the index can be related to direct physical parameters such as collision time and vibration period of the structure. It is also shown that these parameters can be estimated using the static finite element analysis and a simple dynamic analysis of a spring-mass model with three degree of freedom. Based on the above study, a flowchart which shows the procedure to model or to idealize the given collision problem is proposed and its validity is demonstrated through numerical examples.

View full abstract

The conversion method to CTOD from measured clip gauge displacement for a through thickness center cracked specimen and a three point bend CTOD specimen has been already established, for the purpose of obtaining the fracture toughness of the material. However the method for a semielliptical surface flaw has not been established yet because of difficulties of considering the three dimensional effect for the COD.
In this paper, we assumed that COD for a surface flaw on the cross sectional plane perpendicular to the plate surface, which is the plane with a single edge crack, can be approximately given by the superposition of CODs for the single edge cracked plane under bending loading and for the center cracked plane under tensile loading, of which the center line corresponds to the plate surface. The bending and the tensile load are determined by the measured mouth COD and crack angle at the surface of the cracked body with a surface flaw.
Tensile and bending tests for specimens with various sized semi-elliptical surface notches were carried out to confirm the above idea. In the tests, mouth COD and crack angle were measured by using double clip gauge method and simultaneously, the COD inside the surface notch was also measured by using silicon gum casting method. As a result, the converting method to CTOD at the deepest point of a surface flaw is driven from the measured mouth COD and crack angle at the surface which are obtained by double clip gauge method.
The testing result of wide plate estst with a semi-elliptical surface fatigue crack on a multipass welded joint, of which crack tip located at local brittle zone within HAZ, was also analyzed. It was obtained that the critical CTOD for a surface crack of which the tip locates local brittle zone corresponds to the lowest critical CTOD obtained from the three point bend standard CTOD tests of the joint with a through thickness fatigue pre-crack.

View full abstract

Fatigue life has a wide scatter and this fact makes fatigue design of structural members difficult. In order to make fatigue life distributions clear, it is required to prepare many number of specimens and to repeat the fatigue tests under the same loading condition. For such fatigue tests, it usually takes much time or cost. It is still more for fatigue tests such as random fatigue tests or corrosion fatigue tests in long life range.
In this study, in order to reduce the time for experiment, a fatigue testing method using a multi-notched specimen which has many notches of the same shape and its reliability analysis method are developed. The fatigue test procedure by multi-notched specimen is as follows : First, loads are applied from the start of cycling up to certain cycles (N1). Then fatigue cracks are observed with microscope and the number of notches (m1) where fatigue cracks have initiated is counted. Again, loadings of certain cycles (ΔN) are continued (N2=N1+ ΔN). Then the number of cracks (m2) newly initiated during ΔN cycles are counted. Repeating the process until the longest crack initiated in the specimen reaches the length that is defined before the test, a set of data (N1, m1), (N2, m2), …, (Nk, mk) is obtained as an event.
The likelihood function of this event can be calculated employing multinomial distribution. Two parameter Weibull distributions are used as fatigue life distributions in this study. The Weibull parameters are estimated by the Bayesian reliability analysis or the likelihood ratio analysis.
From the numerical calculations assuming a multi-notched specimen with sixty notches and the experiments using a multi-circular holed specimen, it is concluded that two or three multi-notched specimens are enough to estimate the fatigue life distributions and the time for test can be decreased below one-tenth comparing with the fatigue tests using a single notched specimen, with maintaining the reliabilities of the estimated fatigue life distributions.

View full abstract

The purpose of this paper is to set up a general method for estimating the reliability of ocean structural member with/without leak sensor through periodic in-service inspections. Emphasis is put on the classifications of all results which are obtained by the inspections into probability events and to make the probabilities of occurrences of each event clear at arbitrary inspection times.
Choosing a connecting component of TLP tendon, which has a possibility of failure before and after the fatigue crack initiation, reliability analyses of inspections are carried out. And the effects of the fatigue properties of member, the failure rate of member, detectability of a crack by inspection and the shape of initiated crack on the relationship between structural reliability and inspection interval are discussed. Further, the reliability analyses of inspections under the condition of the leak sensor is attached inside of the pipe and under the condition of the crack initiation life and crack propagation life are random variables at the same time are developed.

View full abstract

Estimation on the strength of bolted wood joints is indispensable in designing wood structures. According to the previous investigations, it has been proven that the yield theory which is a kind of limit analysis method is useful in estimating collapse loads of bolted wood joints, and several formulas have been given for the most common joint models such as single shear joints and symmetrical double shear joints.
In actual wood structures, however, joints are much more complicated where asymmetrical double shear joints are commonly used, and usually several wood members which cross in each direction are connected with various kinds of nails, screws and bolts with each type of joint. Moreover, there have been very few papers investigating into the effect of frictional force on joint strength which plays an important role in the estimation of the strength of the joints.
In this paper, the behavioral characteristics of simple wood joints, single shear joints and asymmetrical shear joints with one bolt were examined by taking into consideration of the effect of the frictional force, then, a multiple joint which consisted of four members and was connected with four kinds of bolts was investigated. All models tested were of the type mainly used in ships. It was found that the behavior of these wood joints was satisfactorily explained by the yield theory and by the usual beam theory on elastic foundation.

View full abstract

The strength of bottom slab of PC-barge under the operation of alighting on the sea bed was studied experimentally and theoretically. Large cone penetration tests were performed on three model slabs. The penetration-resistance curves were obtained. The initial destruction mode was the punching shear, and then the concrete crashed. Before crashing out of the concrete, the punching shear strength of JACE's formula was in good agreement with the resistance. After crashing out of the concrete, the resistance was generated from tensile forces of reinforcements and prestressing strands.
The bottom slab strength against the local contact forces of the foundation soils, the side shell strength against the local ice pressure loads, and the strength against the ship collision were discussed. The local contact forces of the foundation soils were too small to cause damage to the bottom slab. However, the local ice pressure loads were very huge, so that it was almost impossible to design the side shell free from damage. The steel-concrete sandwich wall would be the better structural element against the local ice pressure loads than the PC-slab dealt with in this paper. The strength against the ship collision was evaluated with the energy absorption capability. The absorbed energy of the PC models were much more than that of the steel models.

View full abstract

The authors have developed a new composite steel-concrete sandwich system for the strength members of huge offshore structures, where concrete is placed in between steel plates. In the previous papers, a series of experiments under static and repeated loadings, and theoretical analyses using the nonlinear finite element method were carried out. It was clarified that the composite sandwich structure has various excellent properties.
In the present paper, the authors carried out two types of experiments. The first type of experiment was to study the strength of the composite structure under distributed loading which corresponds to ice loadings acting upon offshore structures in the Arctic. The second type of experiment was carried out using the composite structure reinforced by various types of shear connectors. It was found that T-stiffeners and long stud bolts are quite effective to increase the ultimate strength and the amout of deformation.

View full abstract

The mean crushing loads, plastic buckling loads and crushing strokes in the axisymmetric collapse mode for circular cylinders of various radius-to-thickness ratios were analytically and experimentally investigated. The finite element analysis (using axisymmetric solid elements) and the crush tests were conducted for 8 mild steel cylinders with different R/t ratios. The results obtained are summarized as follows :
(1) The plastic buckling loads due to Gerard's theory agree well with the maximum loads obtained in the experiments with R/t>20. But if R/t is below 20, Gerard's solutions underestimate the loads.
(2) The mean crushing loads which the present finite element code predicted, agree well with the experimental results and are improved solutions in comparision with the existing rigid-plastic solutions and the empirical formula. The above numerical results confirmed that the thin-walled assumption can not be maintained for relatively thick cylinders which are considered approximately thin in the linear problem.
(3) For the mean crushing load in case of axisymmetric collapse the modified Alexander's solution is in better agreement with experimental results with R/t>30 than the other existing theories and the empirical formula. When the axisymmetric and the non-axisymmetric collapse are both considered, Magee's empirical formula is sufficiently accurate for the thin to thick walled range.
(4) The crushing half wavelengths and crushing strokes are in good agreement with the solutions of the tangent modulus theory and the modified Alexander's theory respectively, confirming the results of Ref. (2).

View full abstract

Many studies have been done experimentally and theoretically for the general instability of a cylindrical shell reinforced with ring stiffeners caused by external hydrostatic pressure. In the case of shallow stiffeners, it has been found that the theoretical model has a goodagreement with experimental results for a cylinder simply supported at both ends. However, these theories do not work well for the problem where the ring stiffener become large.
An effective breadth of cylindrical shell with the ring stiffener has also been studied by several researchers. However, the equations derived by them for determining the breadth do not have strong theoretical background.
In this paper, a reinforced cylindrical shell with ring stiffeners is analyzed based on, the energy method, where local deformation between ring stiffeners is superposed on overall buckling mode as in the same manner as Kendrick's research. As far as the continuity conditions at the junction of the shell plate and ring stiffeners is concerned, displacement compatibility is imposed in the present paper instead of strain continuity conditions employed in the many previous researches. The present method accurately determines the critical pressure not only for shallow but also for deep ring stiffeners. Several problems are solved and compared with experimental results and the solutions by other reseachers. These comparisons show that the proposed method is effective for determing the critical pressure of the reinforced cylindrical shell. The effecive breadth of shell plate is also determined from the results of critical pressure.

View full abstract

In the first report, two of the authors developed the approximate methods of calculation of Shear Lag proposed by Kuranishi M. and Miyagawa M. to the multi-cellular box girder like as a bulk carrier. And after some numerical calculations on the PANAMAX-TYPE of bulk carrier, they concluded that the warping and warping normal stresses would become large in the gunnwale part, bilge part and the vicinity of ship centre line of double bottom part, and so on, when ship is loaded alternatively.
In this report, the authors try to confirm the reliability of the approximate methods of calculation developed previously by three dimensional Finite Element Idealization on 5-hold parts of the PANAMAX-TYPE of bulk carrier, and then try to make clear the effects of number of side girders upon the magnitude of warping and the warping normal and shearing stresses in double bottom structures.
By these investigations, the following results were obtained :
(1) The results of calculation by the approximate method developed previously agree fairly well with those of calculation by 3 D-FEM.
(2) In this calculation, the shearing force diagram of hull girder should be corrected by the effects of transverse bulkheads.
(3) The number of side girders in double bottom structure little affects the magnitude of warping and warping normal and shearing stresses, so far as the total cross sectional area of side girders remain constant.

View full abstract

It has been widely recognized that Al technique or knowledge engineering should be incorporated in engineering practice in order to improve product's quality or reduce human labors. Expert systems for supporting ship manoeuvring or subdivision design of tankers have been developed in recent years.
The present paper proposes expert system prototype for supporting buckling design of longitudinal strength in bulk carriers, which experienced engineers can perform. In ship design, it is a key role to minimize a routine work composed of analysis and synthesis, so that satisfactory results can be easily obtained by keeping safety of structure. There may be two important difficulties in building a system, that is, efficient data-control supporting re-analysis, and programming of complex ramification for plastic analysis or designer's know-how. This study show the validity of Al technique to resolve these two difficulties and minimize human labor irrelevant to intellectual design practice.

View full abstract

An Idealized Structural Unit model of a tubular bracing member is developed considering the influence of local buckling which takes place after the ultimate strength is attained. That is,
(1) The simplified method which was proposed in the 1 st report for a tubular member subjected to eccentric axial thrust is expanded so that the ultimate strength can be evaluated under general end loads and distributed lateral load. This method is applied to determine the ultimate strength of a member in the framework of the Idealized Structural Unit Method (ISUM) analysis.
(2) An approximate method is proposed to determine the internal forces at a cross section after the ultimate strength is attained. Using these internal forces, strain distribution is determined. The occurence of local buckling is checked based on this strain at every load increments.
(3) After the local buckling has started, its influence is reflected upon the full plastic strength interaction relationship through plastic nodal displacements of the element. A formulation to derive the elastic-plastic stiffness matrix is described considering the influence of local buckling.
Several examples are analyzed applying the newly developed element. The calculated results are compared with those by the finite element method analysis and experiment, and its validity and usefulness are demonstrated.

View full abstract

Though a lot of experimental and theoretical investigations on the strength of stiffened plates in compression have been carried out, there are few to suggest the rational correlation between panel and stiffener associated with their local buckling and collapse. The main purpose of the paper is to present a method of analysis which may be effective not only in estimation of ultimate strength but also in mode shape analysis taking account of the interaction between panel and stiffener.
The proposed procedure is composed of two parts, elastic large deformation analysis and rigid-plastic analysis. The fundamental concept of method was proposed by Fujitao and it has been developed to be applicable to the problem relating with panel-stiffener interaction. The presented method has been verified through comparing its results with the experiment and FEM calculation and it has been found to be practical for the estimation of ultimate load and post-buckling behavior of stiffened plates with flatbar stiffeners.
Through the mode shape analysis, has been obtained a clear scope for the local collapse mode, which had been most commonly observed in the experiment conducted with continuous multi-span models.
The critical slenderness ratio of flatbar stiffener has been studied through the analyses. They show that there is a threshold of the slenderness ratio, h/t_s, over which the dominant buckling mode in a stiffener changes from a panel-buckling-induced one to an inter-frame overall and the buckling strength rapidly decreases. The critical value is found to be about 18, which indicates the existing rule requirements are a little too conservative.

View full abstract

A number of papers on structural mechanics and strength of tubular joints of offshore structures have been published. Most papers of them treated simple nonstiffened joints and only small number of papers treated stiffened joints of members with small diameter, where simple gussets or ring stiffeners were used as stiffening elelmets.
Foundamental knowledges are needed, which are useful to initial design of stiffened joints of a large diameter member with small diameter members such as column to brace jojnts in semisubmersibles.
In this paper, effects of several kinds of stiffening elements used in joints mentioned above are discussed by both experiments using acrylic models and numerical calculations using a FEM program newly developed for the purpose of behavior analysis of shell structures with complicated construction. The joint models mainly discussed are T type of joints subjected to axial load of brace. A TY type of joint model is also studied experimentally and theoretically. The types of stiffening elements used in the models were diaphragm, vertical bulkhead, ring girder, vertical girder, piercing brace and so on. Each specimen of joint was fabricated as combination of a standard non-stiffened joint structure and a set of stiffening elements inserted in a chord member of the joint.
Finally, taking weights of stiffening elements into consideration, stiffening efficiency of joint models treated are compared. This comparison resulted in such a conclusion that a joint stiffened by a diaphragm and a partially piercing brace is the most effective as a T type of jojnt.

View full abstract

Active control of a tower-like offshore structure with plural hinges is discussed experimentally. This tower consists of several components which are connected mutually by hinges. Each component is a light weight structure, equipped with thruster. The natural frequency of the first mode of the tower is lower than the frequency of the waves. This kind of tower is suitable for the stationary base of activities in deep waters. Advantages of this light weight structure are the possibility of reduction of construction cost and weight of the tower. Although there is possibility of resonance with waves in the higher mode of motion or vibration and furthermore large drift can be caused in current by comparatively small rigidity, active control system is used to reduce these undesired response.
The models of the tower used in the experiments were consisted of three pipe elements connected by hinges and comparatively small. High accuracy was required for the displacement sensor. Ultrasonic sensors which employed high frequency of 1 MHz were originally designed for experiment. The motion of the each pipe was calculated from the measured distance between the transducer on the each pipe and the receiver on the monitoring post. Collected data were processed by a microcomputer and control force was calculated. Control force was generated by thruster which was built in the the body of the pipe and this thruster was also designed for this experiment.
In the first place, optimal control was applied to the unstable model fully submerged in the test tank. This situation corresponding to survival condition was successively stabilized by feedback control. In this case increased feed back gain cuased vibration of small amplitude on the models. Then drift in the currents was controlled by feedback control and then feedforward control was combined. This feedforward combined control was effective and showed good performance in respect of stability. Configuration of the models were changed from upright to zigzag configuration and upright to inclined one. In the latter case the model was successfully controlled by changing the system matrix from upright one to inclined one. Lastly correspondences between the experiments and the actual system were discussed.

View full abstract