A finite-difference method of the Navier-Stokes equations is developed to simulate viscous flows with a free surface around a ship. The body-fitted coordinates system which does not fit to the free surface is used in order to cope with ships of arbitrary forms easily. The nonlinear free surface condition is implemented by special method. Simulated results of free surface flows around a Wigley's hull at low Reynolds number are presented and compared with the experimental data.

View full abstract

The diffraction of regular waves by an advancing ship bow is studied both experimentally and numerically. The diffracted wave patterns are visualized by wave pattern pictures and the wave formation is analysed by the grid-projection method. Significant effects of nonlinearity are found, in particular occurrence of circular detached or oblique attached stationary waves depending on the Froude number. Numerical simulations of this nonlinear wave for mation about a bow are undertaken by a finite-difference method based on the Navier-Stokes equation. The computed features are in good qualitative agreement with the measurement.

View full abstract

The full Navier-Stokes equations for the flow past a Wigley model was computed using the Implicit Approximate Factorization scheme, at the Reynolds number R_e=10⁶. The Baldwin-Lomax eddy viscosity model, which is the two-layer algebraic model, was used for simulating the high Reynolds number flow. The present scheme showed good numerical stability property, and the computation took 5 hours on a large scalar machine.
The wake distributions thus obtained showed reasonable agreement with the measurements. The strong pressure oscillation, which existed in the previous computation, has been reduced dramatically by modifying the boundary condition. Though slight wiggles persisted near A. P., the pressure distributions showed good agreement with the measurements. The shear stress distribution, which was obtained using the velocity distribution, showed rather poor agreement with the measurements.

View full abstract

In order to investigate a turbulence flow with a magnetohydrodynamic (MHD) effect, a formulation of k-ε modeling with the MHD effect and its numerical simulation for an infinite channel flow are presented. The k-ε equations derived here show that a magnetic field absorbs the turbulent energy k and its dissipation rate ε. The damping effect of the magnetic field on turbulence is shown by the result of numerical simulation.

View full abstract

Following the former reports, the author discusses the oscillation of a two-dimensional body in otherwise still viscous fluid.
He introduces at first an adjoint equation of Navier-Stokes one.
Thence, he formulates also the reciprocity theorem by making use of linear solutions in the same way as in the former reports.
As a numerical example, he carries out approximate calculations on steady flows and non-linear oscillating forces induced by an oscillating circular cylinder.

View full abstract

The Rankine-source method has been combined with an integral method for boundary layer and wake so as to investigate the influence of boundary layer and wake on the free surface flow around a ship model. The head loss due to viscosity has been taken into account in the equation of the boundary condition on the free surface. The numerical solution of the free surface flow with the boundary layer and wake around 6 m long Wigley model has been obtained with aid of the technique of matrix calculation for source densities on the panel array on the hull surface along with center plane in wake and on the free surface. The computed results as for pressure distributions on the hull surface, wave profiles along the hull side and pressure, frictional and total resistances have been shown in comparison with measurements. They display good results by considering the existence of boundary layer and wake around the hull of the model.

View full abstract

Accuracy of longitudinal-cut wave analysis technique and model size scale effect appearing on wave- and wave pattern-resistance are discussed. Making use of geosims of Series 60 hull, both standard and additional ITTC Cooperative Experiments are conducted in nine organizations of Japan. Varieties and trends of the resistance components ensued through measurements and analyses procedures are surveyed.
Results are summarized as follows.
Selection of truncation point is important for getting reliable results. Effect of difference of program code on analysis results is negligible among the present organizations. Model size scale effect is recognized on wave pattern resistance, but not on wavemaking resistance. Damping of wave amplitude for the actually propagating waves is stronger than the asymptotical estimation by the linear theory. The interference of ship waves and its boundary layer and wake is considered one of the most important factors in considering Reynolds number effect.

View full abstract

This paper presents a practical method to design propellers. In the method, the form of a propeller blade section profile is represented by eleven parameters. The parameters can be divided into three groups.
(1) parameters to determine strength
(2) parameters to affect largely pressure distribution on blade surface
(3) other parameters
Values of second group parameters are determined by solving simultaneous equation under prescribed pressure distribution. Coefficients of the equation are derived from solution of propeller analysis method such as Koyama's lifting surface theory.
By the presented method, it is possible to design propellers with high performance in cavitation and efficiency, under various design conditions including wake distribution.

View full abstract

The author performed cavitation test (cavitation observation, propeller open test, fluctuating pressure measurement and noise measurement) in uniform flow to compare the characteristics between the conventional (original) and the propeller with small blade on the blade tips (bladelet propeller).
From these test, it was suggested that the reduction was confirmed about cavitation noise and fluctuating pressure amplitude, due to the reduction of the strength of tip vortex by the small blades, compared with the conventional propeller, which was observed at the model tests as shown in the first report. The reduction of sheet cavitation region was also observed. However, the cloud cavitation was occurred at the root of bladelet.

View full abstract

In this paper, the resistance increases of the ship in irregular waves are dealt with. The resistance increases of the merchant type hulls in irregular waves are obtained by the theoretical calculation
According to the studies on these results, it is shown that the resistance increase coefficient is a function of the Froude's number, the non-dimensional mean wave period and the hull characteristics. If the Froude's number and the non-dimensional mean wave period are constant, the resistance increase coefficients of a hull are expressed by the maximum value of them, and the rate to the maximum value. Using these results, the simple estimating formulae for the resistance increase of the ship in irregular waves are obtained by the statistical analysis.

View full abstract

In the previous paper, the author presented an approximate calculation method of nonlinear two-dimensional hydrodynamic problem of a ship with forward speed in large amplitude waves.
In this paper, 2 nd report, it is assumed that the three-dimensional velocity potential is composed of two-dimensional velocity potentials which are obtained by the preceding method at every cross section. And using Bernoulli's equation, the hydrodynamic pressure and the hydrodynamic force are obtained.
Pitching motions, heaving motions and shearing forces at SS. 7 1/2 in regular head waves are calculated by present method. As for the fundamental frequency component, the agreement between experimental results and calculated ones is poor. But, through the correction of forward speed and three-dimensional effects which are not taken into account exactly in calculated results, calculated results shows quite good agreement with experimental ones. As for the bi-harmonic components, the agreement between them is fairly good.

View full abstract

Experimental observations teach us that the deck wetness does not necessarily occur repeatedly even in regular waves high enough to cause it when the wave period is close to the natural rolling period of the ship. The shipping water caused by the first one or two waves exceeding the critical wave height of deck wetness reduces GM, lengthens the natural rolling period to lead to the reduction of the rolling angle, and stops wet deck from occurring repeatedly.
In order to understand the stability of ships in waves exceeding the 'critical wave height of deck wetness, we propose in this report a simplified time domain simulation to describe the ship motions from the start of deck wetness to the capsize or to the stationary oscillation around a heeling angle while taking into account above mentioned rather transient effects of shipping water. In the formulation of the time domain simulation, we use an empirical formula relating the amount of the water flowing onto deck, every time deck wetness occurs, to the excess of the predicted relative wave elevation over the bulwark-top.
We compared the numerical simulations with the time histories of motions in experiments with the cylindrical models in beam seas, from the start of deck wetness to the capsize ; we thereby concluded that the numerical simulations describe that process well.

View full abstract

This study is aiming at quantitative evaluation of crew's judgement for ship handling in waves.
This paper, as the first report of the study, explains the outline of the onboard monitoring equipment named Ship Motion Analyzing Computer System (SMACS) developed by the authors, being capable for recording simultaneously ship motion responses in waves, encounter waves and the reasons for ship's course and/or speed changes.
Some of the data obtained from three patrol boats of Maritime Safety Agency of Japan with SMACS are also shown in the paper.

View full abstract

Hydrodynamic derivatives including the effect of frequency are estimated by means of. transient manoeuvring tests. Planar motion mechanism (PMM) was used for this purpose, but partially changed to allow transient motions.
Utilizing the Fourier analysis, linear derivatives are obtained in wide frequency range and they coincide with those derived from regular PMM tests. But the reliability of the results deeply depends on the characteristics of input transient motions and on the quality of measured forces. Then, the effect of maximum amplitude and time parameters applied to the transient motions are studied.
Each pairs of derivatives (e. g. Y_υ and Y_υ, N_r and N_r.) satisfied the Kramers-Kronig relationship in nature. First, Hilbert transform was applied to the experiment results, and the relationship was confirmed in general. But, there are unnegligible differences, because of the limitation in measured frequency range. Next, supposing the causal system, each transfer function is assumed to a rational function with a finite order of Hurwitz polynomials. If a proper set of coefficients in the transfer function is identified, we need no more to confirm the Kramers-Kronig relationship. Bode's diagram of the transfer function is used for parameter identification, and each experiment result is represented by Eq. (4. 6) fairly well

View full abstract

Second-order diffraction problems for a three-dimensional body are solved by the use of simple sources and eigen functions. Numerical results of second-order velocity potentials are available as well as second-order wave forces, which are usually computed from first-order velocity potentials only, invoking Green's theorem.
Numerical analyses are done for a vertical circular cylinder in deep water so as to verify the validity of the method. Agreement between the results by the present method and existing results are satisfactory. It is also confirmed that the second-order velocity potentials on the weather side penetrate the water much deeper than the first-order ones, i. e., the second-order velocity potentials surpass the first-order ones in the depth.

View full abstract

In order to investigate the increased damping for a moored semi-submersible during lowfrequency motions in waves, free oscillation tests were conducted both in still water and regular waves. Test results show that the wave effects as well as the viscous effects have on the low-frequency damping and it seems that the interaction effects among columns play an important roll in the increased damping due to wave effects in short waves. Calculations and experiments in regular wave groups were also compared to investigate the effect of the increased damping due to waves on the low-frequency motions.

View full abstract

A prediction method for the wave induced motions of multiple floating bodies is presented. The hydrodynamic interactions among the bodies are taken into account in their exact form within the context of linear potential theory. The motions of the floating bodies, the diffraction forces due to waves and the radiation forces associated with the motions are solved simultaneously. Comparisons to experimental measurements show satisfactory agreements.
The present method is also extended to examine the motions of floating bodies connected with each other by linear elastic structural members.

View full abstract

In this paper, special attention is given to the basic assumption frequently used in riser dynamic analysis. This is that motions of a floating vessel can be calculated independently of long line structure dynamics and then motions of a floating vessel can be used in the riser dynamic analysis.
Wave actions on a floating vessel are estimated based on three dimensional potential theory and motions of a floating vessel are predicted. Riser dynamic response is theoretically calculated using finite element techniques where wave loads are estimated from Morisons' type equation. Loads from surface vessel displacements are also considered.
Finally, comparison is made between experimental results and theoretical calculations and effects of the low frequency motions on riser dynamics are verified. Results in two directional irregular waves are discussed.

View full abstract

Japan Marine Science and Technology Center carried out successfully the TLP sea test in 1986. The installation site is 41 m deep and 2. 6 km NNW offshore of Yura harbor. In the test the performance of the tension mooring system was evaluated in the real ocean environments. This paper discusses the experiences on the TLP sea test.
The TLP consists of a platform, tension legs, a sinker and a monitoring system. The platform is a semisubmersible with 12 column-footings and the principal dimensions of the platform are length overall 34 m, breadth overall 24 m, height 13. 5 m, draft 5. 5 m and displacement 527 ton.
The platform is tautly moored by the 4 tension legs which are at sea bottom connected to the sinker weighing 536 ton in water. Each tension leg is a chain filled and covered with rubber applied to absorb the shock loads and prevent the mooring line from local abrasion. The anchor connector is made robust and monitored by using underwater TV camera at any time. The total initial tension is 125 ton which is equal to 24% of the platform displacement. The maximum wave height experienced during the test period is 7. 3 m which is just under the critical condition not to hit the bottom face of the upper structure.
This research and development was performed through the Special Coordination Fund for Promoting R & D of the Science and Technology Agency of the Japanese Government, as a part of the Research on the Utilization of Marine Space by Coastal and Offshore Structure.

View full abstract

In the designing of the single point mooring (SPM) systems, slow oscillation or weather vane motion of the moored vessel is a critical problem, because the cost performance of the system on the view point of the rate of operation is affected by the excessive mooring force induced by the oscillation.
As for the theoretical approaches to estimate the oscillation, numerical simulations in the time domain have been developed and used widely, however the reliability of such simulations are not sufficient so far. Therefore it is said that model tests can not be avoided at the final or important stage of the design.
The dynamic stability analysis is another theoretical approach, which is a simple technique but has hardly been employed in the practical designing. A main reason is that the analysis can not give the amplitude of the oscillation or the mooring force but only show the stability of the moored vessel, in other words, whether the slow oscillation arises or not.
But on the basis of the slow oscillation being a critical factor, simple technique is better to be applied practically in designings, even if the limited informations about the slow oscillation can be obtained.
In this paper, practical applications of the dynamic stability analysis to the primary planning of SPM systems are discussed. And it is pointed out that some particulars of SPM systems, such as length of mooring hawser, mooring position, optimum shape of the moored vessel and effective control systems for the slow oscillation etc. can be examined by this analysis.

View full abstract

A construction method of huge offshore structures has been conceived ; that is, several unit structures constructed on land are assembled and jointed on sea. The method has such advantages that conventional docks can be utilized and the construction period can be reduced. However, as this method obliges the work on sea to be carried out under the environmental influence of wave, wind and current, the investigation of the influence, especially the wave induced vertical bending moment of a joint part is important at the design stage.
This paper presents a calculation method which gives the motions and wave loads at a joint part of a floating body assembled from two unit bodies. The hydrodynamic forces can be derived from solving the boundary value problem of two bodies floating in regular waves by means of a 3-D singularity distribution method.
The equations of the motions including wave loads as unknown external forces and moments of the two unit bodies are solved by using the conditions at the joint part.
From good agreement between the results of the calculation and model tests, it is confirmed that the introduced calculation method is effective.

View full abstract

In the first paper, a simple altitude control system for an autonomous vehicle, based on distance observation by four ultrasonic fan beams to different directions, was studied. And it was assured that the proposed system could control the vehicle to keep the predetermined altitude for most cases.
In this report, further study is made to clarify practical feasibility of the proposed system, and the following results are obtained.
(1) By using a variable directional ultrasonic beam to measure the minimum distance to the obstacle, the guidance logic can be simplified and the vehicle can avoid obstacles of the same size as the turning radius of the vehicle.
(2) For a vehicle with variable cruising speed, its attitude can be controlled well only by changing the control parameters in proportion to the thrust power.
(3) Angular velocity saturation of rudder motion causes oscillation of the attitude control, which can be suppressed by applying an optimal control algorithm based on the linear quadratic problem formulation.
(4) A three dimensional mathematical model is formulated, and the controllability of the vehicle attitude around each of three axes is investigated. By changing the vehicle attitude in 3 axes at the same time, it is assured that attitude in each axis can be controlled independently without serious interferences for the applied model.

View full abstract

In the present paper the reciprocal theorem between the ship hull vibration and the fluid vibration, that is, underwater sound is studied.
Using the reciprocal theorem of velocity potential in the fluid medium, the reciprocity of sound in the fluid systems is introduced. On the other hand, using the reciprocal theorem of mechanical vibrations, the reciprocity of ship hull vibration systems is introduced.
To combine with these reciprocities, the reciprocal theorem on the total vibrating systems is proposed, and the relationship between radiation and scattering underwater sound problem is clarified by the theorem. This reciprocity is well known as Haskind's relation in water wave engineering.
To verify its usefullness and accuracy, some numerical examples are shown for sound scattering problem of 2-dimensional semi-submerged circular cylindrical shell, and it is numerically demonstrated that the shell deflection amplitude in sound scattering problem is proportional to the Kotchin function in sound radiation problem.

View full abstract

This report presents a result of authors' study on longitudinal strength of destroyer type ships. As the first stage, we examine the design longitudinal wave loads such as vertical wave bending moment and horizontal wave bending moment and totalize wave stress induced by them.
The method of calculation is based on the concept of total system which has popular technique to predict the extreme value of wave induced loads and stress.
As the result, the extreme value of vertical wave bending stress for the probability of exceedance Q=10^-8 shows a large value as much as the conventional strength calculation by supposing a trochoidal wave with a height equal to L/20.
By using an approximate formula of extreme wave load for merchant ships proposed by Akita et al., the extreme wave load of destroyers are examined. The values can make good approximate with the formula. Regarding the above result, the standard values of wave load of the ships are proposed for the convenience of initial design.

View full abstract

Rational design of ship's longitudinal strength should, be based on accurate evaluation of bending moment and shearing force. The nonlinear strip theory and model experiments reveal that the internal force by waves in rough seas differ from the ones obtained by the linear theory. Rules of classification societies are based on the long term prediction of structural responses, which is performed under the assumption of linearity in response and uses of the linear superposition method. According to the recent studies, the effect of nonlinearity becomes to be taken into consideration in the rules of classification societies.
In the present paper, calculations are performed for a fine container ship and for a fullbodied bulk carrier using the nonlinear strip theory. Numerical results are obtained for regular waves of various wave height, wave length and wave encounter angle and for irregular waves of ISSC spectra by the nonlinear strip theory, and are compared with the result of O. S. M. : the linear theory. Results obtained are as follows :
1) Sagging moment increases as wave height increases by the effect of nolinearities, and the moment is about 170% of the one obtained by O. S. M. at the wave height 8 m in regular head sea for the container ship.
2) On the contrary, the effect of nonlinearity in hogging moment is not so significant as in sagging moment. Therefore, long term prediction of hogging moment is obtained appropriately by the linear theory.
3) Approximate prediction of long term response of bending moment is performed taking into account of the nonlinearity, and compared with the present rules.

View full abstract

In order to investigate ship structural response in ocean waves, various wave loads (hydropressure, inertia force, vertical bending moment, etc.) should be taken into consideration, including effects of phase differences between themselves.
The 1 st report1), one of approaches to the above-mentioned, deals with confirmation of accuracy of estimation method on wave load (focusing on phase differences between wave load components), development of “DIScrete Analysis Method (DISAM for short)” based on 3-D FEM and investigation on structural response of double bottom for midsh if) part in ocean waves.
As the continuation of the 1 st report, this paper deals with investigation on structural response of double bottom in ocean waves from wider point of view, and the following results are obtained :
1) For fore and aft part, in spite of increase in inertia force, stress level is low in comparison with midship part. This phenomena is considered as the effects of the following :
(a) Difference of section shape between fore/aft part and midship part.
(b) Cancellation of various wave load components.
2) Correlation between stress in different directions has the following tendency.
(a) For bottom plate and inner bottom plate, longitudinal and transverse stresses have almost the same phase ; therefore, the maximum values of them arise at almost the same time.
(b) For girder, the large phase difference between longitudinal and shear stress is expected ; therefore, the above two components do not have the maximum value at the same time.

View full abstract

This paper describes summarized results of the analysis on hull response data of a large bulk carrier in service which is obtained through on board test of hull strength monitoring system. The purpose on installing this system on board of ships is to provide the navigator with accurate information on hull response in heavy weather so that the ship master, can make an appropriate judgement for ship operation and maneuvers.
Response data of deck stress at midship and vertical acceleration at bow etc. for 2 years is analyzed from view point of studying relation between hull response and loss of speed in rough seas. The results show that extreme value of vertical acceleration at bow seems to have bounds in rough seas due to loss of speed. It is also found that speed loss ratio of the ship in rough seas is fairly large compared with container carrier and pure car carrier.
Theoretical calculation by Strip Method is made in addition, and the results show considerable decrease of vertical acceleration at bow due to speed reduction in head and bow waves.

View full abstract

Classed ships are repaired according to the degree of damage after class survey. Then the ship conditions inclusive of damage are described in the survey report in detail. However it is difficult to judge objectively the soundness of the ship from the report as it is because of a great number of different conditions. So we arranged statistically the damage data of hull structures from survey reports, and we tried to evaluate the soundness of ships quantitatively by application of quantification theory and multicriteria analysis to the data.
The main conclusions obtained are as follows ;
(1) It is found that ships can be quantitatively classified into two groups depending on the level of soundness where the influential items such as age, ship kind and structural members are represented as quantities by application of quantification theory.
(2) It is possible to evaluate the soundness of ship individually by means of concordance and discordance index which are obtained by the multicriteria analysis on the structural items, in which the weights given by quantification theory are also used.
(3) A ship which indicates a negative value in the discordance index at a survey has a tendency to indicate increasing in its value at the next survey. Reversely, a ship with a positive value in the index at a survey has a tendency to indicate decreasing in it at the next one. In other words, this means that deterioration and effect of maintenance are well reflected in the index. As the result, it is concluded that the index can be an important measure on maintenance to judge ships condition individually.

View full abstract

Crack damages are sometimes detected in offshore structures for petroleum activities. In this paper, influences of cracks on the performance of structural members and systems are theoretically investigated. A computer program SYSTAN 2 based on the Idealized Structural Unit Method is developed for an elastic-plastic analysis of two-dimensional structural systems containing cracked members. The influence of cracks on the full plastic strength interaction relationships is considered.
Example calculations are carried out with SYSTAN 2. First, the ultimate strength of a both ends simply supported tubular member with a circumferential through-thickness crack is evaluated under axial load. It has been found that compressive ultimate strength depends on the slenderness ratio of a member and the magnitude of initial deflection as well as the crack size. On the other hand, tensile ultimate strength depends only on the size of a crack.
A quasi-dynamical analysis is also performed for the transverse section of a semi-submersible drilling unit with assumed crack damages in the lower horizontal brace. The following observations have been made :
(1) The intact structure behaves elastically under the extreme cyclic wave loading.
(2) With a moderate crack damage in the brace, the cumulated plastic deformation at every wave cycle levels off after some wave cycles.
(3) With a more significant crack damage, the progressive plastic collapse of the transverse section takes place.
Further work is needed to account for the elastic-plastic crack tip behaviour and the crack growth effect.

View full abstract

The crush tests for seven specimens of square tubes under axial compression are carried out, whose results are compared with the existing theoretical solutions and the empirical formulas with respect to the buckling load, the wavelength of the crushing mode and the mean crushing load. And also the finite element analysis is conducted by using the code developed in the previous report. The obtained results can be summarized as follows :
(1) As for the buckling load the existing theoretical as well as the finite element solutions agree well with the experimental results.
(2) As for the crushing wavelength the theoretical prediction fails, however, the finite element solutions which minimize the buckling loads are in quantitatively good agreement with the experimental values. The theoretical values given by Wierzbicki which minimize the rigid-plastic solutions for the mean crushing stresses also agree well with the experimental results.
(3) As for the mean crushing load there is not a good agreement between the experimental results and the existing rigid-plastic solutions in which the strain hardening is not fully taken into account, while the empirical formula given by Magee et al. using the tensile strength instead of the yield stress has sufficient accuracy in spite of its simple form.
(4) The complicated crushing deformations can be successfully simulated by the finite element method. The obtained mean crushing loads are about 25% smaller on the average than the experimental values because of the thick-walled effect, however, there is a good agreement between them from a qualitative point of view.

View full abstract

In the present paper the flattening problem of steel tubes under lateral compression is taken as a part of the comprehensive studies on the crushing strength of structural elements, in which the finite element analysis and the crush tests are conducted for several circular and square tube specimens.
All of the major factors in the quasi-static crushing problem such as large displacements, large strains and frictional contact are taken into account in the present finite element analysis. The obtained numerical solutions, which agree well with the experimental results with respect to the crushing deformations as well as the load-displacement relations, verify the validity of the nonlinear finite element method in the crush analysis.
There remains some problems such as the effect of finite length for circular tubes, the influence of initial geometric imperfections and the thick-walled effect for square tubes, however, these problems are not so essential from a practical point of view. Therefore it can be concluded that the finite element code developed in the present study is useful for, the flattening analysis of a single tube at least with various cross-sectional shapes. The analysis of interactive behavior of several tubes and the dynamic analysis are the future problems.

View full abstract

The tension leg platform (TLP) has been studied since the 1970 s by Paulling et al., but has not been used so practically excepting Hutton TLP in the North Sea. The main reason is reliability on the strength of the tether. Therefore, most studies on TLP have related to the strength of the tether. The wave response of TLP and the tension of the tether were studied by Paulling, and Yoshida.
Since TLP has been said to be an aseismatic structure, the earthquake force has not been considered on designing as an external force to the tether. It is true that the horizontal component of the earthquake can be ignored because the flexural stiffness of the tether is nearly zero. However, the tension fluctuation of the vertical component of the earthquake is an important factor on designing. Studies on the earthquake load of TLP is only little described in Penzien's paper, but it has not yet been clarified. The authors have made for years theoretical and experimental investigations on the earthquake response of TLP.
In TLP motion analysis, the problem is usually simplified by idealizing the tether as simple springs. While this approximation appears to be generally adequate for shallow waters. The relative weight ratio (tether/platform) increases very rapidly as the water depth increases. Thus, it is necessary to treat the TLP under the earthquake as a continuous system rather than as a rigid body.
In this study, motion equations on a tension leg buoy (TLB), taking the tether weight into consideration, were prepared to clarify the earthquake response.

View full abstract

Human activity expanding in the field of ocean development demands larger and longer offshore structures. These large structures can easily deform under hydrodynamic forces and dynamic responses become serious problems even though the static strengths are enough. The stiffening of these structures needs large amount of material and is not feasible from the viewpoints of economy. Suppression of the dynamic response by the control technique is supposed to play an important role in the future ocean development. Furthermore the designs presupposing the control of dynamic response are expected to be realized.
As a preliminary trial to control the dynamic response of the offshore structures, two simple tower-like offshore structures are chosen and the dynamic responses in the regular waves are controlled experimentally in the wave tank. One is a simple rigid pipe type structure supported at the lower end by pin joint. The other is a beam type structure fixed at the lower end. The formulation of the optimal control shows that the control is divided into two parts. One is the closed-loop control where the force proportional to the errors between the present state variables and their target value is applied to the systems. The other is the open-loop control based on the definite wave forces. The spectrum of the ocean waves in the steady state is a function of wind velocity and the phase of element waves are decided by analyses of the surface elevation. So the wave forces acting on the offshore structures can be handled as the definite forces.
In this research, two types of open-loop control are employed apart from the optimization. One is the type to compensate the whole wave forces acting on the structure. The other is the type to compensate the major Fourier component of the fluctuating wave forces by using the three states of the thruster, forward, stop and backward. The displacement and deformation of the structures were measured by using ultrasonic measurement system. The surface elevation was measured by a capacity type wave height meter. These data were sampled and processed by 16 bits microprocesser and the thrust was realized by screw propeller type thruster.
The performance of the control was satisfactory. The response of the controlled structures was roughly 30% of the uncontrolled system.

View full abstract

A leg fixing device, which connects together the leg and the hull of a jackup rig, is indispensable when the jackup rig develops oil and gas in harsh environment such as the North Sea, and it takes severe load through its several teeth in contact with the several teeth of the opposite leg rack. Accordingly, it is very important to know the strain distribution, deformation and ultimate strength of one pair of mutually contacting opposite teeth, and the load distribution and ultimate strength of several pairs of mutually contacting opposite teeth from the design point of view.
In this paper, four kinds of scale models of leg fixing device, three were 1/4 scale (one pair of opposite teeth) and the rest 1/8 scale (six pairs of opposite teeth), were tested up to the ultimate load state, and the analysis of the models was performed by the finite element method as elasto-plastic contact problem. The results of calculations were compared with those of experiments and the following conclusions were obtained.
(1) Characteristics of strain field of one pair of opposite teeth from elastic state to ultimate strength state were shown by the calculation and the calculated equivalent plastic strain distribution showed that the mechanism of failure of one pair of opposite teeth derives from shear.
(2) Like the contact problem of circular cylinder and flat body, both calculations and experiments showed that the relation between load and displacement of one pair of opposite teeth was concave in elastic state and changed convex after elasto-plastic state started.
(3) Both calculations and experiments confirmed that the ultimate strength of one pair of opposite teeth was obtained by the simple equation proposed in Ref. 1).
(4) Simplified analysis method to estimate the load distribution of several pairs of opposite teeth was shown and found useful by the comparison between the results of calculation and experiment.
(5) Concerning n pairs of opposite teeth, the calculation showed that uneven distribution of tooth load in elastic state became almost even in ultimate strength state and both the calculation and experiment showed that the ultimate strength of n pairs of opposite teeth was n times the ultimate load of one pair of opposite teeth.

View full abstract

To prevent the catastrophic failure of a large storage tank, it is necessary to ensure not only that the long brittle crack will be arrested by the material toughness, but also that the arrested crack will subsequently be stable. After the long and fast running crack is arrested, the bulging force applied to the crack flanks will grow and re-initiate the unstable ductile crack.
R-curve concept combined with J-integral is adopted as a theoretical tool and wide plate tests are carried out to estimate the fracture resistance of a material for LPG storage tanks. The tested material is the 34 mm thickness 2 1/2% Ni steel plate manufactured by the special thermomechanical process, SHT-DAC. The specimens are 3 m in width and 4. 6 m in length so as to reproduce long arrested cracks in storage tanks. The specimens were loaded in the 10, 000 tonf test rig under the room temperature and as the ductile crack extended, they were unloaded in less than 10% for measuring the crack growth.
The applied driving force is calculated from J-integral, the Battelle's equation on the bulging effect.
The ductile resistance of the test material is 200 N/mm² as dR/da for long arrested cracks. For a large LPG storage tank of 40 m in diameter, the strake height should be limited under 3. 2 m to avoid the ductile instability of the long arrested crack.

View full abstract

It is well known that fracture toughness value is affected by strain rate. As for strain rate, the dependency of fracture toughness on fracture parameter rate has been usually studied. In the linear fracture mechanics, fracture parameter rate (dK/dt) is directly related to strain rate. However, in the non-linear fracture mechanics, the relation between fracture parameter rate (dδ/dt, dJ/dt) and strain rate has not yet been made clear. Therefore, in the safety assessment of real structures, non-linear fracture parameter can not be applied in case of which the effect of the strain rate on the fracture toughness needs to be considered.
In this paper, fracture initiation point is assumed to be the tip of the intensely deformed non-linear zone defined by Rice, which is considered to show the maximum triaxial constraint. Then the strain rate at the supposed fracture initiation point is given in connection with remote strain rate which corresponds to the strain rate in non-cracked plate. By use of the newly difined strain rate, strain rate-temperature parameter is determined, which represents the effect of strain rate on fracture toughness. Compact tension tests and COD tests with different strain rate and temperature, as well as K_Id tests are carried out to verify above conception.

View full abstract

In order to evaluate dynamic fracture toughness of structural steel, three point bending (3 PB-COD) or compact tension (CT) test is usually carried out by using drop weight or high tensile rate testing machine.
Since, in static test, applied load to the specimen can easily measured exactly by load cell of test system, but in dynamic test it is difficult to get or decide working load on test specimen because the effect of acceleration affects on the output of load cell in test system.
In this paper, the estimation of working load on test specimen was investigated applying a simple mass and spring model to the loading system of test equipment.

View full abstract

Constant loading tests and monotonic loading tests have been carried out at two temperature levels, i. e. room temperature and -70°C at which unstable brittle fracture occurs after stable fibrous crack growth, by using double edge-cracked tensile specimens and 3-point bending specimen of 4 mm thickness extracted from mild steel. Fibrous crack growth, time-dependent deformation and unstable fracture transition behaviors have been observed. According to above results, consideration has been conducted on the effect of loading mode on fibrous crack growth and unstable fracture transition.
Fibrous crack growth under constant loading is independent of kind of loading mode, i. e. tensile and bending. Though, near-tip strain distributions of tensile and bending specimens are different, equivalent plastic strain at 0.2 mm ahead of crack tip is the determined by the value of COD at the original crack tip. Fibrous crack growth and time-dependent deformation behaviors under constant loading at -70°C is the same as those at room temperature. Critical CODs, (δ_m) _d (δ_u) _d obtained by constant loading test are lower those obtained by monotonic loading, δ_m, δ_u. Particularly, (δ_u) _d becomes about half of δ_u, and it is almost the same as the COD δ_i at fibrous crack initiation.

View full abstract

A numerical method is developed to estimate the failure behaviour of ice, and the model tests are carried out to verify this method. This method employs the thick-walled shell element based on the concept of the Rigid Bodies-Spring Models. This method is equipped with some functions to deal with characteristics of ice, such as material nonlinearities, geometrical nonlinearities, differences of material properties through thickness, buoyancy and self weight effects, and cracking behaviour. The verification tests of this method are carried out in the IHI ice tank on both level ice sheets and composite ice sheets including model ice ridge, and it is shown that the results of simulations have good correspondence with experimental results.

View full abstract

The present paper proposes a new dynamic failure process simulation model considering the effect of matrix shear failure as well as random fiber breaks based on a shear-lag theory in which an additional time variable is also incorporated by taking the mass of fiber and matrix elements into account. An exact time-dependent stress redistribution process in a composite failure model is evaluated by means of a finite difference scheme based on an increment method for unidirectional carbon (C) and carbon (C) /glass (G) hybrid FRP models.
It is observed that the time-dependent stress concentration factor evaluated by the simulation is in good accordance with the Hedgepeth's analytical solution. It is also clarified that the stress in carbon fiber is relieved to some extent due to the presence of neighbouring glass fibers in hybrid C/G hybrid FRP models.
The stress wave propagation is successfully simulated by visualizing a time-dependent change in nodal displacements as well as pursuing a dynamic failure process of a composite failure model. These simulations are shown to be effective in characterizing different failure processes and stress wave propagation behaviors of unidirectional CFRP under varied higher strain rates.
Finally, a simple damping term is taken into account in the dynamic failure process simulations. It is shown that a clear difference exists in the dynamic failure process and the resulting tensile stress-strain diagram of CFRP and C/G hybrid FRP under varied higher strain rates.

View full abstract

There are instances where employing the preheat temperature for preventing root cracking as determined from single-pass welding cracking test results for the continuous multi-pass welding errs on the side of safety.
The authors devised a preheat temperature selection procedure and successfully developed a handy formula to prevent the root cracking in continuous multi-pass welding.
Parameters employed in the preheat temperature selection procedure and the formula thereby obtained are summarized as follows :
(1) The susceptibility to weld root cracking in continuous multi-pass welding is less than in single-pass welding, mainly because of decrease in hydrogen at the root due to post weld heating effect. This effect is expressed by thermal factor (ΣDδt) which is determined by thermal cycle.
(2) Nearly the same critical thermal factor [(ΣDδt) _cr] for root cracking applies to both continuous multi-pass welding and in single-pass welding. (ΣDδt) _cr is given by the following weldment cracking parameter (C_w)
(ΣDδt) _cr=7.94×10^-4C_w-0.123
C_w=0.32H_Vmax.+30.5 log H₀+F₁ × σ_w
where, F₁ × σ_w is a local stress parameter for root cracking. (3) Critical cooling time to 100°C [(t₁₀₀) _cr] for root cracking is given as :
(t₁₀₀) _cr=exp [1.22 {ln1/α (7.94×10^-4C_w-0.123) +9.79}]
α= 1 : Single-pass welding
2 : Continuous multi-pass welding

View full abstract

Marine casualities of tankers normally involve immeasurable risk of danger threatening the loss of human lives in the large scale of fire on the sea surface.
International Convention for the Safety of Life at Sea, '83, requires oil fire test for the actual lifeboat with sprayed water in the pool on the water surface.
But, the heat load from fire in the pool might be smaller than that of fire in the sea and it will be variable according to the wind velocity during test. Moreover, fire test might cause unjustifiable level of pollution and unreasonable cost of experiment.
Accordingly, it has become necessary to predict the relationship between the flow rate of sprayed water on the external surface of lifeboat and the temperature of lifeboat or inboard air without conducting fire test for the actual lifeboat.
This paper deals with some experiments of sprayed water, fire tests of lifeboat, and calculation of heat flow into lifeboat.
The main results are summarized as follows :
1) Spectral absorption coefficient of liquid water and absorptance of sprayed water were measured to know the effectiveness of sprayed water over the external surface of lifeboat in the fire. These data are also applicable to the planning of water cooling device on embarkation deck and to the decision of flow rate on the external surface of shell to protect the explosion of tank hold surrounded by fire.
2) Four kinds of fire tests were conducted in the pool for the actual lifeboat. The measured temperatures of inside surface of boat and inboard air were in good agreement with the temperatures calculated from the simulation model of heat flow into lifeboat.
3) The alternative method to evaluate fire resistance of lifeboat was proposed without conducting fire test for actual lifeboat.

View full abstract

The supplied quantity of heat energy for keeping in low viscosity of loading oil is necessary to be estimated with accuracy in a tanker carrying high pour point oil or high viscous oil. It is difficult to assume it exactly, because the heat transfer coefficient through tank bottom is various due to depending on the configulation of bottom structure and the physical properties of loading oil.
In order to investigate the heat transfer through tank bottom, the measurement of heat flux is carried out using the tank in the shape of bottom structure on the scale of one-third in which the paraffin as one kind of high pour point oil and the high viscous lubrication oil are heating up with three heaters. The predicting formulae on variable thickness of solidified oil and heat flow at bottom as time proceeds are derived from the theory and the experimental values. These formulae are applied for estimating the heat transfer coefficient of actual ship. As the results, the heat transfer coefficients through tank bottom excepting the effect of ship oscillation are assumed to be 78 W/m²K for high viscous oil and 4.7 W/m²K for high pour point oil.

View full abstract

The factors, which affect a durability of anti-corrosive coating system for offshore structure use, are analysed and quantified. Six factors are extracted as follows ; zone of exposure, tool for painting, number of coats, surface preparation, number of coating materials, and lamination structure. Twenty kinds of materials, which compose coating systems are chosen. An effective thickness of a coation system, normalized with equivalent thickness ratios of the coating materials, is supposed.
The effects of the factors and the equivalent thickness ratios on the durabilities are determined, based on the data at real environments, using the 1 st kind of quantification theory and the method of least squares.
The analytical results are shown and discussed. The estimated durabilities by the analysis and those at real environments correspond fairly good. Here, the relation between the durability and the effective thickness of the coating system is assumed to be linear.
It is concluded that the proposed effective thickness is nearly proportional to the serviceable period of the coating system.

View full abstract