Plastic response of ship structure subject to ice loading

Abraham, Jacob (2008) Plastic response of ship structure subject to ice loading. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (10MB)

Abstract

Stiffened plates are the basic structural building blocks of ships and many onshore and offshore structures. The present research explores the plastic response of a stiffened plate subjected to lateral ice loads. The Finite Element Method (FEM) is extensively used to study the plastic behavior of stiffened plates. In order to gain confidence in using FEM, a validation study of a full scale experiment is initially presented. Following the examination of experimental and numerical results for a loaded grillage, the report examines three separate but related questions concerning the design of ice capable stiffened panels. -- Ship structure design normally considers single frames. This is justified because under uniform loading, all frames behave similarly - so it is reasonable to consider frames singly. In case of ice loading which is non symmetric, the symmetric boundary condition does not accurately represent the true structural behavior. The difference in load carrying capacity between frames in isolation and frames as part of a grillage, subjected to an unsymmetrical loading is studied. -- The capacity of a stiffened plate depends on many factors - geometric properties, material properties, loading type etc. The current IACS Polar Rules for Ships contain plastic limit state models of frame capacity. These limit states are analytically derived using relatively simple energy methods and validated by finite element analyses. The contributions from large deformations and membrane stresses are ignored and hence these analytical solutions may not accurately estimate capacity of all frames. The reliable methods to estimate capacity are either to conduct a full scale experiment or a nonlinear finite element analysis. These methods are either very expensive or too complex. There is a need for a simple regression equation which can predict capacity taking into account all the non-linear behavior of the structure. There are ten factors which influence load carrying capacity of a frame. The study of ten factors at two levels (at high and low levels of each factor) requires 1024 (2¹⁰) ANSYS analyses. A significant reduction in the number of analyses is achieved by using “design of experiments” (DOE) method. A new regression equation for estimating load carrying capacity of the frame is proposed and validated using independent FE analyses. -- The total load carrying capacity of a stiffened plate is contributed by both shell plate and stiffener. In most situations, stiffened panels will be sensitive to buckling under axial loads. In the case of ice reinforcement, the loads are primarily normal to the shell, with minimal axial loads. The concern for frame buckling remains, although the issue is less well understood. Some stiffeners, especially those with slender webs, show a tendency to fail by local web buckling, tripping and shear buckling, causing a sudden loss of capacity and resulting in collapse of the structure. The IACS Polar Shipping Rules (URI2) contains a requirement aimed at the prevention of web buckling by specifying a maximum web height / web thickness (hw/tw). While URI2 employs plastic limit states, the stability ratio is based on prevention of elastic buckling. In some cases these stability requirements have a significant impact on the design. The current rule limiting values of hw/tw are very conservative and do not adequately reflect the conditions that lead to instability. The FEM coupled with the DOE method is used in the study. Six factors which influence stability of a flat bar stiffener are identified. The study of six factors even at two levels (at high and low levels of each factor) requires 64 (2⁶) possible combinations of factors to be considered. A significant reduction in the number of cases is achieved by employing DOE method. The main factors affecting the plastic stability of a frame are quite different from the usual elastic buckling parameters. A new relationship is proposed for calculating the limiting web height and web slenderness.

Actions (login required)

View Item