The problem of structural redesign of plates for static deflection and modal dynamics objectives is formulated and solved by the method of large admissible perturbations. The perturbation approach to redesign is first used to develop response equations for the objective plate design based on its specifications and the baseline plate design. The equations of the objective state are strongly nonlinear implicit expressions of the variable plate thickness. A large admissible perturbations algorithm is developed to solve the plate redesign problem and define the optimal objective state. The latter is reached incrementally with a prediction-correction scheme without repeated finite element analyses. Systematic numerical applications in redesign of a cantilever plate of 216 degrees of freedom are used to investigate the effects of number of extracted modes and redesign variables. It is shown that the large admissible perturbations theory can be used efficiently to redesign plates for multiple specifications that require changes to the baseline design and its response of the order of 100%.

The market of LNG (Liquefied Natural Gas) carrier is remarkably expanded in the last four or five years, and lots of LNG vessels are being built in many shipyards in the world. Membrane-type MARK-III LNG CCS (Cargo Containment System) is used more and more in the construction of LNG carrier, and it has already taken considerable market share in the business of LNG vessel. No matter how many researches have been carried out on the structure of LNG CCS, most of them are mainly focused on its macroscopic behaviors, e.g. structural response of LNG CCS under sloshing load. MARK-III secondary barrier is a matter of primary concern recently, and as already known, its major function is to protect the inner hull structure from the leakage of LNG when the primary barrier of corrugated membrane fails. A closed boundary of secondary barrier for liquid tightness is mainly sustained by the boding between flexible and rigid triplexes, where the adhesive material such as epoxy green glue is applied. The thickness of adhesive glue is about 0.4mm which is extremely thin compared with those of the other structural components of MARK-III CCS. The conventional macroscopic approach hardly gives proper description about the structural behavior of secondary barrier which requires much finer representation with the resolution of glue thickness. Most recently, even though there is an example of research on the structural responses of MARK-III secondary barrier by carrying out structural analysis using microscopic approach, it still is necessary to verify the results of structural analysis base on the experimental evidences. This research deals with an experimental study on the structural behavior of the secondary barrier of MARK-III LNG CCS. The full-scale specimen of MARK-III CCS is prepared and installed in cryogenic chamber which is quite large enough to completely enclose the specimen. As the actual secondary barrier is loaded mainly with thermal loading due to cryogenic temperature and mechanical loading due to hull deformation, the specimen undergoes cryogenic temperature maintained by the chamber and mechanical loading given by the actuator of testing machine. The structural response of secondary barrier in the specimen is maintained and controlled in such a way that the response is almost the same as that of actual secondary barrier in LNG CCS. Through the intensive study on the type and size, the specimen is so designed as to sufficiently realize the structural behavior of the secondary barrier in the actual operating condition. Since the strain gauge is elaborately installed into the thin layer of adhesive glue in the secondary barrier, it is possible to measure its precise response and to capture the realistic structural behavior of the glue. FBG (Fiber Bragg Grating) sensor is also installed in the same specimen, and the structural response of the secondary barrier is measured simultaneously, which provides data of comparison to confirm the reliability of experimental results. It is necessary to verify the performance and feasibility of the strain gauge and FBG sensor prior to actual application into the specimen, because the strain gauge and FBG sensor work in extremely thin layer under cryogenic environment. Therefore, simple tensile test is carried out, and the strain gauge and FBG sensor are examined. During cooling down process, thermal loading is increased until the temperature of secondary barrier reaches −110°C, which is maintained for a few hours to stabilize the structural response of specimen. Continually, the 4-point bending test is carried out to give additional mechanical loading which is monotonously increased until the loading reaches the ultimate strength. The strain gauge and FGB sensor measure and record the strain at each testing location during the whole process of experiment. The specimen of EC (Conventional Epoxy) glue has the smallest ultimate strength, but nevertheless it has sufficient safety factor with respect to the level of loading in actual vessel, and the EHP (Epoxy Glue with the Treatment of Hot Pad) and PU (Polyurethane) glues have much more. The experiment is simulated numerically using FEM. The material data are directly obtained through various material tests. Microscopic approach is adopted, and therefore extremely fine mesh model of which element size is almost equal to glue thickness near the secondary barrier is developed, which makes it possible to represent realistic structural behavior of adhesive glue precisely. The loading and boundary conditions are carefully arranged to embody the experimental circumstances correctly. Finally, it is possible to estimate the degree of discrepancy between the results of experiment and numerical simulation, and the correlation factor can be obtained by studying the discrepancy. The correlation factor is the final result in this research, which can be applied to the structural analysis for actual secondary barrier of MARK-III LNG CCS and improve the results of the analysis.

Ships operating in littoral sea are likely to be subjected to accidental loads especially stranding. Once she has damage on the hull structure, her ultimate strength will be reduced. This paper is to investigate the effect of stranding damage size on the ultimate strength of ship structure by using a series of collapse tests. For the experiment, five box-girder models of 720mm × 720mm in section and 900mm in length were prepared. Of the five, one has no damage and each four have a diamond shaped damage of different size which represents the shape of rock section in seabed. Among the damaged models, three were made by cutting the plate and one by pressing to represent stranding damage. Experiments were carried out under pure bending load and the applied load and displacements were recorded. The ultimate strength is reduced as the damage size increases, as expected. The largest damaged model has the damage size of 30% of breadth and its ultimate strength is reduced by 21% than that of no damaged one. The pressed one has lower ultimate strength than cut one. This might be due to the fact that the plate around the pressed damage area effect unfavorably on the ultimate strength. The models are analyzed with LS-DYNA and the results are compared with the results from the experiments. From the intact model, buckling occurred in the analysis but not in the experiment. The ultimate strengths of other models from the analysis are fairly close to those from the experiments. The errors are within 10%. With the results from the experiments and analysis, an estimation method of ultimate strength of damaged ship on bottom is derived.

No abstract

The automation of fatigue durability analysis for welded bogie frames according to the UIC-code is proposed by using the ModelCenter, which enables several tools used in fatigue durability analysis to be integrated, i.e. I-DEAS, ANSYS and BFAP, and an iterative analysis of geometrical changes in transom support bracket to be performed. In additions wrapping programs to control I/O-data and interfaces of these tools were developed. The developed automation technique brings not only a significant decrease in man-hour required for durability analysis, but also provides a platform for multidisciplinary engineering activities.