energo otc 18325
TRANSCRIPT
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 1/11
OTC 18325
Assessment of Fixed Offshore Platform Performance in Hurricane IvanPuskar, F.J., Spong, R.E. and Ku, A., Energo Engineering; Gilbert, R.B. and Choi, Y.J., The University of Texas at Austin
Copyright 2006, Offshore Technology Conference
This paper was prepared for presentation at the 2006 Offshore Technology Conference held inHouston, Texas, U.S.A., 1–4 May 2006.
This paper was selected for presentation by an OTC Program Committee following review ofinformation contained in an abstract submitted by the author(s). Contents of the paper, aspresented, have not been reviewed by the Offshore Technology Conference and are subject tocorrection by the author(s). The material, as presented, does not necessarily reflect anyposition of the Offshore Technology Conference, its officers, or members. Papers presented atOTC are subject to publication review by Sponsor Society Committees of the OffshoreTechnology Conference. Electronic reproduction, distribution, or storage of any part of thispaper for commercial purposes without the written consent of the Offshore TechnologyConference is prohibited. Permission to reproduce in print is restricted to an abstract of notmore than 300 words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, OTC, P.O.Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
AbstractHurricane Ivan is one of several hurricanes that have
damaged or destroyed fixed offshore platforms in the Gulf of
Mexico in recent years. These events provide a unique
opportunity to determine the effectiveness of structural design
standards and regulations and develop recommendations for
changes, if needed. Specifically, Ivan provided an opportunity
to evaluate the API RP 2A (RP 2A) design process for fixed
platforms to ensure that it provides for well designed
structures.
The first part of this paper describes the general impact of
Ivan on fixed platforms in terms of survival, damage or
destruction. Specific findings and trends are reported related
to global platform performance as well as component
performance. The second part describes a quantitative
assessment to determine the adequacy of the RP 2A design
process. The approach uses a probabilistic based process that
compares analytically predicted platform damage and survival
to that actually observed during Ivan. The result is a Bias
Factor that reflects how well RP 2A predicts platform
behavior under hurricane loads. The work was funded by the
Minerals Management Service (MMS).
Introduction
Ivan was one of several hurricanes in the last dozen yearsthat have significantly damaged or destroyed fixed offshore
platforms. Prior hurricanes are Andrew in 1992 and Lili in
2002. Katrina and Rita in 2005 also caused significant
platform damage and destruction. These types of events
provide an opportunity to determine how fixed platforms in
the Gulf of Mexico perform in hurricanes on both a qualitative
and quantitative basis. The qualitative basis includes a review
of the typical types of damage to topsides and jacket, as well
as the general trends observed, such as the number and type of
platforms with wave-in-deck damage. The quantitative basis
involves the comparison of the observed damage with what
would have been predicted by RP 2A which is the basis for
design of fixed platforms in the Gulf of Mexico. This
provides a quantified assessment of the accuracy of RP 2A
and if it is adequate for design. This paper describes these
assessments for Ivan based upon an in-depth study performed
for the MMS focusing on fixed platforms (no caissons)
[Energo Engineering, 2006]. These types of assessments have
been performed previously for Andrew and Lili [Puskar, et
al., 1994 and 2004]. Recent hurricanes Katrina and Rita in
2005 provide similar opportunities, but have yet to be studied.
Ivan CharacteristicsIvan developed off the west coast of Africa in late August
2004. By September 5th it was a hurricane about 1,100 miles
east of the southern Windward Islands. The hurricane
strengthened running south of the Dominican Republic and
passed within about 20 miles of Grand Cayman on the 12th
By the late afternoon on September 15th, Ivan was in the east-
central Gulf of Mexico approaching the deepwater offshore oi
and gas facilities. During this time, the hurricane was a
Category 4 storm on the Saffir-Simpson scale, with maximum
sustained wind speeds of more than135 mph.
Figure 1 shows the storm track through the key offshore
oil and gas blocks. Also shown in the figure are the fixedplatforms that were destroyed during the hurricane. Ivan
tracked North over the deepwater facilities in the Mississippi
Canyon blocks and up into the Viosca Knoll (VK) and Main
Pass (MP) block areas. The majority of the destroyed o
damaged fixed platforms resided in the VK and MP block
areas. Ivan continued its Northerly track through the eastern
edge of the Mobile block area, making landfall as a major
hurricane with maximum winds of 130 mph on the early
morning of September 16th
just west of Gulf Shores, Alabama.
General Platform DamageA total of seven fixed platforms were destroyed in Ivan as
shown in Table 1. Figure 1 shows the location of theplatforms destroyed. One of the seven (MC 20A) was toppled
by a mudslide, while the other six failures are thought to be
attributed to metocean loads (i.e., wind, wave and current)
exceeding the capacity of the structures. The seven destroyed
platforms are from data provided by the MMS. Note tha
additional platforms may have been later decommissioned by
the operator as a result of damage sustained from Ivan.
There were also a number of other fixed platforms tha
sustained varying degrees of damage during Ivan. Some o
the damage and failures were not considered a surprise, since
the most of the platforms that failed or sustained major
damage were older facilities. These platforms were generally
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 2/11
2 OTC 18325
designed to lower metocean conditions and generally have
lower global strength characteristics (e.g., weaker joints, less
robust bracing patterns, etc.) than platforms designed and
constructed to newer industry practices. Additionally, these
older platforms typically have lower topside deck heights
which make them significantly more susceptible to wave-in-
deck, which can dramatically increase the loads on the
platform and cause damage. The extent of topside damage onmany of the platforms, both new and older vintage, indicated
Ivan caused extremely large waves and associated wave crest
heights.
The majority of the platforms that failed or sustained major
damage during Ivan were in water depths between 200 to 350
feet and had deck heights at or below the current RP 2A
minimum deck elevation requirements. The resulting damage
to the topsides included deck structure failures and
deformations generally as a result of wave-in-deck. Wind
damage was also observed on quarters and building structures.
The damage to the jackets included jacket leg buckles and
separations, bracing failures (e.g., parted and buckled
members), joint failures (e.g., crushed joint cans and brace
punch through) and conductor bracing failures as discussed
later.
It is important to note that even though damage and
destruction of platforms occurred during Ivan, advance
warning allowed thousands of offshore workers to be safely
evacuated from Gulf facilities prior to the storm reaching the
area [API, Hurricane Readiness Conference, 2005]. There
were also no significant environmental effects.
Although a number of platforms sustained damaged, the
majority of the platforms in the path of Ivan weathered the
storm unscathed or with only minor damage. Figure 2 shows
the percent breakdown of undamaged and damaged fixed
platforms in the path of Ivan. The path represents the
approximate boundaries of the hurricane strength winds, takenas 35 miles to each side of the storm centerline track, although
some of the damaged platforms were outside of this path.
Figure 3 shows the breakdown of the damaged and
undamaged platforms based upon year installed. It is evident
that older platforms sustained more damage than newer
platforms. This is the same observation as for prior
destructive hurricanes Andrew and Lili [Puskar, et. al., 1994 &
2004]. This clearly indicates the improvements in industry
design practices with time and those newer platforms are
much less susceptible to destruction and damage in hurricanes.
Wave Crest and Wave Height Observations
The appropriate deck height for new design as well as forstructural assessments of existing platforms used in RP 2A
Section 17 is a hotly debated topic since Ivan, as well as after
Katrina and Rita. Figure 4 shows the deck heights of the
platforms in the path of hurricane Ivan. The figure shows that
the majority of platforms with decks above the RP 2A Section
2 (new design) minimum deck elevation criteria did not
sustain major damage. The figure also shows a cluster of 200-
350 ft water depth platforms with decks lower than the RP 2A
Section 2 deck criteria that either failed or sustained major
damage during Ivan. One item to note in the figure is there
are a number of deck heights which appear to be questionable
since they are over 55 feet. The deck height data shown in the
figure was obtained from the MMS platform database. It i
suspected that some of these deck heights are the cellar deck
top of steel or in some cases the drill deck instead of the
required cellar deck bottom of steel.
Figure 5 shows the RP 2A design and Section 17
assessment wave height curves compared to the maximum
wave height at the associated platform location per the Ivan
hindcast [Oceanweather, 2004]. The comparison indicates themaximum wave heights during Ivan were generally in excess
of the current RP 2A Section 2 wave height criteria for new
design. The maximum wave heights were computed using the
maximum significant wave height (Hsig) at the site during
Ivan and then converting this to the maximum wave heigh
(Hmax) as computed by the Forrestall distribution. The figure
indicates that it is likely that the platforms in the path of Ivan
particularly the older platforms were exposed to loads in
excess of their original design. The majority of the platform
survived due to the inherent safety factors in the designs.
Additional study of the hindcast data compared to field
observations indicates that several platforms had wave-in-deck
damage, yet this would have not been predicted by the Ivan
hindcast in terms of the crest elevation. In other words, the
predicted maximum Ivan crest elevation is less than the
platform’s deck elevation, yet the platform sustained wave-in-
deck damage. This may be due to several factors including a
lower hindcast Hsig than actual Hsig, uncertainties in
converting Hsig to Hmax, uncertainties in computing the cres
elevation from Hmax, unusual wave crest shapes, or other
factors. This is a technical issue that needs to be explored
further in other studies.
Typical Ivan Fixed Platform DamageThis section provides examples of the typical types of
fixed platform damage caused by Ivan. The damage is broken
down as Topsides Damage (wave-in-deck and wind induced)and Jacket Damage
Topsides Damage -Wave-in-Deck
The majority of the fixed platforms that sustained damage
had evidence of wave-in-deck. The damage includes deflected
structural members on the underside of decks, and in many
cases, damage to equipment and support systems (piping
cable trays, etc.) on the lower decks. Wave inundation on
older platforms with lower decks is not necessarily a surprise
However, some of the newer platforms (1990’s design) also
experienced wave-in-deck, although no major jacket structure
damage occurred, but it did cause significant downtime and
repair costs.The structural damage to the topsides consisted o
distorted lower decks (plating and support under deck
structure), equipment foundation deformation, and in some
cases destroyed equipment shelters on the lower decks.
There was also non-structural damage that was in some
cases more pronounced than structural damage. This
consisted of damaged equipment (power controls, generators
etc.), cable trays, and support utilities located on the lower
decks of the platform. Displaced or missing grating, damaged
handrails and stairs and damaged quarters hampered recovery
efforts as these components need to be in good order prior to
repair activities. Getting these as well as support and safety
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 3/11
OTC 18325 3
systems (power, fire water, etc.) up and running restricts the
immediate and/or permanent manning of the platform,
requiring work to often be done on a day-trip basis.
Generally, the non-structural damage associated with wave-in-
deck was one of the greatest contributors of platform
downtime following Ivan.
Topsides Damage - WindWind was also a contributor to topside damage as shown
in Figure 6. Numerous platforms exhibited signs of topside
structural failures due to wind loading including damage to
building cladding and displacement/damage of light weight
structures and equipment. One of the more significant wind
related issues was the damage to the temporary crew quarters
on the Petronius compliant tower. The quarters and heliport
toppled over into the center of the deck under wind loads,
resulting in serious damage and considerable downtime
[Wisch, 2005].
Jacket DamageThe majority of the underwater jacket damage was
confined to older platforms. Underwater jacket damage
includes jacket leg failures, joint failures, brace failures and
conductor guide frame failures. Examples are provided in the
following.
Jacket Leg FailuresLocal leg buckling was observed on four of the platforms
that sustained major damage. Three of the SP platforms are of
similar design, installed in late 1960s in approximately 230 ft
water depth. All three have an 8 pile with 8 skirt pile
configuration and are orientated in the same direction. Wave-
in-deck was observed on all three platforms and local buckles
were observed on the North/Northwest legs. Ivan approached
the platforms from the southeast and it was the leeward sidelegs under higher compression loading that buckled.
A MP platform also sustained leg buckling and separation
on the two diagonally opposed legs. The orientation of the
platform and photos of the observed damage are shown in
Figure 7. The platform is a four pile platform and the A1 and
B2 legs were separated. The X-bracing was also separated at
two locations near the leg damage. Note that the wave action
and subsequent movement of the platform caused the leg to
expand outward at the both ends, almost as if it were an
external ring stiffener, as seen in Image A. Similar damage
was observed in Lili [Puskar, et.al., 2004].
Joint FailuresJoint failures including cracks, punching and crushing
were observed on many of the platforms that sustained major
damage. Several examples are shown in Figure 8. Image A
shows a 24-inch diameter X-brace joint crushed under wave
loading from Ivan. The platform was designed in the late
1970’s. Since then, RP 2A has improved joint design
formulations. In this case, a joint can (i.e., the thicker walled
section of the through member) was present in the design.
However, it was only marginally thicker than the connecting
members and failed.
Figure 8, Image B shows an example of joint punching
failure. The brace was pushed through the chord member and
demonstrates a classic punching failure. This damage wa
located on conductor guide framing.
Brace FailuresThe majority of the platforms with major damage
sustained jacket brace failure. Most of bracing damage wa
local buckling of the bracing. Figure 9, Image A, shows
several examples, including a buckled 24” diameter X-braceNote that in this photo the marine growth was not cleaned off
instead it popped-off as the brace deformed. Marine growth
that has popped-off in this manner is often a clue during
inspections that some form of damage has occurred and
further inspection is required. Figure 9, Image B, shows an
example of a severed brace. The brace is 26-inch diameter x
½-inch wall thickness and the material yield strength is 50 ksi
Note the ends of the brace have been flattened out. This
occurred after the brace separated, and then the brace ends
came into repeated contact caused by the back and forth
motion of the jacket during the storm.
Conductor Guide Frame FailuresThis type of damage has been observed in Andrew and Lil
and is the result of fatigue damage due to the upward and
downward loads as waves pass through the structure. An
example is shown in Figure 10 including the location of the
conductor tray on the platform. In extreme storms like Ivan
these normally low-stress high-cycle fatigue issues become
high-stress low-cycle fatigue that quickly escalates to this type
of damage. The first conductor guide framing below the
waterline on many platforms is between the -20 ft to -40 f
elevation and if not properly designed can be susceptible to
this type of damage. The conductor tray may even come ou
of the water as the trough of an extreme wave like in Ivan
passes the structure, and this may in addition cause wave-slam
loads. The design characteristic typically causing this problemis the plating often found around the conductor guides, which
dramatically increases vertical wave area/load (compared to
conductor frames with tubular framing only). Another cause
is the fact that many older conductor trays are attached to the
legs via long-span horizontal bracing, often without any
vertical bracing to the tray itself, making the tray susceptible
to up-and-down vertical wave motions. This movemen
ultimately results in a fatigue problem. Note in Figure 10
Image A, how the steel coupon from the chord remains on the
end of the separated conductor guide brace. This type o
separation is characteristic of a fatigue failure where the crack
initiates at the weld toe on the chord member, and over time
the crack propagates in the chord material and around theweld. Eventually the brace completely separates from the
chord and may fall off of the platform.
Quantitative AssessmentThis provides a comparison of a platform’s actua
response to the hurricane Ivan (destroyed, damaged or
survived) versus what the load and resistance recipe in RP 2A
predicts in terms of an analytical response. In other words, i
a platform was destroyed in Ivan – would this have been
predicted by RP 2A? A probabilistic Bayesian updating
process was used based upon an approach first used in 1994
for Andrew and repeated in 2004 for Lili. Details of the
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 4/11
4 OTC 18325
approach can be found in the associated references [Puskar, et.
al., 1994 & 2004] and are not repeated here. The Andrew and
Lili studies show that there is about 15% conservatism
inherent in RP2A once all known factors of safety are
removed.
The approach results in what is known as a “Bias Factor,”
which indicates the ratio of the actual capacity of the platform
determined by observation to its analytical capacitydetermined using RP 2A. If a platform survives after a
hurricane, while RP 2A analyses predicted it should have been
destroyed, this platform has a Bias Factor greater than 1.0. In
this case it would imply that the RP 2A analysis recipe is
conservative. The Bias Factor is computed with all known
factors of safety (FS) in RP 2A removed (i.e., the bias is in
addition to the normal RP 2A FS).
The prior work for Andrew and Lili resulted in Bias
Factors of approximately 1.1 and 1.25 respectively. The bias
is approximately 1.15 when Andrew and Lili are combined.
These results imply that RP2A is doing a good job in terms of
fixed platform design, with an inherent conservatism of about
15%.
For this study, the Bias Factor was recomputed considering
Ivan, based upon six platforms – 2 destroyed, 3 damaged and
1 survived. The results are shown in Figure 11. The resulting
Bias Factor for Ivan is 1.0, which means the prediction
matches the observation almost exactly. The Bias Factor for
Ivan was then combined with Andrew and Lili to determine a
combined Bias Factor of 1.10 considering all three hurricanes.
Note that the combined Bias Factor was calculated through a
complicated probabilistic process and is not obtained by
simply averaging the three individual Bias Factors.
The Ivan Bias Factor is lower than for Andrew and Lili.
The lower Ivan results may be explained by the particular
selection of the specific platforms used to determine the Bias
Factor, mostly damaged or destroyed. The inclusion of moreplatforms that survived Ivan, would increase the Bias Factor,
but there was little information on survived platforms
available to this study (most operators study damaged
platforms and not those that survive). There is also a
possibility that some of the damaged platforms had prior
unknown existing damage that was not taken into account in
the assessment. Hence the Ivan Bias Factor is believed to be
conservative.
Another factor for the lower Ivan Bias Factor may be the
large number of wave-in-deck damaged and destroyed
platforms and the associated uncertainties, such as wave crest
elevation and the associated wave-in-deck loads. As noted
previously, wave-in-deck issues should be investigatedfurther.
Overall, the Quantitative Assessment for Ivan indicates
that the RP 2A fixed platform design approach has a Bias
Factor of about 1.0. When combined with Andrew and Lili,
the Bias Factor increases to 1.10. These results indicate that
RP 2A is doing a conservative job.
ConclusionsIvan provided an opportunity to evaluate fixed platform
performance in extreme storms. The results were generally as
expected, with most of the destroyed or damaged platforms of
older design. As in hurricane Andrew and Lili, most of the
destroyed platforms were thought to be the result of wave-in-
deck loads.
There were a significant amount of platforms with wave-
in-deck damage and wind damage that caused the platforms to
be shut-in for an extended period while repairs were made
There was significantly more wave-in-deck damage for Ivan
than for Lili and Andrew, indicating very large waves for
Ivan. In particular, the wave crest elevations as determinedfrom observed deck damage were exceptionally high.
The quantitative assessment indicated once again that RP
2A does a good job of predicting platform performance, with a
Bias Factor of 1.0 for Ivan and 1.10 (or about 10 percent
conservatism) when combined with prior results for Andrew
and Lili. The lower Ivan value is thought to be a combination
of the platforms selected for the assessment, which were
mostly destroyed or damaged and are thought to provide a
conservative estimate (i.e., lower Bias Factor than actual).
AcknowledgementsThe authors wish to thank their respective organizations
for the opportunity to publish this paper. We also wish to
thank the MMS for sponsoring the effort.
ReferencesABS Consulting, “ Hurricane Lili’s Impact on Fixed Platforms”
Final Report to the Minerals Management Service, June, 2004.API, " Hurricane Readiness and Recovery Conference," Sponsored by
the American Petroleum Institute, Houston, Texas, July 26-272005.
API, Recommended Practice for Planning, Designing and
Constructing Fixed Offshore Platforms, API RP 2A, Twenty
First Edition, 2nd Supplement American Petroleum Institute(API), Washington, D.C., October, 2005.
Energo Engineering, “ Assessment of Fixed Offshore Platform
Performance in Hurricanes Ivan Lili and Andrew,” Final Repor
to the Minerals Management Service, Report Number 549January, 2006.
Laurendine, T. “ Hurricane Ivan Impact to Offshore Facilities and
Status on Section 17 Assessments”, Presentation given during2005 Hurricane Readiness and Recovery Conference –Production Facilities Break-out Session, Houston, Texas, July
27, 2005.MMS, “ Impact Assessment of Offshore Facilities from Hurricanes
Katrina and Rita”, News Release #3418, January 19, 2006.Oceanweather Inc., Hindcast Study of Hurricane Ivan (2004)
Offshore Northern Gulf of Mexico, December 2004.O’Conner, P. “Observations from Pompano and Nakika”
Presentation given during Hurricane Readiness and RecoveryConference – Production Facilities Break-out Session
Presentation on Hurricane Ivan Damage Observations, Houston
Texas, July 27, 2005.Puskar, F. J., Aggarwal, R. K., Cornell, C. A., Moses, F. and
Petrauskas, C., “ A Comparison of Analytical Predicted Platform
Damage During Hurricane Andrew”, Proceedings, 26thOffshore Technology Conference, OTC No. 7473, May 1994.
Puskar, F.J., Ku, A. and Sheppard, R.E., “ Hurricane Lili’s Impact on
Fixed Platforms and Calibration of Platform Performance to
API RP 2A,” Proceedings, Offshore Technology ConferenceOTC No. 16802, May 2004.
Wisch, D. “Some Observations - Petronius and VK 900”
Presentation given during Hurricane Readiness and RecoveryConference – Production Facilities Break-out Session, HoustonTexas, July 27, 2005.
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 5/11
OTC 18325 5
1 MC 20 A 475 1984 L1 49 8-P destroyed
2 MP 98 A 79 1985 L1 57.5 TRI destroyed
3 MP 293 A 247 1969 L2 45 8-P destroyed
4 MP 293 SONAT 232 1972 L2 42 4-P destroyed
5 MP 305 C 244 1969 L2 46 8-P destroyed
6 MP 306 E 255 1969 L2 46 8-P destroyed
7 VK 294 A 119 1988 L2 32 B-CAS destroyed
8 MP 296 A 212 1970 L2 46 8-P major (D)
9 MP 277 A 223 2000 L2 50.3 4-P major (D)
10 MP 279 B 290 1998 L2 53.5 major (D)
11 MP 138 A 158 1991 L2 55 4-P major
12 MP 311 B 250 1980 L2 39.5 8-P major
13 MP 296 B 225 1982 L2 49.2 8-P major
14 SP 62 A 340 1967 L2 40 8-P SK major
15 SP 62 B 322 1968 L2 44 8-P SK major
16 SP 62 C 325 1968 L2 48 8-P SK major
17 VK 900 A 340 1975 L2 46.3 8-P major
18 MP 281 A 307 1999 L2 52 4-P major
19 MP 289 B 320 1968 L1 45 8-P major
20 MP 290 A 289 1968 L2 42 8-P major21 MP 305 A 180 1969 L2 45 8-P major
22 MP 305 B 241 1969 L2 46 8-P major
23 MP 306 D 255 1969 L2 46 8-P major
24 MP 306 F 271 1978 L2 49 4-P SK major
25 VK 786 A-Petronius 1754 2000 L1 55 C-TOWER major
26 VK 780 A-Spirit 722 1998 L1 49 4-P minor
27 VK 823 A-Virgo 1130 1999 L1 47 OTHER minor
28 MP 261 JP 299 2001 minor
29 MP 298 B-VALVE 222 1972 L2 43 4-P minor
30 MP 144 A 207 1968 L2 62.2 4-P minor
31 MP 252 A 277 1990 L2 50 4-P SK minor
32 MP 280 C 302 1998 L2 52 minor
33 SP 60 D 193 1971 L2 49 8-P minor
34 VK 989 A-Pompano 1290 1994 L1 55.8 4-P SK minor
Damage Category
(Note 1)
No. Area Block Year of
Installation
Exposure
Category
Deck Height
(ft)
Structure TypeWater Depth
(ft)
Note 1: Damage Categories: Destroyed – Complete Structural collapse of the platform. Major – Severe structural overload to the primary load bearing
members. Major (D) – Major damage and later decommissioned. Minor – Some structural damage but generally to secondary structures.
Table 1 – Platforms Damaged by Hurricane Ivan
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 6/11
6 OTC 18325
#
#
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
::
:
:
:
:
::
:
:
:: :
:
::
:
:
::
:
:
:
:::
:
:
:::
:
::
:
:
:
:
:
:
::
:
::
:
:
::
:
:
:
:
:
::
:
:::
:
:
:
::
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
: :
::
::
::
::
:
:
:
: :
:
:
:: :
:
::
:
:
:
:
::
:
:
:
:
:
:
:
:: :
:
:::
:
:
:
:
:
:
::
:::
:
::
:
:: :
:
:::
:::
:
:
:
:
: :
:
:
:
:
::
:
:
:
::
:
: :
:
:
:
:
:
:
:
:
: :
::
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:::
:
:
::
:: :::
:
:
:
:
::::
::::: :::
:
:
::
:
:
:
:
:
::
:
:
:
::
: ::: ::
:: :
:::::
::::: :
::
:
:::
:
:
::::
::
:: :::
:::: ::
::: :::
:
:::: :::::: :::
:
: ::
:: :: :
:: :::
:
::
::
::
::::
::::::
: :::
:
:
::
:::
:
::
:::
::
:
: :
::: ::: :::
:
::
:
:
:::
:
::
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
::
::
:::
:
:
::::::::::::
:
: :
::
:
:
:
:
:
::
::
:
:
:
:
:
:
::
:
:
:
:
:
:
:
:
:
:
:
:::::::
:
:
:
:
::
:
:
::
:: :
:
:
:
::
:
:
:
:
:
:
:
:
: :
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
: :
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
: :
:
:
:
:
:
:
: :
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:::
:
: :
#S
#S
#S
#S
#S
#S#S
#S#S
#S
#S
#S
#S#S
#S
#S
#S
#S#S
#S
#S
#S
#S #S#S
#S
#S#S
#S
#S
#S
#S
Mobile
Chandeleur Area
Viosca Knoll
Chandeleur Area, East Addition
Main Pass Area
Breton Sound Area
Main Pass Area, South and East Addition
Viosca Knoll
West Delta Area
South Pass Area
South Pass Area, South and East Addition
South Pass Area, South and East Addition
South Pass Area, South and East Additionelta Area, South Addition
MP 277'A'4-Pile in 223' Water DepthInstalled in 2000
MP 296 'A'8-Pile in 212' Water DepthInstalled in 1970
MP 279 'B'4-Pile in 290' Water DepthInstalled in 1998
MC 20'A'8-Pile in Mudslide Area475' Water DepthInstalledi n 1984
MP 305'C'8-Pile in 244' Water DepthInstalled in 196946' Bottom Deck Height
MP 306'E'8-Pile in 255' Water DepthInstalled in 196946' Bottom Deck Height
MP 293'A'8-Pile in 247' Water DepthInstalled in 196945' Bottom Deck Height
MP 293 SONAT4-Pile in 232 Water DepthInstalled in 197242' Bottom Deck Height
MP 98'A'
Tripod in 79" Water DepthInstalled in 198557' Bottom Deck Height
VK 294'A'Braced Caisson in 119' Water DepthInstalled in 198832' Bottom Deck Height
09/15/2004 6pm88.2W 28.8N135 mphmax. wind
27.49mb pressure
09/15/2004 10pm88.1W 29.3N135 mphmax. winds27.55mbpressure
09/16/2004 2am87.8W 30.2N130 mphmax. wind27.85mbpressure
MATTERHORNMATTERHORN
HORNMOUNTAINHORNMOUNTAIN
RAM POWELLRAM POWELL
NEPTUNENEPTUNE
PETRONIUSPETRONIUS
MARLINMARLIN
VIRGOVIRGO
SPIRITSPIRIT
Figure 1 – Path of Hurricane Ivan Showing the Locations of the Destroyed Fixed Base Platforms.
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 7/11
OTC 18325 7
6%
14%
6%
75%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Destroyed Major Damage Minor Damage No Damage
Damage Category
P e r c e n t a g e o f P l a t f o r m s i n S t o r m P
a t
h
Figure 2 - Damaged Platforms Sorted by Damage Type
2
23
62
4
11
3
4
4
3
2
1
4
0
10
20
30
40
50
60
70
80
Pre - 1978 1978 - 1991 (9th Edition) 1992 - 2000 (19th Edition) 2001 - Present (21st Edition)
Platform Vintage (year)
N u m b e r o f P l a t f o r m s
Destroyed
Major Damage
Minor DamageNo Damage
One Platform Destroyed
by Mudslide
Figure 3 - Damaged Platforms Sorted by Vintage (Year Installed)
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 8/11
8 OTC 18325
20
30
40
50
60
70
80
0 100 200 300 400
Water Depth (feet)
D e c k H e i g h t ( f e e t )
RP 2A - Section 2
RP 2A - Section 17 - L1
RP 2A - Section 17 - L2
RP 2A - Section 17 - L3
Destroyed
Major Damage
Minor Damage
No Damage
MP 98 A
Modified Caisson
Destroyed
MP 138 A
4-PileJacket Damage
VK 294 A
Braced Cassion (Designed for
wave inundation)
Destroyed
Questionable Deck Hei hts
Section 2 - L1
Section 17- L1
Section 17- L2
Section 17- L3
Figure 4 - Deck Heights of Platforms in the Path of Andrew Compared to API RP 2A Minimum Deck Elevation RequirementsThe deck heights were taken form operated supplied elevations to the MMS. As indicated, some of the deck elevations above 55’ may be
inaccurate since few platforms have decks this high.
20
30
40
50
60
70
80
0 50 100 150 200 250 300 350 400
Water Depth (ft)
I v a n H i n d c a s t W a v e H e i g h t ( f t )
Section 2 - L1
Section 2 - L2
Section 17 - L1 - DL
Section 17 - L2 - DL
Destroyed Platforms
Major Damaged Platforms
Minor Damaged Platforms
Section 2 - L1
Section 2 - L2
Section 17 - L1
Design Level
Section 17 - L2
Design Level
Figure 5 - Hindcast Maximum Wave Heights at Locations of Platforms in the Path of Andrew Compared to API RP 2A Wave Height Criteria
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 9/11
OTC 18325 9
Image A
Image B
Image D
Image C
Figure 6 - Topsides Wind Damage from IvanImages A to D show a variety of deck equipment damaged by wind.
B2B1
A1 A2
P l a t f o r m
N o r t h
Legseparation
Conductors
Legseparation
Leg
Leg
Pile
Image A Image B
Leg
Platform Orientation
Figure 7 - Jacket Leg Damage from IvanImage A shows the leg where it severed and was flattened due to the back and forth motion of the waves.
Image B shows a severe near a circumferential weld.
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 10/11
10 OTC 18325
Image A
Image B
Figure 8 - Joint Damage from IvanImage A shows a collapsed 24” joint can that was undersized.
Image B shows a brace punched completely through the chord.
Image A
Image B
Figure 9 – Brace Damage from IvanImage A shows a buckled 24” brace.
Image B shows a 26”brace that completely severed.
7/30/2019 Energo OTC 18325
http://slidepdf.com/reader/full/energo-otc-18325 11/11
OTC 18325 11
Couponremains
Couponremains
Conductors
Image A
Image B
Typical Fixed Platform Framing
First elevation of
conductor bracingbelow waterline
(typically between-25’ to -40’ feet).This is theconductor framingelevation most
susceptible todamage.
Waterline Elevation
Typical boatlanding structure
ConductorsPumpCaisson
Legs
Figure 10 - Example of Conductor Guide Framing Damage from IvanThe figure on the right shows the typical location of conductor guide framing located near the water line that is prone to this type of damage.
Figure 5Bias Factor Comparison
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0.6 0.8 1.0 1.2 1.4 1.6 1.8
Bias Factor for Jacket Strength
P r o b a
b i l i t y D e n s i t y
Combined
Andrew
Lilly
Ivan
Ivan, mean=1.00
Lilly, mean=1.24
Combined, mean=1.10
Andrew, mean=1.09
Figure 11 - Comparison of Bias Factors Determined for Ivan, Lili and Andrew