Mechanism of the occurrence of wave breaking with vortex sheets is discussed by thinking the self-similar flow in strict consideration of the gravitational effect. Approximate realization of the self-similar flow with wave breaking was experimentally confirmed by using a ship model with triangular similar sections whose keel line varies as AX², where X is the longitudinal length. It is shown that the zonal breadth of wave breaking is proportional to ε X²/F² and it is related to the zone of the existence of longitudinal waves as a result of applying a linearized theory under the assumption that ε<<1 and F<<1, where ε is an appex angle and F is defined as U_∞√A/g. Examinations of local solutions under the nonlinear free surface conditions showed an existence of the singular solution with a vortex sheet on the production of folded wave surface into two. Its solution explains the mechanism of the initiation of wave breaking phenomena.

View full abstract

A finite-difference method named WISDAM-II is developed for viscous flow about a steadily advancing ship involving free-surface waves. The Navier-Stokes equations in a rotational form are solved by a time-marching scheme in a boundary-fitted coordinate system which is deformed so as to fit the moving free-surface. Making use of the subgrid-scale turbulence model, the three-dimensional separation and the effect of the free-surface waves on the viscous flow are simulated and discussed.

View full abstract

A method has been devised to assure positiveness of k and ε in the computation of the k-ε turbulence model. That is to express k and ε as exponentials of newly defined variables m and n.
The validity of the present scheme is demonstrated by computing two-dimensional channel flows starting with constant initial flow and reversed initial flow. Fully time-implicit three-point wall function method is used as the solid wall boundary condition.

View full abstract

A method for calculating effective wake distribution is developed for a three-dimensional hull form. In the present study the momentum integral equations of the boundary layer and wake are solved by an iterative procedure taking the interaction between propeller and ship wake into account. The velocity field with an operating propeller is divided into effective velocity field and propeller induced velocity field, and the boundary layer approximation is assumed to be applicable to the effective velocity field while the propeller induced velocity field is assumed to be inviscid. To examine the validity of the method the flow measurements were made for a simple hull form, i. e. Wigley's parabolic hull form model. It was found that the method can predict fairly well the velocity field with and without a propeller. The calculated results show that the degree of change from the nominal wake distribution into the effective wake distribution is large in the region where the propeller induced velocity is strong and the velocity gradient of the nominal velocity field is steep.

View full abstract

In the 1st and 2nd reports, direct formulation of boundary element method (BEM) based on the thick wing theory to the analysis of three-dimensional flow around the marine propeller in uniform flow and in steady ship's wake has been successfully presented.
In the present paper, the accuracy of BEM analysis to the three-dimensional flow around marine propeller is examined by the pressure distributions measurement of six kinds of model propeller blade. After comparison BEM calculated pressure distributions with measured ones, it is verified that the BEM analysis by direct formulation to the three dimensional flow around marine propeller is very usefull.

View full abstract

In recent years, a highly advanced lifting surface method with high accuracy has been required in order to design the various types of propellers which demand high efficiency and to reduce ship vibration and noise.
In this paper, an improved lifting surface procedure based on the vortex lattice method is presented. For the determination of a trailing vortex wake geometory, a numerical iterative procedure is described. The numerical results on propeller slipstream give a good agreement with the measured values by laser doppler velocimeter.
The present method is applied to calculate the open water characteristics of a propeller. And a quasi-steady technique based on the present method is adopted to calculate the fluctuation of propeller forces and cavitation patterns behind the ship's wake field.
The advantages of the present method are shown by comparing the numerical results with the experimental data of conventional and highly skewed propeller.

View full abstract

In this paper, from the viewpoint of accurate consideration of unsteady cavitation at the propeller initial design stage, a method to estimate vibration (surface force), noise and erosion, is described.
For this study, systematic cavitation tests were conducted in four different wake fields using four kinds of propellers, which have different blade section profiles (reported in the previous paper) and blade areas. The primary results obtained from the tests are as follows;
(1) Surface force is strongly influenced by the gradient of wake fraction in the circumferential direction.
(2) Noise, and erosion are closely related to the occurance of cloud cavitation, which is considered to be connected with the amount of cavity volume.
(3) In order to reduce surface force, noise and erosion, the improvement of blade section profile is more effective than an increase in blade area.
Based on the above experiments, a practical estimation formula of cavity volume is proposed. In the formula, the attack angle variation of corresponding two-dimensional wing, derived from the unsteady lifting surface theory, is introduced. And cavity volume is assumed to be in proportion to the modified attack angle, in which cavitation unsteadiness is considered in a simplified manner. Using the estimated cavity volume, propeller induced fluctuating pressure is calculated by Holtrop's formula. And it is found that the calculated values of fluctuating pressure agree well with the experiments.
Then, for the convenience of applying the results to the propeller design process, two simple parameters are proposed based on following ideas;
(1) Fluctuating pressure is estimated by the product of maximum cavity volume and the maximum second differential coefficient of attack angle at each radius position.
(2) Noise, and erosion are estimated by the product of maximum cavity volume and the maximum first differential coefficient of attack angle at each radius position.
Experimental results can be explained well by these newly proposed design parameters.

View full abstract

The structure of flow around unsteady cloud cavitation on a steady two-dimensional hydro-foil was investigated experimentally, using a conditional sampling technique. The unsteady velocity around cloud cavitation was measured by a LDV and matched with the unsteady cavitation appearance taken by a high speed camera. This matching procedure was performed, using data of pressure fluctuation measurement on the foil surface. The velocity fluctuations could be divided into a large-scale (low frequency) component and small-scale (high frequency) one, using Fourier analysis. The unsteady cloud cavitation corresponds with the large-scale structure.
In this manner, both large-scale (coherent) and small-scale (incoherent) structure of the flow around the cloud cavitation were clarified and discussed respectively. Main results are given as follows;
(1) Cloud cavitation observed at the present research is a kind of large-scale vortex cavitation whose rotation direction is clockwise when main flow direction is from left to right. The velocity of cloud cavitation is much lower than the uniform flow velocity.
(2) The small-scale velocity fluctuation is not distributed uniformly in the large-scale structure, but concentrated near the edge of it. One of the reasons might be that cloud bubbles collapse there.
(3) Three-dimensional structure of cloud cavitation is a stretched vortex whose configuration looks like a “croissant”.

View full abstract

This paper presents results of free-running model tests concerning ship maneuvering motions due to assist tugboats and simulation calculations based on the mathematical model proposed in this paper. The free-running model tests were carried out by the use of wind fans on the model deck which simulate the tug forces. The mathematical model for such maneuvering motions was developed on the basis of captive model tests which consist of the static drift test with whole range of drift angle and the yaw rotating test with several combinations of towing speed and yaw rotating rate. The simulation calculations were compared with the free-running test results for typical maneuvers with tugboats. Through the present study, the following conclusions are obtained.
(1) The free-running model tests clarify the basic characteristics of ship maneuvering motions due to tug forces and indicate that such maneuvering motions largely depend on ship forward speed.
(2) The simulation calculations show satisfactory agreement with the free-running test results. This indicates that the mathematical model proposed in the present paper is available for describing a so-called low speed maneuvering motion dealt in this study.

View full abstract

In the present report, an attempt is done to develop a computer aided maneuvering system in harbours. The system contains two functions. The first is to make a plan to lead a ship to a berth and the second is to carry out the plan. To make the plan, an approaching pattern is selected and it is built up as a combination of elemental maneuvering patterns. The trajectory from the initial state to the final one is designed.
To assure maneuvering safety, following two concepts are employed ; a false goal and margins of control forces. The first means that a ship approaches to the false goal placed in front of a berth, instead of approaching to the real one. The latter is to design the trajectory with enough margins of control forces.
To lead a ship to the berth, both of feedback and feedforward controls are utilized along the designed trajectory. A simplified mathematical model of maneuvering motions are used to determine control forces. The control forces are supplied in quantized forms by tugs.
A simulation result is finally reported to show a successful approach along the idea proposed in the present study.

View full abstract

A guidance and control system for an autonomous vehicle of a cruising type with a rudder and elevator is studied, which is intended to investigate the surface of ocean ground continuously. In order to collect underwater information continuously taking pictures or measuring chemical characteristics, it is required to keep low altitude from the surface on the condition that it keeps the safety distance.
In the paper, a simplified altitude control system to keep predetermined height is investigated observing the distances to the surface by several narrow ultra sonic fan beams in different directions. The control system is consist of two hierarchical control blocks, i. e. a guidance block to give attitude reference angle calculated from the observed distances to keep desirable altitude, and a local attitude control block to realize the reference angle keeping the stability of the vehicle. As the attitude control system, a linear feedback elevator angle control system with velocity saturation is adopted, taking account of the system dynamics.
In the guidance block, pitch correction angle is calculated from the observed distances. Three different algorithms, linear, polynomial filtering, and mixed type, are proposed. And the system performance of each is evaluated from various surface conditions by simulation study using simplified two dimensional model in vertical (pitch) plane. Non-symmetric smoothing of the correction angle is also introduced to keep the safety distance for rough surfaces. From the study, it is assured that the proposed control system can control the vehicle to keep the predetermined altitude for most cases within 20% margin, and basic feasibility of the system is verified.

View full abstract

It is important to predict the dynamic behavior of a fishing boat moored in harbor. Because, the extraordinary mooring line forces and ship motions cause damage e. g. break of mooring lines or collision with a quay. These dynamic behavior differs from that of a large ship in harbor or a moored offshore structure. That is, a fishing boat is usually moored perpendicular to a straight quay in longitudinal waves. Further, her mooring lines have completely non-linear characteristics.
Then the authors carried out the numerical simulation in time domain ; in which wave forces and mooring forces were evaluated by theories and viscous damping coefficients were derived from free decay tests. We got the stochastic values of surging motion and mooring line forces in irregular waves.
Moreover, we conducted model experiments in a tank. A model ship was moored in front of a model quay. Its motions and mooring line forces in irregular waves were measured by means of an optical tracking system and ring-type strain-gauge transducers, respectively.
As a result, the values computed by our numerical simulation procedure were in good agreement with the experimental values. Then it is shown that our proposed simulation procedure is effective to predict the safety of a fishing boat moored in harbor. In particular, we found that wave exciting surge force can be estimated by Froude-Krylov hypothesis and mooring forces can be estimated by the static mooring characteristics and the surging motion of a ship. Furthermore, we discussed the effect of waves, mooring systems and quay structures on the dynamic behavior of a fishing boat.

View full abstract

In order to solve nonlinear hydrodynamic problems of a ship with forward speed, a nonlinear calculation method of two-dimensional hydrodynamic problems is developed. The free surface boundary conditions are linearized, but an exact body boundary condition permits large amplitude motion of the body. Appling the multi pole expansion method, a calculation procedure which gives an exact solution of the boundary problem mentioned above is developed. Ignoring the effect of the free surface on the bi-harmonic component of the velocity potential, an approximate calculation method is proposed. Calculated results are compared with measured ones. The approximate method gives fairly good estimations.

View full abstract

Hydrodynamic forces acting on bodies with rectangular and circular sections are measured and are found to exist non-linear forces.
Radiation forces acting on bodies forced to sway in simple and in combined two harmonic oscillations in low and high frequencies are measured. Hydrodynamic coefficients obtained from the forced sway combined two harmonic oscillations differ from those of the simple harmonics.
Wave diffraction forces acting on bodies oscillated in low frequencies are also measured and are found to be different from those acting on still bodies.
These differences are considered as non-linear hydrodynamic forces and have good correlations to the increases in the modified Keulegan-Carpenter number proposed in this paper for the case of combined oscillations.

View full abstract

Hydrodynamic forces acting on rough circular cylinders in a harmonically oscillating flow at high Reynolds numbers were experimentally investigated.
Especially effects of the roughness height and roughness density were investigated. Two kinds of roughness height-5mm, 10mm and three kinds of roughness density-20, 60, 80% were used in experiments.
Experiments were carried out using a large circular cylinder with 1.2m diameter and with 1.5m length at NKK's Tsu Ship Model Basin. The cylinder was fixed to the main carriage. The cylinder was forced to oscillate harmonically with the main carriage. In-line and lift forces acting on cylinders were measured at Keulegan-Carpenter numbers over 6 up to 50 and at two Reynolds numbers-0.5, 1.0×10⁶. Drag coefficients, added mass coefficients, maximum lift coefficients and periods of lift force are obtained.
Experiments show that the roughness has a great effect on the hydrodynamic coefficients even at high Reynolds numbers. In the range of the value of roughness parameters used in these experiments, a roughness height has a more considerable effect on the hydrodynamic coefficients than a roughness density.

View full abstract

Due to the lack of facilities, only few experimental works have been carried out in the field of motions of a floating body in multi-directional waves. Therefore, the authors carried out experiments on motions of a floating body in two directional waves from rectangle each other in a model basin and they checked the linearity of waves and motions of a floating body.
From these investigations, the authors pointed out the limitation of linearity and the interesting phenomena of slow drift oscillation of a moored floating body in two directional waves. They also developed the experimental technique of motions of a floating body in multi-directional waves and derived the theoretical prediction method of these motions in multi-directional waves which is based on functional polynomial method.

View full abstract

In the present paper, the powerful numerical method for a nonlinear wave forces acting on the plural vertical cylinder with arbitrary cross-sections was proposed.
According to the present method, a second-order wave force can be obtained from a linear radiation potential without solving second-order boundary value problem. The boundary value problems for a linear radiation and diffraction potentials are solved with the hybrid boundary element method. The two method of numerical integration for a infinite integral on a free surface were showed and examined. One is the approximate integral method, in which infinite region is replaced by finite region and another is the numerical integral method with the Laguerre integration.
The numerical results of maximum and second-order wave forces for the circular and rectangular cylinders obtained by the present method were compared with the experimental results. The property of the two numerical integration and the characteristics of the nonlinear wave forces were discussed.

View full abstract

A series of model tests in a regular wave tank have been conducted to validate the theoretical solutions for the first order and mean drift responses of an articulated column, which have been developed in the first report. Measured results for the first order wave-induced hydrodynamic forces and tilt response as well as for the mean drift overturning moment on a model column are presented in this second report, and compared with theoretical predictions based on the potential flow theory. Agreement between the experimental data and the theoretical predictions is found to be satisfactory over a range of frequencies if the effects due to reflections from the tank side walls are taken into account in the theory. These results seem to suggest the validity of the potential flow theory for estimating the first order and mean drift responses of the articulated column in open sea.

View full abstract

It was well known that slowly varying phenomena in irregular motions of a moored semi-submersible rig could not be neglected in irregular waves or in actual ocean waves. On the other hand, the structural responses of members of offshore structures were usually treated only by the linear response functions owing to the wave frequency.
This paper shows at first the clear experimental results including the slowly varying structural response which are obtained by the model experiments of a standard semi-submersible rig with chain mooring in irregular waves.
And the theoretical linear transfer functions of the motions and structural responses are compared with the measured values, then usefulness of these practical calculation methods is confirmed.
It is investigated that the structural responses with very low frequency are caused by the slowly varying motions and can be approximately estimated by the application of the free oscillation tests results in still water.
Furthermore, the statistical properties of the irregular time histories with the slowly varying phenomena are clarified, and the statistical predictions of the motions and structural responses with the slowly varying phenomena become practical.

View full abstract

Japan Marine Science and Technology Center plans the TLP sea test using a prototype floating offshore structure which has been designed by Ship Research Institute, Japan Ministry of Transport, for the joint at-sea experiments.
The TLP sea tests will be excuted in this and next years. The test will be performed to evaluate the performance and reliability of the tension mooring system in the real ocean environment. The site is 40m deep and 3km offshore facing the Japan Sea. The environmental condition is favorable in summer, which allows us very easy installation work. It becomes, however, severe in winter. The prototype floating ocean structure is not originally designed for TLP. Moreover this test structure is made as small as the motions of the structure and the mooring line tensions can be measured in relatively severe sea conditions upon keeping enough stability.
The air draft is only 5.5m. Consequently the great waves will crash the upper structure causing shock loads in the tension mooring line. By the 1/40 scale model test the critical wave heigh is proved to be 7.2m. Therefore, the test period during which the sea test can be excuted safely is decided from May till October. The stress check for yielding and buckling is proved that the working stresses under the critical environmental conditions are much less than the allowable stresses.

View full abstract

Three-dimensional wave deformation due to a horizontally submerged thin plate is analyzed by the Doublet Distribution Method (DDM). The plate is assumed to be located at a certain depth from sea surface in deep water. The linear wave diffraction theory is employed in this method.
First of all, the influence of mesh refinement in the present analysis to the numerical results is checked by employing a submerged square plate model. Next, as a particular application example, a wave focusing phenomenon by a submerged convex-shaped plate model is examined by the present method. In addition, some hydraulic model tests are performed to verify the numerical results.
The following findings and conclusions are obtained in the present investigation.
(1) The present DDM can solve effectively the three-dimensional wave deformation phenomenon due to a horizontally submerged thin plate. Element size is the most important factor in the present analysis in order to obtain a good numerical result. It is found that the element size should satisfy the following criteria : B/b>7 as well as λ/b>7, where the symbols b, B and λ denote an element size, a chracteristic length of the submerged plate and an incident wave length, respectively.
(2) A wave focusing and the corresponding wave amplifying phenomena due to a submerged convex-shaped plate can be identified in both theoretical and experimental results. In almost all cases, a significant wave amplification phenomenon can be observed on the surface of plate. The maximum amplification factor (the ratio between the observed and incident wave heights) obtained has the value of about 3.0 at the edge of the plate. It is also observed that the wave amplified zone is widened in the shore-side area behind the plate as the wave period becomes shorter and the plate is submerged in shallower water. As a result, it can be stated that the submerged plate model studied herein would be an important structual element for controlling the natural wave field in the future, for example, for extracting electric energy from wave motion.

View full abstract

In the preceding report, we have dealt with underwater sound radiated from a semi-submerged simple circular shell, and shown the basic role of free water surface. However these results are all about the simple circular cylinder so that they might not be applied to actual ships.
This report deals with this point, namely, to know the influence of the hull section shape on the property of underwater radiated sound, we show the numerical analysis method of the mutual coupled vibration of fluid-hull interaction, in which the vibrations of hull are solved by F. E. M. and the sound into fluid by B. E. M..
At first, we calculate the semi-submerged simple circular shell by this method, compare with the preceding report results, and find a good agreement between them. Next, we show the effect of changing drafts of a circular shell and the effect of hull section shapes. Finally, we calculate the realistic cross section of ships and show the effect of the plate thickness and rigidity on underwater radiated sound.

View full abstract

Longitudinal ship motions have been investigated by means of the linear strip theories such as O. S. M. etc. On the other hand, the nonlinear simulation programs for longitudinal ship motion and strength have been developed for the purpose of analysis of slamming and whipping phenomena caused by hydrodynamic impact forces, and they have been successfully applied for several actual disasters to investigate real process of the disasters. In case of rough seas that may cause ship disasters, surging motions become significant, which have been experienced in model experiments and actual ship navigation. Here a difficulty arises; the effects of surging motions have been disregarded in the previous investigations of longitudinal motion and strength.
In the present paper, surging motions of a ship among elevated waves are determined step-by-step by considering instantaneous ship configurations and Froude-Kriloff forces acting on the ship; it is performed by the use of a revised version of the nonlinear simulation program TSLAM for longitudinal ship motion and strength, since the original version is for a ship advancing with uniform speed. As results of the investigation, the increase of thrust can be calculated and compared with the previous experiments, which shows good coincidence. For the longitudinal strength, however, no significant effects appear in the present example calculations.

View full abstract

In ship structure design, estimations of wave loads and structural response to them should be made in relation to sea condition. Since the estimation of wave loads is usually carried out component-wise, accuracy of estimated response must be examined by checking not only magnitudes of responses to component wave loads but also phase differences between them.
From this point of view, in this paper, confirmation was made first on the accuracy of wave load estimation calculation by comparing with results of model experiment, focussing on phase differences between wave load components. Then, DISCRETE ANALYSIS METHOD based on full ship 3-D FEM was developed, and its reliability was confirmed. Next, trial calculations of structural response to superposed wave load components, were made on a 63, 000 DWT bulk carrier. From this calculations, the following results are obtained:
(1) Stress components on double bottom induced by inner and outer pressure mostly cancel each other.
(2) Resultant stress on double bottom due to combination of hull girder bending and local bending is lower than the stress value obtained by simple summation of stress components, owing to the effect of phase difference.

View full abstract

Car ferries, pure car carriers and container ships may suffer from slamming on the stern flare because those ships require broad deck plane and the bottom of near the aft part becomes flat. In the present paper, a response of a container ship to stern flare slamming is studied using the numerical method which authors developed to analyze the dynamic response of ship hulls to the impulsive hydrodynamic force.
The calculations are performed for the ship in following seas because an environment condition data suggests that a ship is operated in following seas as frequently as in head seas. Results obtained are as follows :
1) When the wave length is equal to the ship length or shorter, there is a possibility of occurrence of severe stern flare slamming.
2) It is recommended from a viewpoint of the strength that stern bottoms should have some degree of dead rise angle. Hydrodynamic impact force is not observed in the calculated result for the ship the stern bottom of which has 10° of dead rise angle.
3) 4- and 5-node vibration have significant influence to the structural response of an aft part of ship as well as 2- and 3-node modes.

View full abstract

The effects of various design parameters on the strength of double hulled structures in collision were examined. At first, the scantlings and the strength characteristics of double hulls were investigated, which included double sides and double bottoms. Static destruction tests were carried out using a series of double hulled models which covered a wide strength distribution range. The validity of the simplified method to analyse the strength of double hulled structures in collision which was presented in the previous paper was confirmed by comparing the calculated results and the experimental results. Finally, choosing a double hulled ship of 100, 000 tons displacement as a standard type ship, parametric calculations were done using this method, where the side shell thickness, the transverse web thickness and the double hull breadth were taken as design parameters. It was found that the most effective method to increase the energy absorption capacity is to increase the transverse web thickness.

View full abstract

Primary objective of this report is to provide a fundamental strategy to analyze dynamic response of offshore structures under collision. The key to planning meaningful experiments or numerical analyses is a classification of the given problem based on the overall knowledge of the phenomena. The authors proposed a method in which offshore structures are idealized as spring-mass systems. The characteristics of the collision problems were studied to get the overall knowledge using this method. Such knowledge can be utilized to classify given problems and also to decide how to tackle them. Following the above strategy, collision problems of single pipe and an elastically supported pipe were analyzed using beam and shell Finite Element Methods as examples.

View full abstract

The purpose of this report is to describe the experimental results of dynamic lateral buckling behavior of T-shaped beams against impulsive load. Thin-walled beams are tested systematically in a pendulum type impact testing machine by changing geometrical parameters of specimens as well as loading conditions. Especially, we will study about dynamic loads and energy consumption, which are compared with static results.
The present paper will also describe FEM model incorporating effects of increase of yield stress due to high strain rate in order to improve its capability for impact analysis. Based on numerical results by the present model, we will investigate experimental behaviors of the second lateral buckling mode as well as the first mode which cannot be conducted in the present testing machine.

View full abstract

The non-axisymmetric crush tests were conducted for 11 mild steel cylinders with different R/t ratios, and the obtained results were compared with the theoretical, numerical and other experimental results from various points of view. The conclusions can be summarized as follows :
(1) The plastic buckling loads due to Gerard's theory agree well with the maximum loads obtained in the present experiments.
(2) As for the relation between the radius-to-thickness ratio and the circumferential wavenumber in the non-axisymmetric crushing mode, the validity of the existing solutions given by one of authors was confirmed.
(3) Magee's empirical formula for the mean crushing strength is in good agreement with the present expermental results.
(4) The nonlinear finite element code developed in the previous report gave improved solutions for the mean crushing stresses in comparison with the existing rigid-plastic solutions, which are in good agreement with the experimental results except for the case of relatively thick-walled cylinders.
(5) The actual complicated crushing processes accompanied with traveling hinge lines were successfully simulated by the present finite element code.

View full abstract

Previously, the authors developed the plastic collocation method for elastic-plastic analyses of homogeneous continuum in any geometrical shape. In this method, using ordinary finite elements, plastification is examined at the checking points for plasticity properly selected in the element. Regarding the yield conditions at the checking points as plastic potentials, and applying the plastic flow theory, the elastic-plastic stiffness matrices can be derived without integration over the element. Moreover, suitable selection of the checking points considering characteristics of the stress distributions in the element leads to improvement of the accuracy of the collapse loads.
In this paper, expanding the basic theory of the plastic collocation method, a general theory for the analyses considering strain-hardening effects is proposed. In the plastic collocation method, plastic deformations are concentrated only to the nodes. Then, to take into account of strain-hardening effects, it becomes necessary to evaluate the spread of plastic region in the element and condense strain-hardening effects therein to be represented as the strain-hardening rate with respect to the plastic nodal displacements. Applying this expanded theory, several examples are analysed, and the effectiveness of this method is demonstrated.

View full abstract

In the first report, the nonlinear behavior of perfect plates and perfect stiffened plates (without initial imperfections) subjected to inplane loads were studied and idealized. Basic versions of two idealized units (elements) : the “rectangular plate unit” and the “stiffened plate unit”, have been developed.
In this report, the capabilities of these idealized elements are extended so that they can be accurately applied to a wider range of actual structures. Effects of initial imperfections, i. e. initial deflection and welding residual stress, on the behavior of these elements are studied, idealized and included in the formulation. Lateral load is also considered and its interaction with initial imperfections and inplane load is taken into account.
Assessment of accuracy of these elements is carried out through comparisons with experimental and other theoretical results. An example of analysis on a relatively large structure is also presented to demonstrate the capabilities and efficiency of these elements.

View full abstract

Using an open type wind generator, two kinds of tests on TLP models were carried out. One was the wind tunnel tests for measuring both wind load and coefficients and effects of interaction of elements on the coefficients. The other was the test for measuring response of a model with legs installed in the seakeeping basin, where wind and waves were applied both individually and simultaneously.
On the other hand, the nonlinear analysis in a time domain was carried out. Based on the comparison of the measured responses with the calculated ones using the experimental coefficients, some foundamental considerations are discussed on the following subjects : possibility of evaluation of wind loads by summing up element loads, feasibility of nonlinear time domain analysis, possibility of superposition of wave and wind loading and so on.

View full abstract

Fundamental structural response characteristics of a typical semisubmersible in waves are investigated both by model tests and theoretical calculations.
A 1/50-scale structural dynamic model of 2-lowerhull and 8-column type was made of acryl and the wave tank tests were carefully carried out. The platform motion, mooring tension and the axial and bending stress of the braces and deck girders were measured in detail.
The measured wave-induced responses were compared with three-dimensional motion and frame structural calculations. In this paper stress is put on the axial force responses of major structural members. Structural response characteristics are also studied through simplified estimation of wave loads.
As a result, the experimental responses agree with the calculated ones fairly well except for a particular brace, and the fundamental structural dynamic response characteristics of this type of semisubmersibles are confirmed both experimentally and theoretically. The detailed experimental results presented in this paper are expected to be valuable for the study and design of similar types of semisubmersibles.

View full abstract

The local buckling of tubular members after ultimate strength is attained was investigated from both experimental and theoretical viewpoints. First, axial compression tests were carried out on three tubular specimens similar to the bracing member in an existing semisubmersible drilling unit. It was known that :
(1) After the ultimate strength was attained, local buckling of a cosine mode first takes place at the compression side of bending. The buckling wave spreads about one half circle in the circumferential direction, but has short wave length in the axial direction.
(2) At the same time, load carrying capacity suddenly decreases.
(3) Then, the formations of local denting deformation follow at the foot of initial cosine buckling wave.
A simplified method was proposed to analyze the elastic-plastic behaviour of a tubular member under axial compression. Two models were newly proposed which simulate the post-local buckling behaviour of a tubular member based on the observed results in the experiments. They are COS model and DENT model.
Combining these models with the proposed simplified method, a series of analysis was performed on the test specimens newly carried out and previously done. The results of calculations were compared with the experimental ones, and the validity and the usefulness of the proposed method of analysis were demonstrated.

View full abstract

The method of practical fatigue strength analysis for the hot-spot parts around leg nodes of the Jack-up rigs is seemed to be behind the demand because of their unique operating systems and complicated mechanism of axial force occurrence in braces. In order to estimate the fatigue strength at leg nodes, the forced vibration analysis for irregular waves in short and long-term sea conditions is mandatory.
In this paper, the method of the forced vibration analysis in the regular wave is developed for the irregular waves and the method to estimate the long-term distributions of axial force amplitude in braces is introduced by applying the simple formula in the previous paper. For the estimation, the probability of occurrence of wave incident angle, operating sea depths and whether the wedges are installed or not are considered.
The fatigue strength at nodes can be estimated by the combinations of long-term distribution of hot-spot stress amplitudes calculated from axial force in braces and the S-N curve for the used material.

View full abstract

In structural analysis of frames by the efficient and accurate method (Idealized Structural Unit Method), members are usually assumed to be rigidly connected to each other at nodal points. However, in tubular frames with simple (unstiffened) joints, the joints may show considerable flexibility in the elastic as well as the elastic-plastic ranges. Such flexibility may cause excessive deflections and different internal force distribution in the structure. Increased deflections are of great concern since their effects may be great and important in the collapsing behavior of a structure.
In this paper, the authors carried out theoretical investigations into the local rigidity and strength of tubular joints (T, Y, TY and K joints). Based on the results, they idealized the behavior and developed an effective simple model of tubular joints, which is constructed with spring, rigid bar and beam elements. This model is named an idealized “Tubular joint element”. The model exhibits joint flexibility in the elastic range as well as the elastic-plastic range and the limit strength, following the elastic fully-plastic load-displacement relationships which are assumed based on the results obtained either by theoretical analysis or experiment. Under several typical loading conditions, the behavior of the model was examined, being compared with the results of analyses by the finite element method (the plastic node method). The result confirmed the accuracy and assured the validity of the model.

View full abstract

The ultimate strength is one of very important criteria for safety assessment of offshore structures. In this report, the ultimate strength of jack-up rigs is evaluated by the “Idealized Structural Unit Method”. Two loading conditions are taken into account, i. e. survival and punch-through conditions, which may lead to total collapse of the rigs and, then, are very worthy of investigation.
The investigation draws the following conclusions.
(1) Two failure mechanisms are distinguished, i. e. yielding of chord which occur usually with fixed type jacking unit, and buckling or yielding of braces which occures usually with floating type jacking unit.
(2) In case of jack-up rigs investigated in this study, the max. wave force in survival condition is 1.6-1.8 times of design wave load and allowable penetration in punch-through condition is 4-5 m. This is believed to be typical for jack-up rigs. However, the present design may not be regarded as over conservative considering the many uncertainties involved.
(3) In 3 legged jack-up rigs, failure of one leg would lead to total collapse of the rig. No redundancy is provided by the other two legs.
Based on these conclusions, a simple analytical model is developed and simplified formulae are proposed to estimate the ultimate strength of jack-up rigs in the above mentioned two loading conditions depending upon the failure mechanism and the type of jacking unit.

View full abstract

Recently oil drilling activity tends to go towards deeper water and harsher environments, which requires severer design conditions for oil drilling rigs. For jack up rigs, one of the typical type of oil drilling rigs, the above trend requires enhanced capability, such as stability, soil bearing capacity, leg strength and leg holding capacity. The former three problems can be solved by proper configuration of rig main hull, legs and footings, however, it is difficult and uneconomical to obtain the larger holding capacity by enlarging jacking system or leg holding system independently.
In this respect the jacking unit newly developed by Sumitomo Heavy Industries, Ltd. can integrate jacking system, automatic torque balancer and leg holding system. The unit is developed in combination with Sumitomo standard jack up rig, SU 300 H, for 300 feet water depth and hostile environment use.
Jacking system is of electric driven rack and pinion type, rated 320 tonf per pinion and composed of 4 pinions. As a sophisticated high ratio planetary reduction gear is adopted for first reduction stage, no oil bath lubrication is required for second and third reduction stages, which minimize the maintenance work.
Automatic torque balancer eliminating the dangerous manual torgue adjustment of jacking motor, which is inevitable in conventional units, can equalize the bearing loads among all pinions of the jacking system. The balancer can also adjust the load distribution among the pinions and stoppers when both work simultaneously under the hostile sea condition.
Leg holding system is composed of 8 stoppers, each independently driven by electric motor with constant torque, which can compensate the unfairness of chord racks and the relative displacement between racks and stoppers. This system is furnished above the jacking system and functions solely or simultaneously with the jacking system.
Consolidated leg holding capacity is 6000 tonf per unit when jacking system and leg holding system are simultaneously operated, which leads to enormous leg holding capacity for its compactness. A 6000 tonf capacity comes from the uneven load distribution among pinions and stoppers.
Theoretical analysis was carried out to study the behavior of captioned jacking unit under the design load conditions, which showed fairly good agreement with the experimental results.
A 1/2.5 scale experimental model weighing about 20 tonf was manufactured, tested in every aspect by the 3000 tonf load test rig, and the test was found satisfactory with the obtained results, such as elevating capacity, automatic torque balance and leg holding capacity. The load test was focused on investigating the mechanical and structural characteristics, especially on the load distribution among 4 pinions and 8 stoppers under the towing condition with full length legs raised and operating condition.

View full abstract

To the arctic off-shore structures under huge ice load, it is very interested to apply the composite steel-concrete beams, where concrete is filled between steel plates, as the exterior walls of the structures.
In this paper, the authers carried out static loading test to investigate the strength of the composite steel-concrete beam in the various cases, they are load distribution, span length, size of supporting plate, beam end condition, shear connector and web plate.
The main conclusions are as follows;
(1) The strength of the beam under a symmetrical load to mid-span is greater than that of symmetrical load.
(2) The length of supporting plate placed under the beam have considerable influence on the strength and deflection of the beam.
(3) Where the ratio of span to depth of the beam is 2.0, influences upon the beam strength by end condition are negligible and decrease of the ultimate strength in the case of no shear connectors or no web plates is not found.
(4) The calculated ultimate strength by Niwa's formula for reinforced concrete deep beam is in good agreement with experimental ones.

View full abstract

Few hydrodynamic vibration damper have been developed. As far as the authors know, one is the U-type hydrodynamic vibration damper developed by the NPL of England, and the other is the spring and hydrodynamic damper devised by the authors. The former is the prototype damper for the actual war ship, and the latter is model. Recently Bauer showed that rectangular container filled with two immiscible liquids is very available for reducing vibration.
This paper deals with experimental and theoretical investigation into a method to reduce vibration with the sloshing in a tank. The authors carried out exciter test of the off-shore structure model with the rectangular tank filled partially with water, and also calculated damper effect of the sloshing with the theory, in which the sloshing is simplified into mass-spring systems. The authors obtained the conclusion that the sloshing works as dynamic damper for reduction of vibration and the above-mentioned theory, which was developed for estimating dynamic strength of a tank on land, gives coupled natural frequency in good agreement with measured one.

View full abstract

In general the design of deck panels is performed to be tuned of the lowest natural frequency of panels over frequencies in exciting forces. Recent passenger ships or some types of research ships with wide area for more accomodations, however, result in the occurence of lowered frequencies of panels. This means that it is very difficult to keep the lowest frequency of panels over exciting frequencies, and in such a case, it is necessary to verify that the higher modes of panels do not resonate with the exciting forces and/or those resonant amplitudes are small allowably.
This paper presents a simplified method to calculate the natural frequencies and the responses of deck panels, and also investigates the vibration of main hull girder as an excitation source, considering the effect of long superstructure. Through some vibration tests on actual ships to obtain several data required for correlating the calculation with the design, it is found that the deckcovering materials have very important roles on estimating the damping of panels.
Finally a design-based simplified technique is proposed to predict the characteristics for the vibration of deck panels, together with several recommendations related to the anti-vibration design of deck panels.
It is concluded that it is very important and effective to sustain higher mode frequencies of panels apart from the lowest one, for example, by the panels stiffened in two perpendicular directions and to avoid the resonance in the lowest vibration mode, from the fact that response amplitudes of higher modes are much smaller than that of the lowest frequency.

View full abstract

The Moss Rosenberg Verft spherical tank type L. N. G carrier was built in M. H. I s Nagasaki shipyard recently.
She has tall superstructure in the after body. It was considerd necessary to control vibration level, and vibrations study has been performed by modal analysis and Finite Element Method.
The exciter test and measurement at sea trial has been carried out for confirmation on completion of the ship, and following items were confirmed
(a) Present calculation method is useful for estimation of dynamic characteristic of ship.
(b) Vibration level of L. N. G. C was controlled satisfactorily as we approched and expected.

View full abstract

It is known that when buckling accident occurs, the buckled part easily separates. In this connection, the deteriorating effects of extremely high compressive prestrain on ductility of mild steel were discussed in the present paper.
30mm thick mild steel plates were compressed uniformaly to a few levels of prestrain up to -35%. After prestraining, 2mm V-notch Charpy specimens and compact tension specimens were machined, and tested for evaluating the criteria of Charpy test and fracture toughness. In order to attain to higher compressive prestrain, mild steel plates were deformed by bending to several levels of prestrain up to about -80%. 2mm V-notch Charpy tests were carried out on these plates
The Charpy test results of the materials prestrained by two different ways of plastic deformation were compared, and it was found that there is substantially not much difference between them. The deteriorating effect of compressive prestrain was analysed in terms of prestrain level. Criteria of Charpy test and CT test were found to rise almost linearly against prestrain level. As for the Charpy test criteria, the most sensitive one was _vTr₁₅, and rate of rise of it was 3 to 3.7 deg C for one p.c. of compressive prestrain. The results of CT test were found to be more sensitive, and the rate of rise on the stress of yield point of base metal was 3.5 to 4.1deg C.
In order to examine the possibility of cracking by brittle fracture from the unnotched part in the case of buckling accident, steel plates were bent to various angle and unloaded. When absolute values of compressive prestrain exceeded the level of ductility of the plate, brittle cracks initiated at the compressive side in the course of unloading. From the fractographic examination, it was found that, at the starting point, fracture surface was covered with not only quasi-cleavage facets but facets of slip planes. After propagation of 2.3mm, fracture surface consisted of cleavage facets.

View full abstract

In this study, the effects of power spectrum density of strains on the fatigue life of a pressure vessel steel were examined with random loading fatigue tests. Different three types of the power spectrum density of strains, S₁ (symmetrical, single peak shaped), S₂ (unsymmetrical, single peak shaped) and S₃ (two peaks shaped) were used. One problem in a comparative discussion of the fatigue life by a random loading with that by a constant amplitude loading is the definition or representative expression of the fatigue life under random loading. In this paper, a new expression N_{f (rep)} was defined with the idea of representative frequency. It was found that the fatigue life corresponding to different strains spectrum could be put in order in a lump with r. m. s. value of strains by using the N_{f (rep)} . In the same way, the constant amplitude fatigue data and the random fatigue data were compared with each other using N_{f (rep)} and r. m. s. value of strains, showing a good agreement. The life estimating method TYPE 1 on the basis of r. m. s. value of strains and N_{f (rep)} was proposed. Mean hysteresis energy W newly defined. W corresponding to a particular r. m. s. value of strains showed a little lower value (about 90%) in the random fatigue test than in the constant amplitude fatigue test. Both of the constant amplitude test data and the random fatigue test data were plotted on a log-log coordinate, and each data were located on two lines which had a same slant and the distance of them was about 0.80 in terms of life ratio. The life estimating method was newly proposed, TYPE 2 obtained on the basis of W and N_{f (rep)} . Comparing the two proposed methods, TYPE 1 is much easier to apply, and TYPE 2 needs more information although it gives more precise estimation. In addition, verification of the Miner's law was also done using three kinds of wave counting methods, i. e, the range count, the hysteresis loop count and the peak count method. As a result, it was found that a proper counting method had to be selected according to the r. m. s. value of strains for application of the Miner's law.

View full abstract

Incidents of unexpected ship hull failures have been on the sharp decrease owing to the technological progress in hull structural analysis, prediction of various kinds of loadings and quality of ship production.
The emphasis in the hull design from the aspect of strength evaluation has shifted its focus from the early failures to the fracture resulting from deterioration with time.
Representative of the latter type fractures is that of fatigue. The structural design that take into account the fatigue fracture, or fatigue design in short, is mandatory not only for LNG carriers but also for critical structural components of other ship types as well. To oil drilling rig structural design, the fatigue design is a matter of primary importance.
In the practice of fatigue design, desiding about what factor of safety to use in designing the structural components adequate for avoiding the fatigue crack initiation requires consideration of many factors contributing to the hull structural integrity.
The computerized hull damage record data bank allow access to statistical hull failure data amassed and classified according to the size and locations of fatigue cracks. There, however, are only few literatures that address the problem of just what factor of safety to use in designing hull structural members.
BS 5400, Part 10, the fatigue rule to be complied with in designing bridges, says “In certain cases, a higher probability of failure could be acceptable, for example, where fatigue cracking would not have serious consequences, or where a crack could be easily located and repaired.” but fall short of being specific as to what percentage of fracture probability to employ for which structural component.
The purpose of the author's study is to quantify the structural integrity, which depends for parameters on such ambiguous quantities as the results of inspection, functional evaluation, etc.
An approach often taken in such an instance is to resort to the graph-network theory for mutal appreciation of phenomena and to the fuzzy sets based analysis for quantification.
The ship hull fatigue failure analysis, on the other hand, depends for parameters on the size of statistical data including those on sea conditions. The reliability engineering is employed in such an instance, and recently the safety index has come into use as factor of safety.
This paper assess the importance ratings of individual hull structural members from the fatigue strength standpoint employing the abovementioned techniques.

View full abstract

In order to evaluate the fatigue strength of the marine propellers of aluminium bronze castings, corrosion fatigue tests under synthetic sea water were carried out by using a great number of specimens taken out from three actual propellers different from both mass and size, while the examination of defect distribution was made on them.
Defects were examined in detailes all over the propellers with penetrant or visual method. The fatigue test results were statistically settled considering scatter of fatigue lives.
Then the correlation between the fatigue strength and the defect distributions was discussed by introducing statistics of extremes on the basis of the weakest link theory. As the results, it was concluded that the fatigue strength is dependent upon the defect distribution, that is, the failure probability relates to the existence probability of maximum defect size at the concerned region.
Finally, the reasonable fatigue strength diagram for design was proposed taking into account of survey periods, etc.

View full abstract

Crack extension by fatigue may sometimes cause a catastrophic failure of structure. For the safety assessment of structure, therefore, it is indispensable to estimate fatigue crack growth under service loading. In most cases actual service loads are random, and the estimating method for fatigue crack growth under random loading must be established.
For practical use, it is very preferable that the prediction of fatigue behavior in real structures under random loading could be made using the data obtained by constant amplitude loading test which is simple and convenient as a conventional and routine test. Many investigations have been carried out in order to clarify the relationship between fatigue behaviors under random loading and constant amplitude loading, recently based on the concept of “crack closure” proposed by Elber. In these investigations, however, pure random load has not been applied because of its complexity, and the relationship has not been clarified yet.
In this study, fatigue crack propagation tests under random loading-narrow band random for its simplicity-are carried out as well as constant amplitude loading tests for comparison. Crack closure phenomena of the specimens are automatically recorded over a long period of fatigue cycles using electric potential method aided by micro-computer system. It is shown that consideration of crack closure is essential for examination of fatigue crack growth and fatigue behavior under random loading may be predicted from constant amplitude test.

View full abstract

In the previous paper, a new method for estimation of fatigue strength under random loading condition was proposed, and the method is combined fatigue tests with calculations. This method was confirmed experimentally and shown to be useful in estimating fatigue strength of structures for random loading.
In this paper, using the proposed method, fatigue strength diagrams for butt and fillet welds of mild steel are obtained for random loading with an exponential distribution. In addition strength diagrams for these joints are calculated using DOE guidance and AWS-Code.
Fatigue strength diagrams proposed by authors are compared with those by guidance and code. As a result, the safety factor considered in guidance and code is larger than the one in other types of fracture, such as buckling and yield.

View full abstract

Behavior of fatigue crack in 3% salt water is experimentally examined under the condition of controlled stress intensity factor as well as constant load amplitude. The emphasis is placed on the effects of crack opening load, frequency and test temperature on the crack growth rate.
The results obtained are summarized as follows;
(1) Since there is notable effect of wedge action due to the corrosion products on the crack closure, the influence of corrosion of the fatigue crack growth can be seen in a much better fashion with the diagram of the crack growth rate da/dN vs. the effective stress intensity factor range than that of da/dN vs. the stress intensity factor range.
(2) As the stress ratio increases, the wedge action of the corrosion product decreases and the effective stress intensity factor range approaches to the nominal stress intensity factor range.
(3) Performing tests under several combinations of load cycle frequencies and water temperatures, it is found that the growth rates in 3% salt water can be almost seven times as large as those in air for some cases. Making use of the above an accelerated test may become possible.

View full abstract

Constant loading tests for double edge-cracked tensile specimens and smooth round bar tensile specimens of mild steel have been carried out at room temperature. And according to these results and visco-elasticity theory, time-dependent deformation behavior has been revealed. Moreover, monotonic loading with extremely low loading rate of cracked tensile specimens and smooth round bar specimens have been carried out. Comparing with these results and constant loading tests, consideration has been conducted on the mechanism of stable fibrous crack growth behavior under constant loading.
Under constant loading, time-dependent strain-that is creep strain-appears; and plastic strain in the vicinity of crack tip increases with time. This time-dependent strain contributes fibrous crack initiation and crack growth under constant loading. The crack growth occurs until time-dependent strain is saturated due to strain-hardening in the vicinity of crack tip, that is, fibrous crack extension under constant loading reaches that obtained at the same load level under monotonic loading with extremely low loading rate. Crack growth is arrested when time-dependent strain is saturated, but unstable fracture occurs when holding load level is higher than maximum load under monotonic loading with extremely low loading rate. Accordingly, the time-dependent strain and strain-hardening behavior plays an important role on the mechanism of stable fibrous crack growth under constant loading.

View full abstract