Page 31 - Marine Structural Design

8 Part I Structural Design PrincipreS

Ship collision and impacts from dropped objects offshore
Ship grounding
Firelexplosion
Freakwaves
The term "accidental loads" refers to unexpected loads that may result in a catastrophe causing
negative economical, environmental, material consequences and the loss of human life.
Extreme and accidental loads differ in the sense that the magnitude and frequency of the
extreme loads can be influenced to a small extent by the structural design, whereas active
controls may influence both the frequency and the magnitude of accidental loads.
The design for accidental loads includes determination of design loads based on risk
consideration, prediction of structural response using rigid-plastic analytical formulation
and/or non-linear FEM and selection of risk-based acceptance criteria. Traditionally rigid-
plastic analytical formulation has been popular for design against accidental loads because
large plastic deformation is usually the mechanism for energy absorption in accidents. In
recent years, the nonlinear finite element analysis has been applied to simulate the structural
behavior in accidental scenarios and to design the structure for the performance standards. Use
of the finite element analysis enables us to deal with complex accidental scenarios and to
better predict the structural response.
1.2.3 Design for Fatigue

Fatigue damage and defects may threaten integrity of the marine structures. This concern is
aggravated as the cost of repair and loss of production increase. Fatigue design became an
important subject due to use of higher strength materials, severe environmental conditions and
optimized structural dimension. In recent years there is a rapid development in analysis
technologies for predicting fatigue loading, cyclic stress, fatigue/fracture capacity and damage
tolerance criteria. The fatigue capacities are evaluated using S-N curve approach or fracture
mechanics approach. The S-N curves are established by stress controlled fatigue tests and may
generally be expressed as:
N=K.S-"' (1.3)
where:
N = Number of cycles to failure
S =Stressrange
m. K = Material constants depending on the environment, test conditions, etc.
The S-N curve approach is mainly applied in the design for fatigue strength, and it consists of
two key components: determination of hot-spot stress and selection of appropriate S-N curves.
A bi-linear S-N curve is shown in Figure 1.3 where on a log-log scale the x-axis and y-axis are
number of cycles to failure and stress range respectively. The slope of the curve changes from
-
m to r where the number of cycles is NR (= 5 lo6 for steel).
Discrepancy has been observed between the hot-spot stresses predicted by different analysts or
in different analyses. It is therefore important to derive an optimum procedure and standardize
the analysis procedure as part of the xules/code development. In recent years, there has been a
rapid development in the standardization of the S-N curves. In this aspect, International
Institute of Welding (IIW) has published a couple of new guidance documents on the selection