Assessment of Existing Structures and Repairs - Offshore Structures: Design, Construction and Maintenance - page 494

height with the associated wind speed. The wave contribution was found to be
the most significant factor.
The collapse analysis for a similar platform was performed by USFOS
pushover analysis, as per Soreide et al. (1986) , and it was found that the
minimum RSR was equal to 2.19 at collapse and 1.76 at first failure in the
leg joints at 9.0 m elevation below the mean seal level, i.e., an environ-

mental load from the northwest 119% greater than maximum wave height
(H max ) combined with wind speed equal to 15.6 m/s. Note that 15% and
30% wall thickness loss reduce the RSR values to 2.02 and 1.55, respectively;
however, in this analysis there was no corrosion loss considered for the
platform.
The pushover analysis was performed without taking flooded members.
The ROV survey reported that there were four braces near the sea bed that
were flooded, potentially indicating that some cyclic fatigue action had led to
through-thickness cracking of the members. It is considered unlikely that
these flooded members led directly to the subsequent structural failure, since
ROV footage implied that two were intact after failure.
It is very important to remember that some joints appeared in the post-
collapse subsea survey to have discolored surfaces and marine growth over
cracks, which are good indications that the failure of the joints happened
some time before platform structural failure.
The collapse analysis was performed regardless of the flooded member data
because cracks in joints increase with time and could have caused a sudden fail-
ure with a high storm.
In-place analysis using the SACS model showed that 90% of the joints had a
unity check (U c ) higher than 1.2. Also, some joints at elevations

−

− 11, − 28 and

− 46 had U c higher than 2.0.
In addition, the minimum pile capacity factor of safety in compression for
storm condition was equal to 1.08, and the minimum factor of safety for tension
was 1.00, which violates API RP2A safe conditions.
Collapse analysis is performed by taking pile-soil interaction, and in this
case the RSR was equal to 1.25. An other collapse analysis was performed
by using a dummy pile stub to define the RSR for the jacket, which in this
case was equal to 1.65. Note that the joint flexibility option was included in
the analysis.
From the dynamic point of view, the natural frequency is calculated through
the eigenvalue problem, which takes the form:

K ϕ = lM ϕ (7.26)

ϕ

where K and M are the stiffness and mass matrices, respectively, and is the

model shape factor of the structure. The software program calculated the lowest
eigenvalue λ and the corresponding eigenvector. From the above, stiffness is the