The electronic pdf versions of the documents found through http://www.dnv.com/ are the officially binding versions. Copyright Det Norske Veritas.
![[Table of Contents]](/dwicons/b_toc.gif)
|
DNV-OS-C502 Offshore Concrete Structures
|
App.D
A 100 Seismic analysis
App.D A
101 Two levels of seismic loading on an offshore concrete structure
shall be considered:| — | strength level earthquake (SLE),
which shall be assessed as a ULS condition |
| — | ductility level earthquake (DLE), for which ductile
behaviour of the structure assuming extensive plasticity is permissible
provided the structure survive. |
LNG containment structures shall be designed in accordance with
DNV-OS-C503. The LNG storage tank shall be designed for both the
SLE and DLE earthquakes. Systems which are vital for the plant system
shall remain operational for both SLE and DLE.
App.D A
102 If ductile response of specific components of the structure
under the SLE event is predicted or considered in the analysis,
such components shall be designed for ductile behaviour, in accordance
with Sec.6. Expected best estimate
of stress/strain parameters associated with ductile behaviour
may be adopted in the analyses. Due consideration shall be given
to the effects of overstrength with respect to the transfer of forces into
adjoining members, and for the design of those failure modes of
such members that are not ductile, such as shear failure. For those
cases where the structure can be designed to the DLE event applying
normal ULS criteria, no special detailing for ductility is required.
App.D A
103 Seismic events may be represented by input response spectra
or by time histories of significant ground motion, in accordance
with DNV-OS-C101 Sec.3 E800. Where the global response of the structure
is essentially linear, a dynamic spectral analysis shall normally
be appropriate. Where non-linear response of the structure is significant,
transient dynamic analysis shall be performed.
App.D A
104 Seismic response of a structure is highly dependent on the
natural periods of the structure over a range of modes. This relies
upon accurate assessments of the platform mass and stiffness, and
a best estimate of soil stiffness.
Such parameters shall be carefully assessed and, if necessary, the
sensitivity of the structure to changes in these parameters shall
be evaluated.
App.D A
105 Interaction of the structure with its foundation is particularly
significant for seismic analysis. The foundation shall be simulated
with sufficient accuracy in global structural analysis to ensure
accurate assessment of natural periods of vibration and a suitable
distribution of soil loads into the structure.
App.D A
106 Two principal types of seismic analyses are suggested for
fixed concrete structures:
| — | direct soil-structure analysis |
| — | impedance function/substructure analysis. |
App.D A
107 For direct soil-structure analysis, the caisson may be modelled
as a rigid structure connected to a flexible simulation of the foundation.
In substructure analysis, the caisson may be considered as a rigid
circular disk for the computation of impedance functions.
App.D A
108 Consideration shall be given to the range of likely values of
soil stiffness in the analysis. In particular, the possible degradation
of soil properties during high-level seismic events, such as the
ductility level earthquake, shall be considered. Appropriate non-linear
or reduced soil stiffness properties shall be used.
App.D A
109 Soil properties, particularly shear wave velocity, dynamic
shear modulus and internal damping are dependent on the shear strains
used. These values should be adjusted for the expected strains appropriate
to the seismic excitation and the variation in vertical effective
stress and voids ratio due to the presence of the structure.
App.D A
110 The simulation shall include a representation of the mass of
the structure, in accordance with Appendix
B A300. Enclosed fluids can be included as a lumped mass
where the height of water column is short.
App.D A
111 Unless a detailed evaluation is made, critical internal damping
of not more than 5% shall be used to simulate structural
and hydrodynamic damping for seismic analysis. Any increased value
shall be subject to justification based on expected response. Values
of soil damping shall be determined based on the soil type present.
App.D A
112 For the strength level earthquake, linear dynamic global structural
analysis may be performed using the response spectrum approach.
Spectra used shall be in accordance with DNV-OS-C101 Sec.3 E800,
but the analysis shall incorporate the effect of the platform mass
on near soil motions, if appropriate. If degradation of soil properties
and non-linear soil-structural interaction are significant, a non-linear
dynamic time history analysis procedure should be adopted to address
these effects, although the platform structure may be modelled as
linear elastic. Where seismic isolation or passive energy dissipation devices
are employed to mitigate the seismic risk, a non-linear time history
procedure will be required. If degradation of soil properties, non-linear
soil-structural interaction or base sliding is significant, a non-linear
dynamic time history analysis procedure should be adopted to address
these effects.
App.D A
113 Sufficient modes shall be included in the analysis to provide
an accurate estimate of total global response. At least two modes
shall be considered in each of the two principal horizontal directions
and a torsional mode about a vertical axis. This requirement may
be considered satisfied if it is demonstrated that for the modes
considered in the analysis, at least 90% of the participating
mass of the structure is included in the calculation of response
for each principal horizontal direction.
App.D A
114 One design spectrum may be used equally in each principal
horizontal direction, combined with 2/3 of this spectrum in
the vertical direction unless a lesser value can be justified based
on site-specific data. These spectra may be combined together modally
using the complete quadratic combination method (CQCM) and directionally
using a square root sum square (SRSS) approach. Alternative methods
are permitted with suitable justification that all seismic action
effects are included.
App.D A
115 Secondary spectra may be developed for the analysis of components
such as deck or conductor frames to evaluate the response of substructures,
appurtenances and equipment not modelled for the global analysis.
Alternatively, the design of local components may be based on equivalent
pseudo-static analysis of such components, based on maximum vertical
and horizontal accelerations obtained from the global seismic analysis.
App.D A
116 Action effects from seismic analysis shall be combined with
similar results from gravity loading to produce action effects for
structural design. Appropriate directions of seismically induced
forces shall be considered to maximize these action effects.
App.D A
117 For the ductility level earthquake, non-linear seismic analysis
may be performed using a time history or transient approach. Unless
time histories are available by scaling or by other means, they
may be developed numerically from the design spectra. Multiple time
histories are required to represent the random nature of seismic
ground motions. At least three sets of tri-axial statistically independent
time histories shall be developed and used, and the maximum response
parameters shall be adopted in the design. The requirement for statistically independent
time histories may be considered satisfied if the correlation coefficient
between any pair of time histories is less than 0.3. The time step
for integration shall be selected to ensure accuracy and stability
of the non-linear dynamic solution, normally it should not be more
than 1/(12f).
App.D A
118 The computer model for ductility level earthquake analysis
shall include discrete models of all primary components of the structure
using either linear elastic or material non-linear simulations.
Deflection effects shall be evaluated and gravitational loads included
in the analysis to ensure that second order effects are simulated
with sufficient accuracy.
App.D A
119 The action effects on components that are simulated as linear
elastic in either the SLE or the DLE analyses shall be evaluated
and used to confirm that these components satisfy ULS criteria.
Components that demonstrate ductile response shall be so designed,
and assessed against acceptance criteria relevant for the actual
limit state with respect to all relevant response parameters.
![[-]](/dwicons/i_colpse.gif)