In this paper, the influence of plate thickness on the fatigue strength of non-load-carrying fillet weld joints is discussed, from the viewpoint of the residual stress and the stress concentration.
The calculation methods of the weld residual stresses in cruciform joints and T joints are presented, and the residual stresses are expressed in functions of plate thickness, yield strength, and heat input. The calculated residual stresses are compared with the experimental ones, and are in good agreement with.
The combined effects of mean stresses (or of the stress ratio) and the weld residual stresses are applied to the fatigue strengthes of the cruciform joints and the T joints, and it is confirmed that the difference in the fatigue strength between the two types of joint is due to the difference of the residual stress distributions.
In the last part of the paper, the estimation method on the fatigue strength of the non-load-carrying fillet weld joint is presented. By means of the method, the fatigue strength of the joint is estimated, with the fatigue strength of the base metal, the stress ratio, the stress concentration due to the joint, and the residual stress. On 67 sets of the test results, the applicability of the estimation method is confirmed.

View full abstract

Recently, Al-alloys are widely used in large scale welded structures with various degrees of plate thickness. For example, 1 m/m thick plates are used on honey-comb structures and on shell plate of hulls. Medium thick plates (4-9 m/m) are used in structural members of ships and 10-60 m/m thick plates are used for large structures such as pressurized or refrigerated storage tanks.
Many researches have been carried out on the fatigue strength of Al-alloy welded joints. However, because of difference in previous and advanced welding techniques, there are difficulties to apply past research results to the subject newly designed. The present research was carried out with the following purposes.
(1) To Collect data on the fatigue strength of welded joint with various bead shapes within plate thicknesses of 1 m/m to 60 m/m and to analyse the data by bead shape factors.
(2) To Propose a new calculation formula of stress concentration factors around weld beads on thick plate joint.
(3) To Propose a fatigue strength assessment correlated to the stress concentration of the bead toe.
(4) To Assess the control method used in the yard to keep a certain value of fatigue strength.

View full abstract

The most characteristic work in shipbuilding industry is to bend thick-steel plates. It is often said that plate bending technique is one of the most difficult one.
In order to solve those technological problems, a lot of investigations were conducted in the past 30 years in Japan. A series of universal press with multiple piston heads have been studied, in order to automate plate bending process. The other approach was contributed by 4th Subcommittee of the Systems Technology and Computer Application Committee. They proposed that such shipbuilding skills should systematically be analysed to be applicable for researches on Artificial Intelligence and Expert System.
The aim of this paper in to develop a simulator for plate bending by Line-Heating. A Simplified non-linear elastic plate bending model is introduced instead of thermal elasto-plastic and large deformation analysis. Engineering work station is used in the main system and FEA is carried out in numerical calculation. And the system supports good interface. So user can understand plate bending intuitively and acquire know-how of it. The simulator is useful to a training facility for technicians.

View full abstract

This paper presents a method of applying the probabilistic fracture mechanics technique to the selection of better acceptance crieria for CTOD test, primarily in terms of the required fracture toughness and the number of test specimens. Particular reference was given to welded mists of marime structures subjdcted to sea wave induced stresses, such as ships and offshore structures. In carrying out the analysis, sea wave induced stress, crack size in weldments and fracture toughsess of materials were taken as random variables.
It was proposed to decide the required fracture toughness in a probabilistic manner by estimating the distribution of fracture driving parameter which depends on mechanical environmental conditions such as boundary force and crack size and by specifying the risk level of required fracture toughness which corresponds to the importance of structural component to be evaluated.
From the viewpoint of structural safety and reliablility, it is rational that the materials of fracture toughness distribution characteristics with low fracture probability should be given high probability of acceptance in an acceptance criterion. The superiority of acceptance criteria was considered in terms of fracture probability vs. probability of acceptance. The dicussions are made on the numerical results obtained for some specified values of the pamameters inuolved. It is found out that the most excellent acceptance criterion varies with the given conditions, which implies that a single criterion can not be adopted as the universally best one.

View full abstract

In order to make fracture assessment for defects in weldments with respect to brittle fracture, a few methods have been standardized. Those present assessments are all conducted in a deterministic manner. Therefore it is not clear which level of reliability defects evaluated acceptable during a given in-service period correspond to from the viewpoint of brittle fracture.
This paper presents a probabilistic method to evaluate the reliability for defects to survive during a given in-service period, considering defects in welded joints of marine structures subjected to wave loads as reference. Though in the case of probabilistic evaluation it is necessary to take both the probability characteristics of fracture driving parameter caused by the randomness of wave loads and the probability characteristics of fracture toughness caused by the scatterness of material properties into consideration, in this paper only the former was taken as random variable assuming that the longterm distribution of wave loads is the exponential distribution and wave loads occur independently of appearence sequence. The variation of the fracture stress limit with the growth of crack size and the exceeding probability of the required fracture stress were taken into consideration to propose the formulation of the reliability analysis.
Discussions were carried out based on numerical results. Defects evaluated acceptable by present assessment methods have the different reliability depending on the given conditions despite the same in-service period. Even if the given conditions are same, fracture probability of accepted defects increases in substance with the in-service period. Therefore it is required to consider the concept of reliability in fracture assessment for defects.

View full abstract

A new TMCP-manufactured high-toughness, low-temperature steel plate was developed for use in fabricating the cargo tank and hull structure of LPG carriers, which could offer both “the adequate performance in arresting the brittle unstable crack propagation in base metal” and “adequate toughness at the fusion line of welded joints produced by high-heat-input welding process.”
This report reviews the concept followed in developing the high-toughness, low-temperature steel plate and the results of evaluation made of : (1) base metal crack propagation arrest performance, (2) fracture toughness at the fusion line of welded joint produced by high-heat-input welding process, (3) weldability, (4) cold-workability, and (5) line-heating workability.
The above studies showed that the newly developed steel plate could ensure adequate safety and reliability for use in cargo tanks and hull structure of LPG carriers loaded with the cryogenic cargo, with a high level of crack propagation-arrest capability, excellent high-heat-input weldability, and shop workability.

View full abstract

It has been reported that a contract was signed for building an LNG carrier of independent (self supporting) prismatic tank Type B system. This news must be received with a great interest as this is the first case of adopting the independent prismatic tank Type B system for a large LNG carrier.
The idea of the independent prismatic tank Type B system was crystallized in 1970's and a group of ship builders started the development of its design standards in which Nippon Kaiji Kyokai (a Classification Society-NK) played an active role. With the design standards thus developed, number of LPG carriers and Ethylene carriers of this system were built with the NK class in the beginning of 1980's, and nothing has been reported of serious deficiencies and troubles.
Nevertheless, on the occasion of adopting this system for the large LNG carrier, the authors attempted to review the design standards to verify their reliability. For this purpose, the authors reviewed the tank strength analysis, tank test procedure, design and inspection records of ships in service, and made an empirical research on probability of failure of the tanks.
The study concluded that the prevailing design standards for the system were considered quite reasonable and this independent prismatic tank Type B system could be one of the LNG tank systems of high reliability.

View full abstract

An efficient method to generate narrow band stationary Gaussian random load sequences for fatigue test is proposed. Fatigue crack propagation tests under stationary Gaussian random stress intensity factor process were carried out with 3 different power spectral densities to see the effect of load profile on the growth rate. The reciprocals of the experimentally obtained growth rate were plotted on the Weibull probability paper to find that slight change in the band-width has little effect on the probability distribution of the resistance to the fatigue crack propagation. Simulations of crack growth using the data obtained by the previous tests showed a good agreement with the experiments. It is confirmed that the probability distribution of fatigue crack propagation life can be estimated by simulations provided that statistical data are supplied by the test method previously proposed by the authors. (ref. 7)

View full abstract

Some points in the mathematical model of cathodic protection problem in infinite field are discussed. The boundary element method applicable to 3-D 1/2-infinite, 1/4-infinite and 1/8-infinite field cathodic protection problems is derived. Using an appropriate fundamental solution for each case, only the interface between the structure and electrolyte needs to be discretized. Both triangular linear and rectangular bilinear elements are employed to discretize the boundary surface of 3-D field more efficiently and accurately. After several pilot computations for accuracy analysis, the electropotential distribution on the interface between seawater and a ship body under cathodic protection is calculated, and the numerical results show that the present method is accurate and efficient.

View full abstract

There exist two outstanding streams of researches on the optimization of ship routing operation : the optimum solution of ship speed distribution under the weather and the optimum steering control to reduce the propulsive resistance. Up to now, they have been studied separately each other.
In this paper, the fuel saving criterion for the navigating ship is formulated precisely on the basis of the propulsive motion equation under the weather condition. It consists of some vital ship operation variables such as displacement, ship speed (main engine output), course deviation, rudder angle and many physical characteristics data determined definately for a given ship.
Although it is rather complicated with lots of ship's paramaters and main engine performance data, it is proven that this totalized criterion can be hierarchically decomposed to two performance indices by applying the optimum control theory and the non-dimensionalization technique for the ship maneuverability parameters. As a result of this hierarchical decomposition, the ship speed distribution under the weather can be evaluated with the value of performance index for the optimized steering control in use.
This feature ensures the real optimization of the ship routing operation, justifies approximately the seperate approach of the above two streams and suggests the possibility of integration of optimum ship operation control technique.

View full abstract

Ship life depends on many factors. The factors will be divided into three groups, the law of nature, performance and economy. The laws of nature are mainly fatigue and corrosion which reduce the retained structural toughness in aged ships. The performance factors of social impact will sometimes act key point. For example, marine pollution has a great influence on types of ships. In economy factors, investment and its recovery are basic ones that should be analyzed, and cost sensitivity of each of the factors should be analyzed on the point of effect on life. And some investigations show low finantial merit in ship life extention. The authors investigated the factors for proper ship life in basic and fundamental aspect, and confirmed that it should be taken heads to link the factors together. The points to be cleared in engineering point of view are the properties of corroded plates, and some results show very low fatigue strength and fracture toughness that warn low strength redundancy of aged ships.

View full abstract

Just after shipbuilding industries have released several kinds of large scale of CAD/CAM systems, the necessity of Computer Integrated Manufacturing system (CIM) have been recognized in Japanese shipyards. In the previous paper, a shipbuilding definition system (or a modelling for shipbuilding) was discussed. As the result, a System of Design and Assembling for Shipbuilding (which is called SODAS) was proposed. In SODAS, there are some systems which include ship design system, system of cutting of designed ship structure into pieces and system of assembling those pieces to a ship block.
The aim of this paper is to refine assembling system in SODAS. In the construction stage of a ship, a lot of work should be done by using particular facilities in shipyards. Therefore two models are introduced here. One is facility model such as crane, lift-magnet and welding machines. The other is operation model in which operation data such as welding condition, root gap should be managed. An object oriented language on engineering work station Sun-3 is also used to represent a ship and shipyard model.

View full abstract

SHIP & OCEAN FOUNDATION studies on Pilot Model for CIM in shipbulding. One of the subjects in this project, we researched on modeling for Piping Arrangement in Shipbuilding. This paper proposed Piping Arrangement Model based on Object Oriented and Entity-Relationship method. Following this concept, we developed Prototype System by Smalltalk80, this is object oriented program language and enviroment.
The Prototype System applied to midship part of oil tanker and obtained efficiency of proposed Piping Arrangement Model.

View full abstract

Recently, the necessity of CIM in shipbuilding has been recognized in our shipyards. And also SHIP & OCEAN FOUNDATION studies it on pilot model. In this project, process planning system is one of the important application system.
The aim of this paper is to propose one example of process planning expert system.
The proposed system is applied to the typical block of oil tanker and the efficiency of proposed system could be obtained.

View full abstract

The process planning expert system is discussed in this paper. The parts and structure to be constructed are expressed in terms of the Entity-Relationship model. The expert system has three particulars. At first, the production system is based on the modified Role method which was developed on the previous work of authors. Secondly, this system can recognize the substructure in the whole structure. The substructure is also process planned and treated as an assembly in the process planning of the whole structure. Thirdly, the process planning expert system consists of sub expert systems. Each sub system is in charge of the production at each stage, such as the Role generation, substructure recognition, process planning of the whole structure and so forth. With using this system, the system can save execution time and it can be made very clearly. The system was applied to the simplified structure of one block near the midship. The system proved its ability to give an appropriate process planning.

View full abstract

The thermal and moisture states of the environment in enclosed spaces, such as the reefer hold of a refrigerated cargo ship or the air-conditioned living spaces, are manipulated by a forced circulation of air with controlled temperatures. The natural tendency to form thermal layers and heat pockets in the absence of properly balanced heat and humidity distributions in the air circulation, could make the living spaces uncomfortable and damage the refrigerated cargo. This makes it necessary to investigate and establish the various factors that govern this situation, through a detailed analysis of the problem.
This paper makes a precise numerical investigation of the thermal and humid environments in an anisotropic turbulent air flow subjected to gravitational effects, by including the diffusion terms with buoyancy effects also, in the momentum, heat and humidity transport equations. The air flow velocity, temperature and humidity distributions are experimentally measured in a thermally insulated box chamber, in which moist hot air blows and spontaneous evaporation from water pans kept on the floor, were employed to control the temperature and humidity distributions. Numerical and experimental results are compared with each other to justify the validity of the present theoretical formulation of the problem of turbulent heat and moisture transfer with buoyancy effects.

View full abstract

The various factors and conditions for the design of ship equipments contain several indeterminable elements, which could often introduce a certain amount of fuzziness in the designed values. In the previous part of this paper, a method of solution to such problems was attempted by evolving a Cartesian Product Model using the Fuzzy theory. This part investigates the prediction methods for the amount of fuzziness that could be introduced by the indeterminancy of the governing factors in a strongly parameter dependent phenomenon, based upon a combination of mathematical modelling and fuzzy relationships, using a fuzzy determination model.
As a practical example, the phenomenon of the incidence of ship fires and the conditions for its outburst and spread are investigated by dividing it into the flashover of the burning cell and its spread into the adjacent cells. The flashover threshold curve for the outburst of fire in a cell is determined first, and the fuzziness involved in the chances of a flashover is predicted in the threshold curve, using the fuzzy relationships. The conditions under which the fire would spread into the adjacent cells, and the range of thicknesses for the noninflammable intercell wall materials necessary to prevent that danger, are determined then by introducing the indeterminable values into the fuzzy determination model.

View full abstract