Mar 07
Design/Analysis Procedures for Fixed Offshore Platform Jacket Structures
Mar 07 Page 2 of 110
1.0
INTRODUCTION
The most common offshore solution for shallow to medium water depths takes the form of piled-jacket with deck structures, all built in steel. A typical North-Sea jacket is shown in figure 1- Appendix A.
The size and the weight of the jacket structure depends on the number of facilities to be provided on the deck (typically referred to as topsides), water depth, environmental loads imposed on the structure. The number of legs, plan dimensions and brace member configuration are function of topsides area requirement, loading, water depth and environment.
The jacket leg spacing at the top is determined by deck leg spacing and at mud level, by foundation capacity requirement. The vertical batter (double batter) is limited to 1/6 to 1/8 of the height. If the jacket is to be launched, one face has to be vertical, so that the launch trusses are continuously supported. If jackup rig operations are to be carried out then the platform North face must be vertical. As mentioned before, jackets are founded to the seabed by means of piles. Design of these piles is dictated by the jacket reactions at the mudline and the response of the soil to the imposed reactions at the mudline.
Jacket structures have to be designed to maintain structural integrity for the duration of field life (typically called the in-place condition). Here the structure must be designed for strength as well as for fatigue. In addition there are several phases which the jacket structure has to go through before it can be operational, viz. - Fabrication in a yard with access to the sea - Lift, loadout, transportation to offshore location and installation on location
For inplace conditions the structure is designed to resist combinations of design loads that include self-weight and other operational loads as well as environmental loads due to wave, current, wind, earthquake etc. The most commonly used code for designing jacket structures is API-RP2A WSD (Working Stress Design). Other codes include
Mar 07 Page 3 of 110
DNV rules and Lloyds rules. Topside structures are typically designed using the AISCcode along with the AWS code for welding. Figure. 2 shows the components of a jacket platform. These These main ones among these are outlined below 1) Jacket legs 2) Elevation and plan bracings 3) Joints 4) Appurtanances like boat landing, riser guard, stairs etc.
Jackets serve the following purposes: •
Provide the support for the production facilities installed on the deck keeping it stable in the imposed environmental conditions.
•
Provide lateral support to the well conductors and the pipeline riser and also provide protection to them.
The jacket structure is braced in both the horizontal and vertical planes. The braces are also tubulars and connect the jacket legs to each other and reduce the leg effective lengths. Figure 3 shows the typical jacket bracing configurations in use.
The jacket is founded to the sea-bed by means of open-ended tubular steel piles. The pile resists the inplace forces acting on the structure by means of skin friction as well w ell as end bearing resistance. Additionally, lateral load resistance of the pile due to the surrounding soil is required for resisting the horizontal forces imposed on the structure.The number, arrangement, diameter and penetration depth of the piles depend on the environmental loads and the soil conditions at the location. Typical diameters of the piles vary between 1524 mm to 2000 mm though they can be higher. Typical depths of penetration can vary between 90 – 110 meters below the mudline.
Mar 07 Page 4 of 110
There are 3 main types of pilings 1) Pile-through-leg concept, where the pile is installed in the main legs of the jacket. 2) Skirt piles through pile sleeves at the jacket-base, where the pile is installed in guides attached to the jacket leg. Skirt piles can be grouped in clusters around each of the jacket legs. 3) Vertical skirt piles are directly installed in the pile sleeve at the jacket base. This arrangement results in reduced structural weight and easier pile driving. In contrast inclined piles enlarge the foundation at the bottom, thus providing a stiffer structure. The 3 types of piling arrangements are shown schematically in figure 4.
The jacket structure essentially supports the deck. The major functions on the deck of a jacket platform are: 1) Well control 2) Support for well work-over equipment 3) Processing facilities for separation of gas, oil and non-transportable components of the well fluid. For example water, waxes and sand need to be seperated out from the well fluid before transporting back to an onshore facility for further processing 4) Compressor modules and pumps required to transport the product ashore 5) Power generation 6) Living quarters for staff
There are four concepts employed in the f abrication of decks •
the single integrated deck
•
the split deck in two four-leg units
Mar 07 Page 5 of 110
•
the integrated deck with living quarter module
•
the modularized topside consisting of module support frame (MSF) carrying a series of modules
A typical deck structure welded on to the top of the jacket is shown in figure. 5
A deck structure typically consists of the following components 1) Deck legs 2) Deck beams 3) Deck plating / grating 4) Deck bracing
As mentioned previously, the jacket and deck structure has to be designed to withstand the inplace loading conditions imposed on it. In addition they must be able to withstand the forces imposed on it during the operations which typically take place from the time the structure is fabricated in the yard to the time it is installed at site. These conditions are collectively termed as preservice conditions and include typically 1) On shore lift 2) Trailer/skidded loadout on to the transportation barge deck 3) Transportation from the yard to the offshore location 4) Installation of jacket which includes jacket launch, upending, offshore crane lift for decks and modules
A few other inservice conditions which the structure must be designed for are 1) Accidental Boat impact 2) Wave slam 3) On Bottom stability
The next few sections describe the inservice and pre-service conditions for which the jacket and deck structures must be designed. designed.
Mar 07 Page 6 of 110
2.0
DESIGN FOR INPLACE CONDITION
The first premise in the design of jackets is that the jacket natural period is well separated from the wave periods normally encountered in the inplace condition. This ensures that the structure responds in a statically and not dynamically to the imposed wave loading.
Typically jackets have natural periods in the first mode ranging from 2 to 3 seconds. The wave period is typically between 10 to 16 seconds. In such a case the structure can be analysed for the forces imposed on it quasi-statically. In case the structure natural frequency approaches the predominant wave frequency then the analysis must take care of response amplification at the wave period.
Quasi-static design begins with the classification of design for inplace condition into two categories a) Design for strength b) Fatigue design
In each of the above categories a set of loads act on the structure which must be designed to resist these loads effectively. The next section describes the major loads acting on a jacket structure.
2.1
LOADS ACTING ON JACKET STRUCTURE
The loads acting on a jacket structure are typically classified into 1. Permanent (dead) loads. 2. Operating (live) loads. 3. Environmental loads. 4. Construction - installation loads. 5. Accidental loads
Mar 07 Page 7 of 110
In the subsequent sub-sections we examine the loads acting on the structure in the operating or in-place condition.
2.1-a Permanent loads
Permanent loads are those which are present during the entire life time of the jacket structure. The major permanent loads acting on the jacket structure are: a. Dead weight of the structure i.e. weight of the structure in air. This includes the steel weight of the jacket and deck In cases where the annular space between the legs / skirt sleeves and the piles is grouted then the weight of such grouting must be considered in the structural design. Another load whose weight must be considered is the weight of ballast. During the upending operation the the jacket legs are selectively flooded so that the jacket which is floating in the water after being launched can upend and then sink to the sea-bed where it sits before the piles are driven. The weight of this ballast water must also be considered in the structural design. Consideration must also be given to jacket appurtanances like boat landing, riser guard, stairs, ring plates on the jacket legs etc. Dry weights of attachments, fittings and fixtures including architectural finishes, sanitary and plumbing fittings, utility fixtures also find place in this list
b. The deck houses all the processing facilities. This category of permanent loads on the deck includes the weights of equipment, attachments or associated structures which are permanently mounted on the platform. A typical example of a load of this nature is the piping dry weight. Other examples include pressure vessels, pumps, piping, mechanical equipment, cables, switchgear, tanks, HVAC ducting etc.
c. All the tubular members below the water line are subjected to hydrostatic pressure. All non-flooded members will be subjected to compressive hoop stresses and must be
Mar 07 Page 8 of 110
designed to prevent hydrostatic collapse. Additionally all non-flooded members displace water due to which they are subjected to buoyancy forces. This forces reduces the overall reaction due to the dead weigh of the structure alone.
d. Protective coatings (e.g. paints, galvanising, sprayed metal coating, fire resistant coating materials etc.) must also be considered in this list
2.1-b Operating Loads
Any offshore installation in its operating condition is subjected to a number of continuous / non-continuouse operations. These offshore operations result in operating loads. A typical example example of a operating load is the piping operating load i.e. during platform operations piping will have operating fluid flowing through it. The weight of the piping along with its contents is different from the dry weight of the piping alone. This category is also described by the weight of all non-permanent equipment or material, as well as forces generated during operation of equipment. More specifically, operating loads include the following: a. The weight of all non-permanent equipment (e.g. drilling, production), facilities (e.g. living quarters, furniture, life support systems, heliport, etc), consumable supplies, liquids, etc. b. Forces generated during operations, e.g. drilling, vessel mooring, helicopter landing, crane operations, etc. Some loads in this category are genenerated during operations are often dynamic or impulsive in nature and must be treated as such. For example, in designing the helideck according to ABS there are three load cases to be considered viz. distributed loading on deck, impact load due to the helicopter landing and helicopter stowed on the helideck along with wind loads. Clearly these loads are of a non-permanent nature.
Mar 07 Page 9 of 110
Other examples include loads from rotating machinery, drilling equipment etc. that may be treated as harmonic forces. However the basis of design document (BOD) clearly specifies whether such harmonic loading must be considered in the analysis or not. A third non-permanent loading condition is when there is an accidental impact of a supply vessel or transportation barge with the jacket structure. Here again the structure must be designed to withstand this impact force and not lead to catastrophic failure.
2.1-c Environmental Loads
Environmental loads are those caused by environmental phenomena such as wind, waves, current, tides, earthquakes, temperature, ice, sea bed movement, and marine growth. The meteorological and oceanographic conditions (typically referred to as metocean data) at the jacket platform location are determined by experienced and expert consultants. An example of this is Glen’s report which is typically used for design of ONGC platforms.
In the following sections we discuss a general summary of the typical information that is required for design of jacket structures. Final selection of information needed at a site is typically made after consultation with both the platform designer and a meteorologicaloceanographic specialist. Generally the metocean data includes descriptions of normal and extreme environmental conditions as follows:
1. Operating storm, also referred to as operating environmental conditions, that occur frequently during the life of the structure. These conditions are important in terms of the design of the structure for its service life as well as during installation of the platform. These conditions are also referred to as storm with a 1 year return period. The design of the structure under these conditions must be done with no increase in allowable stresses.
2. Extreme conditions (conditions that occur quite rarely during the life of the structure) are important in formulating platform design loadings. This is also referred to as
Mar 07 Page 10 of 110
extreme storm with a 100 year return period. Even though the probability of occurrence is low, the structure must be able to withstand these extreme conditions. An allowance given to account for this low probability of occurrence is by increasing the allowable stresses by 33 % (API-RP2A).
The environmental forces are discussed below. .
1. Wind
Wind forces are exerted upon that portion of the structure that is above the water, as well as on any equipment, deck houses, and derricks that are located on the platform. The wind speed may be classified as: (a) gusts that average less than one minute in duration, and (b) sustained wind speeds that average one minute or l onger in duration. Wind data should be adjusted to a standard elevation, such as 33 feet (10 meters) above mean water level, with a specified averaging time, such as one hour. Wind data may be adjusted to any specified averaging time or elevation using standard profiles and gust factors. API-RP2A recommends the following formulation for calculation of wind speeds above the reference elevation.
For strong wind conditions the design wind speed u ( z, t ) (ft/s) at height z (ft) above sea level and corresponding to an averaging time period t (s) [where t < t o; t o = 3600 sec] is given by: u( z, t ) = U ( z) × [1 – 0.41 × Iu( z) × ln(t/t o )]
where the 1 hour mean wind speed U ( z) (ft/s) at level z (ft) is given by: U ( z) = Uo × [1 + C × ln( z/32.8 )] -2
1/2
C = 5.73 × 10 × (1 + 0.0457 × Uo)
and where the turbulence intensity Iu( z) at level z is given by: Iu( z) = 0.06 × [1 + 0.0131 × Uo] × ( z/32.8 )
-0.22
where Uo (ft/s) is the 1 hour mean wind speed at 32.8 ft.
Mar 07 Page 11 of 110
Once the wind speed at the desired elevation has been computed the wind force at this elevation may be computed as: 2
F = (ρ/2)u C s A
where F = wind force, 3
3
ρ = mass density of air, (slug/ft , 0.0023668 slugs/ft for standard temperature and
pressure), u = wind speed (ft/s), C s = shape coefficient, 2
A = area of object (ft ).
API-RP2A [2] distinguishes between global and local wind load effects. For the first case it gives guideline values of mean 1-hour average wind speeds to be combined with extreme waves and current. For the second case it gives values of extreme wind speeds to be used without regard to waves. Wind loads are generally taken as static. When, however, the ratio of height to the least horizontal dimension of the wind exposed object (or structure) is greater than 5, then this object (or structure) could be wind sensitive. API-RP2A requires the dynamic effects of the wind to be taken into account in this case and the flow induced cyclic wind loads due to vortex shedding must be investigated.
2. Waves
The wave loading of an offshore structure is usually the most important of all environmental loadings for which the structure must be designed. The forces on the structure are caused by the motion of the water due to the waves which are generated by the action of the wind on the surface of the sea.
The resulting waves are irregular in shape, vary in height and length, and may approach a platform from one or more directions simultaneously. For these reasons
Mar 07 Page 12 of 110
the intensity and distribution of the forces applied by waves are difficult to determine. Because of the complex nature of the technical factors that must be considered in developing wave-dependent criteria for the design of platforms, experienced specialists knowledgeable in the fields of meteorology, oceanography, and hydrodynamics should be consulted.
In those areas where prior knowledge of oceanographic conditions is insufficient, the development of wave-dependent design parameters should include at least the following steps: 1. Development of all necessary meteorological data including wind profiles etc. 2. Development of operating sea-states. 3. Development extreme sea-states consistent with geographical limitations. 4. Bathymetric effects like water depth, seabed slope etc. also play a role in jacket design and these must be included in the basis of design.
Once the preliminary work has been completed the wave forces are generated in two steps.
The first step is to compute the sea state using an idealisation of the wave surface profile and the wave kinematics given by an appropriate wave theory. The second is the computation of the wave forces on individual members and on the total structure, from the fluid motion.
Two different analysis concepts are used: 1) The design wave concept, where a regular wave of given height and period is defined and the forces due to this wave are calculated using a high-order wave theory. Usually the 100-year wave, i.e. the maximum wave with a return period of 100 years, is chosen. No dynamic behaviour of the structure is considered. This static analysis is appropriate when the dominant wave periods are well above the period of the structure. This is the case of extreme storm waves acting on shallow water structures.
Mar 07 Page 13 of 110
2) Statistical analysis on the basis of a wave scatter diagram for the location of the structure. Appropriate wave spectra are defined to perform the analysis in the frequency domain and to generate random waves, if dynamic analyses for extreme wave loadings are required for deepwater structures. With statistical methods, the most probable maximum force during the lifetime of the structure is calculated using linear wave theory. The statistical approach has to be chosen to analyze the fatigue strength and the dynamic behaviour of the structure. Wave theories used in concept 1 mentioned above are discussed in the next section.
2.1
Wave Theories
Wave theories describe the kinematics of waves of water on the basis of potential theory. In particular, they serve to calculate the particle velocities and accelerations and the dynamic pressure as functions of the surface elevation of the waves.
Wave loading on members results from the water particles having finite velocities and the jacket structure being an obstruction in the path of these water particles. Wave theories predict these water particle velocities al ong with other variables like dynamic pressure etc. In general all these theories solve the general problem of finding water particle velocities by using a velocity potential which has to satisfy a Laplace equation and appropriate boundary conditions at the seabed, at the structural body face, at the free surface of the sea and the radiation condition at infinity.
th
Airy’s linear theory, Stokes 5 order theory, the solitary wave theory, the cnoidal theory, Dean's stream function theory and the numerical theory by Chappelear are some of the well known wave theories. These theories of varying complexity, developed on the basis of simplifying assumptions, are appropriate for different ranges of the wave parameters. They vary in their treatment of the free surface condition (for example, where the pressure is atmospheric). Linear theory (Airy) satisfies conditions to a first approximation. Stoke's higher order theories (and others) solve the problem to a second order (i.e., the parameter wave amplitude /
Mar 07 Page 14 of 110
wave length ratio would involve squares). For example, combining the first and second order solutions for wave surface elevation makes the wave crests steeper and troughs shallower. Suitable wave theories can be selected based on wave steepness and relative depth (both non-dimensionalised).
For the selection of the most appropriate theory, the graph shown in figure 6 may is typically used (as suggested in API-RP2A). Selection of the most appropriate theory depends on the water depth and the wave height under consideration. It must be mentioned though that Stokes fifth order theory is widely applicable to a range of water depths and wave heights and is the most popular theory being used for design of jackets.
Once the appropriate wave theory has been selected the forces on the jacket members can be computed from Morrison’s equation after accounting for effects like Doppler effect, effect of current on wave kinematics, wave spreading, current blockage factor, marine growth and conductor shielding factor etc. Consideration for all the effects mentioned are detailed in API-RP2A.
2.2
Wave Forces
Wave loads on submerged jacket members can be computed as a summation of drag loading and inertial loading. The equation summarizing these forces is called Morrison’s equation.
Drag loads are produced by flow separation on the down stream side of the member, creating a wake with reduced velocity, leading to local velocity gradients. Drag loading is proportional to the incident velocity squared.
Inertia loading is produced by pressure gradients in the accelerating fluid and is proportional to the acceleration. The velocities and accelerations refer to the orbital motion of the water particles within the waves as distinct from wave propagation.
Mar 07 Page 15 of 110
From API-RP2A Morrison’s equation for wave loading is given below
where F = hydrodynamic force vector per unit length acting normal to the axis of the
member, lb/ft (N/m), F D = drag force vector per unit length acting to the axis of the member in the plane
of the member axis and U , lb/ft (N/m), F I = inertia force vector per unit length acting normal to the axis of the member in
the plane of the member axis and αU /αt , lb/ft (N/m), Cd = drag coefficient, 3
w = weight density of water, lb/ft3 (N/m ), 2
g = gravitational acceleration, ft/sec2 (m/sec ), A = projected area normal to the cylinder axis per unit length (= D for circular
cylinders), ft (m), 2
V = displaced volume of the cylinder per unit length (= π D /4 for circular 2
2
cylinders), ft (m ), D = effective diameter of circular cylindrical member including marine growth, ft
(m), U = component of the velocity vector (due to wave and/or current) of the water
normal to the axis of the member, ft/sec (m/sec), |U | = absolute value of U , ft/sec (m/sec), Cm = inertia coefficient,
= component of the local acceleration vector of the water normal to the axis of 2
2
the member, ft/sec (m/sec )
The values of C d and Cm depend on the wave theory used, surface roughness and the flow parameters. According to API-RP2A, C D = 0.65 to 1.05 for smooth and
Mar 07 Page 16 of 110
rough conditions respectively and C M = 1.6 to 1.2 for smooth and rough conditions respectively. The total wave force on each member is obtained by numerical integration over the length of the member. The fluid velocities and accelerations at the integration points are found by direct application of the selected wave theory.
2.2
LOAD COMBINATIONS
This section describes the load combinations to be considered for inplace design. The load combinations should typically be those that will produce the most severe effects on the structure and consequently result in the highest stresses prosiible from among all the potential loads acting on the structure. Typically, these loading conditions are created by combining environmental conditions (wind and wave loads) with appropriate dead and live loads in the following manner. 1. Operating environmental conditions combined with dead loads and maximum live loads appropriate to normal operations of the platform. 2. Operating environmental conditions combined with dead loads and minimum live loads appropriate to the normal operations of the platform. 3. Design environmental conditions with dead loads and maximum live loads appropriate for combining with extreme conditions. 4. Design environmental conditions with dead loads and minimum live loads appropriate for combining with extreme conditions.
These load combinations are important not only in terms of global member designs but also play a very important part in deciding the reactions at the mudline level and the consequent pile sizes, depth of penetration into the soil and the capacity generated. The capacity generated with a given pile should be sufficient to be more than the maximum pilehead (mudline) reaction times a factor of safety. The factors of safety differ according the condition being analysed and are outlined below
Mar 07 Page 17 of 110
Load Condition
Factor of Safety
1. Design environmental conditions with appropriate drilling loads
1.5
2. Operating environmental conditions during drilling operations
2.0
3. Design environmental conditions with appropriate producing loads
1.5
4. Operating environmental conditions during producing operations
2.0
5. Design environmental conditions with minimum loads (for pullout)
1.5
Environmental loads, with the exception of earthquake load, should be combined in a manner consistent with the probability of their simultaneous occurrence during the loading condition being considered. Earthquake load, where applicable, should be imposed on the platform as a separate environmental loading condition.
The operating environmental conditions should be representative of moderately severe conditions at the platform. They should not necessarily be limiting conditions which, if exceeded, require the cessation of platform operations. Maximum live loads for drilling and production platforms should consider drilling, production and workover mode loadings, and any appropriate combinations of drilling or workover operations with production.
Mar 07 Page 18 of 110
3.0
DESIGN FOR PRE-SERVICE CONDITIONS
These loads are temporary in nature and arise during fabrication and installation of the jacket, deck or modules on the deck.. Typically during the fabrication and erection phases of the project, lifts of various st ructural components generate lifting forces which are transferred to the structure. In the installation phase forces are generated during platform loadout, transportation to the site, launching and upending, as well as during lifts related to installation. These too are transferred on to the structure. It must be noted that the jacket and deck structures are designed for inplace conditions. This does not automatically guarantee that the structural members will be able to withstand the forces imposed on them during the fabrication, erection and installation phases of the project. Re-design may be necessitated if structural failures are indicated in the analysis carried to simulate the pre-service conditions. The next section describes some of the preservice conditions and the forces that they impose on the structure.
3.1
Lifting
Lifting forces generated and consequently imposed on the jacket structure depend on the weight of the structural component being lifted, the number and location of lifting eyes used for the lift, the angle between each sling and the vertical axis and the conditions under which the lift is performed. All members and connections of a lifted component must be designed for the forces resulting from static equilibrium of the lifted weight and the sling tensions. A typical offshore lift operation is shown in figure. 7. A first look at this static equilibrium condition may suggest that the forces are only vertical in nature. However, API-RP2A recommends that in order to compensate for any side movements, lifting eyes and the connections to the supporting structural members should be designed for the combined action of the st atic sling load and a horizontal force equal to 5% this load, applied perpendicular to the padeye at the centre of the pin hole.
Mar 07 Page 19 of 110
All these design forces are applied as static loads if the lifts are performed in the fabrication yard. For offshore lifts, where the lifting derrick or the structure to be lifted is on a floating vessel, dynamic load factors should be applied to the static lifting forces. In particular, for lifts made offshore API-RP2A recommends two minimum values of dynamic load factors: 2.0 and 1.35. The first is for designing the padeyes and padeye connected members i.e. those members and their end connections framing into the joint where the padeye is attached. The second factor is for all other members transmitting lifting forces. For loadout at sheltered locations, the corresponding minimum load factors for the two groups of structural components become, according to API-RP2A, 1.5 and 1.15, respectively.
3.2
Loadout
Once the jacket, deck or module has been fabricated it needs to be moved from the yard to the transportation barge which will then transport it to the offshore location where it is installed. The loadout process leads to the generation of forces when the jacket is loaded from the fabrication yard onto the barge. Loadout could be carried out by: • Lift • Trailer • Skidding If the loadout is carried out by direct lift, then, unless the lifting arrangement is different from that to be used for installation, lifting forces need not be computed, because lifting in the open sea creates a more severe loading condition which requires higher dynamic load factors. Loadouts can also be carried out using trailers. The component being loaded out is lifted allowing the trailers to come underneath the component after which it is placed on the trailer bed. The load is distributed as evenly as possible with the use of loadout and
Mar 07 Page 20 of 110
spreader beams. During loadout lateral loads to the extent of 5% of the loadout weight are also applied to account for any trailer sway. A typical trailer loadout is shown in figure. 8.
If loadout is done by skidding the structure onto the barge, a number of static loading conditions must be considered, with the jacket supported on its side. Such loading conditions arise from the different positions of t he jacket during the loadout phases from movement of the barge due to tidal fluctuations, marine traffic or change of draft, and from possible support settlements. Since movement of the jacket is slow, all loading conditions can be taken as static.
For a skidded loadout a special launch truss with spacing running from deck fabrication position to quay side, with matching tracks on cargo barge has to be done. Additionally, skid shoes have to be provided on launch trusses. Typically these skid shoes are timber blocks secured by clips, block size sufficient to take deck leg reaction. A typical skidded loadout progression is shown in figure.9.
3.3
Transportation
Once the loadout process has been completed the next stage is the transportation to the offshore location on the transportation barges. These transportation barges are not self propelled but are towed by means of tug boats. During the transportation, inertial forces are generated when platform components (jacket, deck) due to the motion of the barge when it is being towed in the open sea.
The 6 motions ( viz. surge, sway, heave, pitch, roll and yaw) of a sea-vessel are shown in figure. 10. These motions are responsible for the generation of transportation forces which must be resisted by the structure being transported.
Mar 07 Page 21 of 110
The forces also depend upon the weight, geometry and support conditions of the structure (by barge or by buoyancy) and also on the environmental conditions (waves, winds and currents) that are encountered during transportation. Clearly transportation in the open sea poses a serious threat to the succesful completion of the project. To minimize the associated risks and secure safe transport from the fabrication yard to the platform site, it is important to plan the operation carefully by considering, according to API-RP2A the following: 1. Previous experience along the tow route 2. Exposure time and reliability of predicted "weather w indows" 3. Accessibility of safe havens 4. Seasonal weather system 5. Appropriate return period for determining design wind, wave and current conditions, taking into account characteristics of the tow such as size, structure, sensitivity and cost. Transportation forces are generated by the motion of the tow, i.e. the structure and supporting barge. They are determined from the design winds, waves and currents. According to API-RP2A, towing analyses must be based on the results of model basin tests or appropriate analytical methods and must consider wind and wave directions parallel, perpendicular and at 45° to the tow axis. Inertial loads may be computed from a rigid body analysis of the tow by combining roll and pitch with heave motions, when the size of the tow, magnitude of the sea state and experience make such assumptions reasonable. Note here that it may also be necessary sometimes to consider the flexibility of the barge deck and the effects it has on the forces generated on the structure being transported. Typically a stowage plan is prepared for the component being transported on the transportation barge. A barge response analysis is carried out using model tests or validated software to estimate the barge responses in terms of the pitch and roll angles
Mar 07 Page 22 of 110
and the heave acceleration values. Out of the 6 possible motions mentioned previously only these three are generally considered for the transportation analysis because they are periodic in nature and the other three viz. surge, sway and yaw are non-periodic. In the absence of any barge response study, the following conditions are specified for use in the transportaion analysis by Noble Denton. These are generally referred to as Noble Denton Criteria and depend largely on the length of the vessel on which the transportation is being carried out. These criteria are outlined in Table 1.
3.4
Launching and Upending
Once the jacket has been transported to the offshore location it is sitting on the transportaion barge. The temporary transportation fastening in the form of sea-fasteners and lashings is removed. The next objective is to launch the jacket into the open sea after which it is straightened and released so that it can sit vertically on the sea-bed. There are five stages in a launch-upending operation (shown schematically in figure 11): a. Jacket slides along the skid beams b. Jacket rotates on the rocker arms c. Jacket rotates and slides simultaneously d. Jacket detaches completely and comes to its floating equilibrium position e. Jacket is upended by a combination of controlled flooding and simultaneous lifting by a derrick barge. The loads, static as well as dynamic, induced during each of these stages and the force required to set the jacket into motion can be evaluated by appropriate analyses, which also consider the action of wind, waves and currents expected during the operation. To start the launch, the barge must be ballasted to an appropriate draft and trim angle and subsequently the jacket must be pulled towards the stern by a winch. Sliding of the jacket starts as soon as the downward force (gravity component and winch pull) exceeds
Mar 07 Page 23 of 110
the friction force. As the jacket slides, its weight is supported on the two legs that are part of the launch trusses. The support length keeps decreasing and reaches a minimum, equal to the length of the rocker beams, when rotation starts. It is generally at this instant that the most severe launching forces develop as reactions to the weight of the jacket. During stages (d) and (e), variable hydrostatic forces arise which have to be considered at all members affected. Buoyancy calculations are required for every stage of the operation to ensure fully controlled, stable motion.
To summarize, design of Fixed Jacket type Offshore structures requires several analyses to be carried out. These can be summed into two major categories: 1. Pre-Service Analyses 2. In-Service Analyses
Pre-Service Analyses includes analysing the structure for various conditions during the fabrication, load-out and installation of the structure including the Jacket and Deck structures.
The In-service analyses covers the conditions that the structure would be subjected to during its actual operation life at Offshore.
The various analyses required are as follows:Pre-Service
In-Service
Onshore Lift Analysis
Inplace Analysis with PSI
Load-out Analysis
Modal Analysis
Transportation Analysis
Fatigue Analysis
Launch & Upending Analysis
Seismic Analysis
Offshore Lift Analysis
Vibration Analysis
On-bottom Stability Analysis
Mar 07 Page 24 of 110
In addition to the above, there are some local member designs required for boat impact loads, accidental dropped object loads etc. The Global Structural analysis for Fixed Offshore structures is carried out using SACS software which has various modules to carry out all of the SACS above mentioned analyses.This report discusses the steps to be followed to complete the above global analyses using SACS software.
Note: In the subsequent discussion many files are followed with a .* extension (ex. sacinp.*, psiinp.* and so on. We are used to a file naming convention where whatever follows after the ‘.’ denotes the file type (ex a letter.doc indicates a word document file, num.xls a Excel file and so on. SACS file naming works in the opposite way. Whatever precedes the ‘.’ indicates the file type (ex. sacinp.* indicates a sacs input file, psiinp.* indicates a pile soil interaction input file and so on). In SACS the ‘*’ indicates the label (filename) that the user can specify. The location of this label input in SACS Executive is shown below.
Label Subsequent sections discuss each of the analyses mentioned above in detail. .
Mar 07 Page 25 of 110
4.0
INPLACE ANALYSIS WITH PSI
Files required: sacinp.*, psiinp.* and jcninp.*
File Details: Sacinp.* 1. Define the basic Jacket framing using the Jacket wizard in Precede-Pro which defines the leg batter, conductor spacing and leg and/or skirt pile details. A new model file is created by clicking on Model>Create New Model in the Interactive Window of SACS Executive and then entering the required data using the wizard.
2.
Title this input file as ‘sacinp.*’ where * is the user-defined title in a working folder titled Inplace. The file contains the model data, member properties defined using Section and Groups. It also contains all the load data, wind areas, Cd, Cm values and Marine growth. This information is entered using Precede and DataGen. Note that the sacinp.* input file may be opened in Precede or DataGen by rightclicking on the input file in SACS Executive.
3.
The first line of this file is the load options line LDOPT. The value for mudline elevation in the LDOPT line is to be taken as the depth marked in the bid document/drawings. With reference to the RS-2 jacket, this value marked in the bid drawings is 80.85m. Also note this value is entered as a negative value
4.
The value for the water depth in the LDOPT line is calculated as follows
Still Water Depth = CD + LAT + (50% of Astronomical Tide) + (Storm Surge)
For purposes of this calculation use the mudline depth (absolute value) entered in step 4 above as the value of CD (Chart Datum). The values for LAT (Lowest Astronomical Tide), Astronomical Tide and Storm Surge are obtained from the bid document.
Mar 07 Page 26 of 110
With reference to the RS-2 jacket S.W.D = 80.85+(-0.183)+0.5*(3.66)+0.61 = 83.107 (rounded to 83.11m)
psiinp.*
1. The Pile and soil properties are specified separately in the ‘psiinp.*’ file. A new psiinp.* file is created by clicking on Data File>Create New Data File>Static>Pile Soil Interaction>Select in the Interactive Window of SACS Executive.
2. A title may be entered in the next window generated.
3. The next pop-up window is the PSI options window. The following options are entered under the ‘General’ tab: +Z as the vertical co-ordinate, Units: MN, Displacement and Rotation Convergence requirement as 0.005, EX (KPGI preference) or CB(Valdel preference) and 100 maximum iterations. Under the ‘Output Options’ tab enter number of pile increments as 100 and density as 7.85MT/M^3. Click the ‘Finish’ button after this. Note that the number of pile increments is nothing but the number of locations along the length of the piles where a unity check will be done.
4. The Pile Groups are defined next specifying the pile section dimensions and properties (Edit Line>Insert Line>PLGRUP(Pile Group)>Select>Yes(for the header). The next pop up window is the Pile Group definition window. Note that a single pile may have a varying cross section and/or varying material properties. This variation is accounted for by defining different section sizes and/or material properties under the same group label. For example, for the RS-2 jacket skirt pile 1 (reference: Pile drawing Number 1238-RS2-6102-L009 REV 0) is defined in three segments with varying wall thickness (5cm,3.8cm and 5 cm) and different segment lengths (5m, 95m and 2 m respectively). To create the first segment (of length 5 meters) shown in bold below enter the following options in the Pile Group definition window (OD=182.9cm,Wall thickness=5cm, segment length = 5
Mar 07 Page 27 of 110
m,Elastic modulus=20.5 (1000 KN/cm^2),shear modulus=8 (1000KN/cm^2),yield stress=34.5 KN/cm^2). Note that these values mentioned are just representative values and will vary. Clicking OK will create the first segment of group PL1
PLGRUP PL1
182.90 5.00 20.5 8.000 34.50
5.
PLGRUP PL1
182.90 3.80 20.5 8.000 34.50
95.
PLGRUP PL1
182.90 5.00 20.5 8.000 34.50
2.
2.624
To create the additional two segments (input lines 2 and 3) right click mouse on the next line in DataGen > Insert line> PLGRUP(Pile Group)>Select and then enter the appropriate details. Once all 3 lines have been created pile group PL1 is completely rd
defined The last number in the 3 line (2.624) refers to the end bearing area. This end bearing area is calculated assuming that the pile is plugged (i.e. soil has entered the pile annulus). This end bearing area is defined only for the last 2 meter segment of the pile group PL1. The other pile (and their segments are defined similarly)
An initial pile penetration depth needs to be assumed. This was specified by way of the segment lengths. In case of the RS-2 jacket by specifying 3 segments of lengths 5 m, 95 m and 2 m we are assuming that the initial pile penetration is a total of 102 meters.
5. Next,
define
the
pile
using
the
PILE
card
(Edit
Line>Insert
Line>Pile(Pile)>Select>Yes(for the header). In the pile pop-up window, specify the PILEHD joint to which the pile is attached, the pile group label which is to be used, the joint to be used to specify the pile batter and the soil table ID to be used. For example
PIL JB5 SK1 PIL1
SOL2
defines a pile attached to the PILEHD joint JB5 (located at the mudline level), having a batter (slope) defined as the batter between the joints JB5 and SK1 with
Mar 07 Page 28 of 110
pile section properties defined by the PILGRUP P L1 and using soil table ID SOL2. Note that soil properties have not been defined yet but will be in subsequent steps.
The remaining piles are defined similarly.
6. The soil properties are obtained from a geo-technical report supplied along with the bid document. Here we refer to the IEOT geotechnical report. All soil properties should be defined in the order of axial (T-z data), axial bearing(Q-z data), torsional and lateral stiffness (P-y data).
7. Note: IEOT soil data is obtained by carrying out a std. pile penetration test using a pile with OD 1.372m and wall thickness 0.051m.
a) The IEOT T-z data is reported with row headings ‘c’ and ‘t’ implying compression and tension values for the test pile. Caution: The IEOT data are readings taken for the std. test pile and not for the soil. What we have to specify are not test pile readings but soil properties. Take for example the T-z value in the ‘t’ row at a depth of 3.80 m under the columns ‘t4’ and ‘z4’ in the IEOT report. These values are mentioned as 0.03 and 10.3, implying that a tensile force of 0.03 MN was required to create a displacement of 10.3 mm at the pilesoil interface. A tensile force on the pile will be a compressive force on the soil. Hence compressive values for the pile translate to tensile values for the soil and tensile values for the pile translate to compressive values for the soil.
To enter the T-z soil data do the following. In DataGen Edit Line>Insert Line>Soil T-z axial Head (T-z Soil Axial)>Select>Yes(for the header) . In the T-z soil axial pop window enter the following information: the IEOT report provides data for 32/33 soil strata. Z-factor (explained later), soil table ID. The following line gets created in DataGen
SOIL TZAXIAL HEAD 32
0.1SOL2T-Z AXIAL
Mar 07 Page 29 of 110
b) When the T-z values in both compression and tension are the same absolute values, the soil card can be entered as follows: In DataGen Edit Line>Insert Line> T-z Soil SLOC T-z(T-z Axial Stratum)>Select, For example to create the soil properties at a depth of 0.0m, enter the following information in the T-z Axial Stratum window: Check the Symetrical T-z option, No. of points on T-z curve = 6, dist. To top of stratum = 0.0 and T-factor = 0.0232 (this is explained later). SOIL T-Z
SLOCSM 6 0.00
0.0232CLAY
SOIL
T-Z
0.0 0.0 0.0 1.7 0.0 3.4 0.0 6.9 0.0 10.3
SOIL
T-Z
0.0 13.7
Note: The IEOT report gives only 5 data (T-z) points at each depth. Yet we have entered number of points on T-z curve as 6. This is because at each strata we have to enter an additional T-z point = (0.0,0.0) implying that for a compressive/tensile force = 0.0 MN the displacement at the pile soil interface = 0.0 mm)
c) In case the compressive and tensile values are different, then the data needs to be input in separate SOIL cards (In DataGen Edit Line>Insert Line> T-z Soil SLOC T-z(T-z Axial Stratum)>Select). For example, the IEOT report has two T-z sets at a soil depth of 6 meters. To account for different properties at the same depth, an additional SOIL card is created with a ‘distance to stratum’ value of 6.001m. Additionally, in this second set, we notice different values of T-z in compression and in tension. Note the first T-z value in the compression set is z=1.7mm for a compressive force of T=0.002 MN; in the tensile set the first T-z value is z=1.7mm for a tensile force of T=0.001 MN.
d) To account for the different compressive and tensile behaviour of the soil, additional SOIL cards are created. F or example
Mar 07 Page 30 of 110
SOIL T-Z
SLOC
11 6.001
0.0232SAND
SOIL
T-Z -0.002 -13.7-0.002 -10.3-0.002 -6.9-0.002 -3.4-0.001 -1.7
SOIL
T-Z
SOIL
T-Z 0.003 13.7
0.0 0.0 0.002 1.7 0.002 3.4 0.003 6.9 0.003 10.3
The second line has negative values for both T and z implying compressive properties of the soil at this depth. These values are obtained from the ‘t’ row of the T-z soil data in the second 6 m depth set. For example the ‘t’ row shows a value of 0.002 13.7 under the colums t5 z5 in the IEOT report. These values are tensile properties for the test pile and translate to compressive properties for the soil. Hence they are input as negative values in the SOIL card. Here leaving the ‘Symmetrical T-z option’ unchecked will create the SLOC implying that soil properties in compression and tension are different.
Compressive values of T-z for the test pile in the IEOT report are input as positive values values in the SOIL card to imply tensile behaviour.
In addition, one extra data point of z = 0.0 at T = 0.0 is created in the SOIL card, hence number of points are specified as 11 ( 5 for compressive data, 5 for tensile data and this additional (.0,0.0) point.
Also note that SACS allows only 5 T-z points to be entered at a time. To create the additional lines needed to specify the remaining 6 data points use Edit Line>Insert Line> Soil T-z (T-z Axial)>Select
e) The Bearing data is entered at various depths from the Q-z data of the IEOT report. Note that the Q-z data in the IEOT report is called T-z Bearing in SACS. To create this set use the following commands in DataGen in a manner similar to the T-z data.
Mar 07 Page 31 of 110
For the Q-z data first use Edit Line>Insert Line>Soil Bearing Head (T-z Axial Bearing)>Select. Here we specify the number of strata for which IEOT provides Q-z data and the Z-factor (explained later).
On the next line use Edit Line>Insert Line>Soil SLOC BEAR (T-z Axial Bearing Strata)>Select. Here we specify the number of points on the T-z Bearing curve. The IEOT report provides 5 data points and we have to create an additional data point of (0.0,0.0). Hence we specify number of points on curve as 6. Also provide the distance to the top of the stratum and the T-factor (explained later). On the next line use Edit Line>Insert Line>Soil T-z (Axial Bearing T-z)>Select to enter the required data.
As with the T-z data, SACS allows only 5 T-z Axial Bearing data points to be entered at a time. To input additional data use Edit Line>Insert Line>Soil T-z (Axial Bearing T-z)>Select
f) For the lateral P-y use the following commands in DataGen in a manner similar to the T-z data.
For the P-y data first use Edit Line>Insert Line>Soil Lateral Head(Lateral Soil)>Select. Here we specify the number of strata for w hich IEOT provides P-y data, the y-factor (explained later), soil table ID and the reference pile diameter. (this is the only place in SACS where a mention of the standard test pile size used to collect soil data is asked for by SACS)
On the next line use Edit Line>Insert Line>Soil SLOC P-y (Lateral Soil Stratum)>Select. Check the Symmetrical P-y option, specify the number of points on the T-z Bearing curve. The IEOT report provides 4 data points and we have to create an additional data point of (0.0,0.0). Hence we specify number of
Mar 07 Page 32 of 110
points on curve as 5. Also provide the distance to the top of the stratum and the P-factor (explained later).
On the Next line use Edit Line>Insert Line>Soil P-y(Lateral P- Y) > Select to enter the P-y values
g) Factors used: In the previous few paragraphs a mention was made of the factors to be used while specifying the soil properties These factors account for the fact that the IEOT geotechnical report uses units different from those needed as SACS input as well as some scaling which needs to be performed. The scaling is required since the IEOT data is obtained using a standard size pile and we need to scale this data to the actual pile size to be used. Calculations of these factors are provided below.
T-Z FACTORS
T Factor: Unit Conversion: MN to KN/cm2 T-factor:
1000/ (pi*Dref*100)
Dref = 137.2 cm; therefore T-factor = 0.0232 Remarks: The ‘T’ values are not scaled according to the diameter; only unit conversion is done.(values depend on the actual surface area.)
Z-Factor: Unit conversion: mm to cm Z-factor:
0.1*Dact/Dref
For RS-2, Dact = 182.9 cm, Dref = 137.2 cm, therefore Z-factor = 0.133 Remarks: The deformation values are scaled in proportion to the diameter; and the value 0.1 is introduced for unit conversion.
Mar 07 Page 33 of 110
Q-Z FACTORS
Q-factor: Unit Conversion: MN to KN/cm^2 Q-factor:
1000/(pi/4*Dref^2) = 0.067 (for RS-2)
Remarks: The ‘Q’ values are not scaled; instead unit conversion is performed. (The values depend on the actual bearing area.)
Z-factor Unit conversion: mm to cm. Z-factor:
0.1*Dact^2/Dref^2 = 0.177(for RS-2)
Remarks: The deformations are scaled according to the cross-sectional areas; which in turn is proportional to the square of the respective diameters. This is true in case of only plugged piles. The factor of 0.1 is introduced for unit conversion.
P-Y FACTORS P-factor Unit Conversion: MN to KN/per running meter length of pile (KN/cm) P-factor: 1000/100=10 Remarks: Unit conversion is performed.
Y-factor Unit Conversion: mm to cm. Y-factor: 0.1*Dact/Dreq = 0.133 (for RS-2) Remarks: Deformations are scaled according to the diameters. The factor 0.1 is for unit c Remarks about the soil data: a) The jack up rig is present on the north face of the jacket (row 1, or the face where the jacket legs do not have a batter). Due to absence of any batter on the jacket legs, the leg pile/skirt piles will also not have any batter. i.e. piles PL1
Mar 07 Page 34 of 110
and PL2 are the piles without any batter. Due to operation of the jack up rig some scouring of sea bed will occur at the north face. Scouring is nothing but the removal of sea bed mud. In case of the RS-2 jacket, scour depth was specified as 3 meters in the bid document. Therefore the piles on the north face, for the first 3 meters, there will be no soil present due to scouring and hence no pile-soil interaction.
b) To account for this two soil tables have been created in the RS-2 psiinp.* file i.e. SOL1 and SOL2 respectively. Piles with section properties PL1 and PL2 use SOL2 for the PSI analysis
c) The difference in the two tables is that in SOL2 for the T-z data an additional card is created for a depth of 3 meters. However the IEOT report gives no soil T-z data for a depth of 3 meters. It provides data a 0.0 m depth and then directly at 3.80m depth. To get around this, the same soil T-z data provided by the IEOT report for 0.0 m depth is used to specify the soil properties at a 3 m depth. By doing this we are simply saying that in the soil stratum between 0.0 and 3.0 meters there is no variation in T-z properties. Note that this addition is done in the SOL2 table only and the piles having section properties PL1 and PL2 (north face piles) use SOL2 for PSI
d) A change also needs to be made in the P-y data of the SOL2 table to reflect the absence of soil upto a depth of 3.0 m for the north face piles. One way is to make the P values upto a depth of 3.0 m equal to zero leaving the displacement values the same as those in the IEOT report. (this is what KPGI has done). John Brown leave the P-values upto scour depth as in the IEOT report but make the displacement (y) values equal to zero.
e) In the RS-2 psiinp.* for ‘some’ reason KPGI has chosen to make the P-values upto the scouring depth (3.0m) equal to zero in both soil tables SOL1 and SOL2. This does not make sense since SOL1 refers to the soil around the piles
Mar 07 Page 35 of 110
with batter (PL3 and PL4) where no scouring occurs and hence PSI takes place from mudline upto the pile penetration depth unlike the north face piles where scouring causes PSI to occur only between scouring depth and pile penetration depth. For our purposes we will make the changes in the P-y data for only one of the soil tables and not both.
f) Another strange modification that KPGI has done is that in the SOL2 table between and depth of 3.0 to about 6.0 m, the P-values entered in the psiinp.* file are exactly half those specified in the IEOT report while displacements are entered as is. This implies that the soil stiffness is halved between 3.0 m and 6.0 m for no apparent reason. For our purposes other than the modifications to reflect scouring on the north face we will enter the data as is.
g) As far as the bearing data is concerned, KPGI for some reason has entered the Q-z bearing data only from a depth of 62m and not from 60.5m as is provided in the IEOT report. Again we will enter the data as is.
h) Lastly IEOT does not provide a torsional constant. For future projects we will use the value used in the RS-2 and hard code it into the psiinp.* file.
jcninp.* 1. The last input file to be created is the Joint Can input file where the load cases for which the joint can unity check need to be performed are specified The SACS model created is a center-line model where the tubulars are modelled as beam elements. In reality, these beam elements are tubulars and a stress calculation needs to be done not at the center-lines but at the tubular walls. This joint can input file specifies the load cases for which the joint can unity check is performed
2. The jcninp.* file is created as follows: Data File>Create New Data File>Post>Joint Can. A minimum gap of 5.1 cm and a maximum gap of 100 cm is created (according to API) and a brace on brace check is performed. The final few lines in
Mar 07 Page 36 of 110
this input file are used to increase the allowable stress (using the AMOD command) by 33% for the load cases ‘Extreme Storm with blanket loads’ and ‘Empty with Extreme Storm’. API allows this increase in load cases reflecting extreme storm cases.
3. Note that that the joint capacity should be atleast 50% of the member strength rd
(Reference Pg.46 API-RP2A) or a 2/3 over-ride should be used (in case 2/3
rd
of
tensile strength is less than the yield strength). For information on the QU factor refer to pg 49 API-RP2A)
4. The last line in the jcninp.* file is the RELIEF command. While modelling we ensure that the tubular beams are modelled with offsets. However in case we have omitted specifying the offset for any beam then putting the RELIEF statement ensures that SACS will do the code check for the tubulars at the walls and not the centreline.
ANALYSIS PROCEDURE 1.
Select ‘Linear Static with Pile Soil Interaction’ option in the Runfile Wizard and select all the above 3 files in the appropriate sections with the following options selected in the Analysis Options window Foundation •
Do not create pile fatigue solution
•
Do not create a foundation superelement
Element Check •
Perform element check
•
WSD AISC 9 (for the deck beams)
•
API 21 (for tubulars)
th
st
PostVue •
Create PSVDB
Joint Check •
Perform tubular joint check
Mar 07 Page 37 of 110
•
Select the jcninp.* file here
Reports •
Override model, Check Joint rection, end forces and UC ranges
Run the analysis.
Output files: psilist.*, seoci.*, psicsf.*, psvdb and psi.run
Checks: Psilst.* file: The output file ‘psilst’ specifies the pile capacities mobilised versus the capacity required and gives a pile UC ratio value. •
F2 – Error/warnings to find errors and warnings
•
F2 - Relative – to find dead weight (preferred way is to go to PrecedePro > Load > Selfweight > for information only)
•
Pile maximum axial capacity summary. Check pile safety factors in compression and tension. F.S >= 1.5
•
Check pile UC ratios. These must be <1.0 for all cases and <1.33 for extreme storm cases
Open the ‘psvdb’ folder and check the member UC ratios graphically. The joint can UC ratios need to be checked in the ‘saclst’ file.
Mar 07 Page 38 of 110
5. 0 SINGLE PILE ANALYSIS
This is not strictly a separate analysis but it is as a pre-cursor in both Dynamic as well as Seismic analysis. In dynamic and seismic analysis the first step it to carry out a modal analysis to determine natural frequencies and mode shapes of the structure. When doing this we simplify the non-linear pile soil interaction by replacing the Pile-soil subsystem with an equivalent pile stub that would give the same reactions at the PileHead joints as the actual pile soil subsytem (this procedure is called single pile analysis). This pile stub is then modelled and the structure natural frequencies and mode shapes estimated. The pilehead joints are those through which the linear structure is connected to the non-linear system. The stiffness and the load matrices of the linear structure are condensed down to the pilehead joints in order to account for the effects of the linear structure in the nonlinear pile-soil interaction analysis5.
The single pile analysis procedure is described below. Here the reference files are not for the RS-2 jacket but for the Bunduq platform located in the following folder (Z:\MECHANICAL
ENGINEERING
\
OEG
\
JB'sData
\
Disc4-Offshore
\
Bunduq(Part3) \VA SACS Analysis \ Rev 5 (For Report)). Refer to the sacinp.sta.gip_va file in particular
1. In a new folder called Single Pile Analysis make a copy of the sacinp.* file and the psiinp.* file
2. As mentioned above we intend to replace the pile soil sub system with an equivalent pile stub which will give the same reactions at the pilehead joints. The pilehead joint reactions in question here are those that are generated when the operating storm loads are imposed on the structure since this represents closely the true behaviour of the piles during actual operation.
3. In case of the Bunduq platform the operating storm loads will be considered for deciding the pile stub-lengths (open the sacinp.sta.gip_va file - in DataGen and search for the string LCOMB using the F5 key).The operating storm loads consist of the dead loads
Mar 07 Page 39 of 110
produced by structural components, crane loads and the directionally dependent wind and 1 year wave + current loads and produce directionally dependent reactions at the pilehead joints. In the Bunduq sacinp.* file these load combinatations are labelled as OX1, OX2 …, OY1, OY2…,OA1, OA2…. & OB1, OB2 etc.
4. For example one load combination OX1 combines all the permanent loads (loadcn 1 through 16) with the wind (loadcn 17) and 1 year wave+current loads (loadcn 85) ○
○
oriented at 0 (0 orientation implies that the wind or wave load is moving from the –X axis towards +X axis). LCOMB OX1
1 1.100 2 1.120 3 1.120 5 1.100 60.8625 70.8250
LCOMB OX1
8 1.000 9 1.000 10 1.100 110.8625 120.8250 130.8250
LCOMB OX1 14 1.100 160.8250 17 1.000 85 1.140
5. Similar combinations are created in the sacinp.* file for all the other directions.
6. The pile stub lengths are decided based the pilehead joint reactions to the loads created in the previous steps. The LCSEL command is used to specify the load cases which are to be analysed. This is specified at the beginning of the sacinp.* file (after the OPTIONS card). For example LCSEL ST OA1 OA2 OA3 OA4 OA5 OA6 OA7 OA8 OB1 OB2 OB3 OB4 LCSEL ST OB5 OB6 OB7 OB8 OX1 OX2 OX3 OX4 OX5 OX6 OX7 OX8 LCSEL ST OY1 OY2 OY3 OY4 OY5 OY6 OY7 OY8
specifies that load cases OA1-OA8, OB1-OB8, OX1-OX8 and OY1-OY8 are to be analysed. Note that in the inplace sacinp.sta.gip_va file many more load cases are analysed. For this analysis only the operating storm load cases are highlighted.
7. The next step is to do a Linear Static Analysis with Pile Soil Interaction. In SACS Executive click Runfile Wizard>Static and then scroll down to Linear Static Analysis with Pile Soil Interaction. Select the sacinp.* and psiinp.* file and use the following options Foundation • Do not create pile fatigue solution
Mar 07 Page 40 of 110
• Do not create a foundation superelement
Superelement • Import Option>No Superelement
Reports • Override model, Check Joint rection, end forces and UC ranges
Run the analysis.
8. The output file is a psilist.* Double click to open this file and search for the string “Pilehead Forces” by using the F2 key in the editor.
9. The output specifies the pilehead forces in terms of the axial force, lateral force, bending moment and pilehead displacements.
10. For each pile note down the load case creating the maximum lateral force (i.e. shear force) at the pilehead. For example in case of the pilehead 001P for the pile PL1 load case OA6 yields a axial force of -7212.06 KN, a lateral force of
267.32 KN, a bending
moment of 153.4 Kn-m and an axial deflection of 0.35 cm
11. Similarly the maximum lateral forces for the other piles are 009P: axial force = -4883.84 KN, lateral force= 288.37 KN, BM=129.5 KN-m
axial
defl = 0.24 cm (load case OX2) 019P: axial force = -4110.93 KN, lateral force= 378.65 KN, BM=171.6 KN-m
axial
defl = 0.20 cm (load case OX7) 081P: axial force = -6945.34 KN, lateral force= 333.71 KN, BM=184.6 KN-m
axial
defl = 0.34 cm (load case OX4) 089P: axial force = -6614.11 KN, lateral force= 344.19 KN, BM=147.1 KN-m
axial
defl = 0.32 cm (load case OX3) 089P: axial force = -7259.84 KN, lateral force= 373.93 KN, BM=173.2 KN-m defl = 0.35 cm (load case OX3)
axial
Mar 07 Page 41 of 110
12. Next make a copy of the psiinp.* file and rename it as the pilinp.*.
13. Open the pilinp.* file in DataGen and make the following changes.
14. Remove the PSIOPT card. Insert a PLOPT card (Edit>Insert Line>PLOPT (Pile Options)>Select) and use the default options along with the following changes: Output Units: MN, Stress and Unity Check Option – NO, Plot option – NO, Pile Material Weight 3
Density – 7.85 MT/m , Number of length increments – 100, Max. Number of Iterations – 100, Deflection Convergence Tolerance – 0.001 and Stress Conc. F actor – None.
15. The original psiinp.* file defines section properties of all the pile groups and also defines all the piles (pile group, batter joint, soil table ID used by pile). To avoid scanning through a lot of output data for each pile at the end of the a single pile analysis analyze only one pile and estimate pilestub lengths one at a time for each pile.
16. In order to analyse only pile at a time the following steps must be ensured:
Take the Bunduq platform for reference. It has 6 piles associated with the pilehead joints 001P, 009P, 019P, 081P, 089P and 009P respectively. First we analyze only the pile associated with pilehead joint 001P. In the pilinp.* file comment out all the other pile definitions viz. 009P, 019P, 081P, 089P and 009P.
The pilinp.* file also defines a single soil table. In case two soil tables SOL1 and SOL2 are defined (as in the case of the RS-2 jacket) an additional step ensues. In the RS-2 jacket for example pile JB5 uses soil table SOL1. If we leave the soil table SOL2 as it is and run the single pile analysis then an error is generated that essentially says that more soil tables have been defined than used by any pile. For this reason also comment soil table SOL2. For the Bunduq platform this is not necessary.
17. The last step is to add a PLSTUB card at the end of the soil data, just before the END card (Edit>Insert Line>PLSTUB(Pile Stub Design)>Select). In this card we input the
Mar 07 Page 42 of 110
lateral force, bending moment, axial load and axial deflection noted down from the Linear Static Analysis with Pile Soil Interaction analysis.
18. In the ensuing Pile Stub Design Card check ‘F’ under the Force or Displacement Option, enter the PileStub joint name (001P), the PileStub lateral force (267.32KN), PileStub Moment (153.4 KN-m), Axial Load (7212.06KN) and axial displacement (0.35 c m), click OK and save the file.
19. In SACS Executive click on ‘Misc’ and then ‘Single Pile Analysis’ using the scroll bar. Click ‘Start Wizard’, select the pilinp.* file saved in the previous step and then click ‘Run’ to execute the Single Pile Analysis.
20. The output file is a pillst.* file. Open this file and search for the string ‘Stub Properties’. The pile stub length required is reported under as ‘Joint to Joint Length’. In case of the 099P pile (look in the file pillst.single_pile it is 338.83 cm.
21. Repeat steps 16 through 20 to find out pile stub lengths for all the remaining piles. Care must be taken to ensure that each pile being analysed makes use of the correct soil table. Also as before ensure that only one PILE card and its corresponding PLSTUB card are defined at a time.
22. Each of the pile stub lengths may be different. The one requiring the longest pile stub length is chosen as the stub length for all the piles. The different pile stub lengths estimated for the Bunduq platform are 001P: 384.65 cm 009P: 385.345 cm 019P: 387.574 cm 081P: 385.995 cm 089P: 385.926 cm 099P: 386.673 cm 23. This concludes the single pile analysis procedure.
Mar 07 Page 43 of 110
6.0
FATIGUE ANALYSIS
Fatigue failure of metals may be defined as the formation of cracks by repeated/reversal 1
application of loads, each of which, in itself, is insufficient to cause static failure . In offshore structures the major of cause of fatigue failure is the repeated wave loading on the structure. Detailed commentaries are available in API RP-2A and in the Technical notes for Structural Design of Offshore Jackets and Topsides.
There are two types of fatigue analysis Deterministic Fatigue Analysis Spectral Fatigue Analysis
In general the thumb rule on which type of analysis are to be used depend on the natural period of the offshore structure a) If natural period < = 3 seconds, use Deterministic Fatigue
b) If 3 seconds < natural period < 10 seconds, use Deterministic Fatigue with Dynamic Amplification Factors (DAF’s) c) If natural period > = 10 seconds, use Spectral Fatigue Analysis
The procedure for Deterministic Fatigue Analysis is outlined below. Files required: sacinp.*, ftginp.* and jcninp.*
ANALYSIS PROCEDURE This procedure is a step wise method in SACS of the general Deterministic Fatigue procedure 1. Create a new folder called fatigue. Make a copy of the inplace sacinp file. Rename the file as sacinp.fatigue.
2. Open the sacinp.fatigue file in DataGen by right clicking on it in SACS Executive and make following changes to this file:
Mar 07 Page 44 of 110
(a) In the LDOPT card change the Water Depth as per bid to account for absence of Storm Surge Still Water Depth = CD + LAT + (50% of Astronomical Tide) With reference to the RS-2 jacket S.W.D = 80.85+(-0.183)+0.5*(3.66) = 82.49 (rounded to 82.5m) Note that the mudline elevation value remains the same i.e. -80.85m
(b) Delete the LCSEL, HYDRO, HYDRO2, UCPART AND AMOD lines following the OPTIONS line in the sacinp.fatigue file. Also since the HYDRO and HYDRO2 lines have been removed, make sure that in the LDOPT line under Simplified Hydro Collapse tab, NOH has been checked so that no hydrostatic collapse check is performed. Not checking this option after deleting the HYDRO and HYDRO2 command lines will result in errors.
(c) The Cd (drag coefficient) and C m (Inertia coefficient) values are normally specified as follows2 Cd = 0.65, C m = 1.6 (for smooth members) Cd = 1.05, C m = 1.2 (for rough members)
While entering these values we increase C d by 7% (i.e. 0.696 for smooth and 1.123 for rough members) and C m by 4% (i.e. 1.664 for smooth and 1.248 for rough members). So the CDM card in the sacinp.inplace file specifies these values as shown below.
CDM
16.00 0.696
1.664
1.123
1.248
The 7% and 4% increase in the drag and inertia coefficients is to account for al l the minor appurtenances that we have not modelled. In this example, the first number immediately following the CDM command (16.00) is the diameter to which these values have to be applied. The CDM command occurs well into the sacinp.* file. The best way to locate is to search for it by pressing F5 in DataGen and searching for CDM.
Mar 07 Page 45 of 110
It is possible to specify these coefficients for each tubular separately using multiple commands. However the simplest method is as follows (Edit>Insert Line>CDM(Drag and Inertia Coefficient>Select). The lines below designate a drag coefficient of 0.696 (smooth) and 1.123 (rough) and a inertia coefficient of 1.664 (smooth) and 1.248 (rough) for all members having outer diameters between 16 cm and 250 cm CDM CDM
16.00 0.696
1.664
1.123
1.248
CDM 250.00 0.696
1.664
1.123
1.248
The diameters 16.00 cm and 250.00 cm are the lowest and the highest diameters that have been used in the RS-2 project and any members having diameters between (and including) these two extreme values will be assigned the said drag and inertia coefficient values.
In the RS-2 KPGI inplace sacinp.* file additional intermediate diameters (between 16 cm and 250 cm) have also been specified. As shown in the SACS manual this is unnecessary.
For fatigue analysis, the drag and inertia coefficients have to be modified as follows
2
Cd = 0.5, C m = 2.0 (for smooth members) Cd = 0.8, C m = 2.0 (for rough members)
These values are specified in RP-2A in the section on fatigue analysis Here again we have to account for all minor appurtenances which have not been modelled by increasing C d by 7% and C m by 4%. Hence the values input in the CDM card in the sacinp.fatigue file have to be changed to Cd = 0.535, C m = 2.08 (for smooth members) Cd = 0.856, C m = 2.08 (for rough members)
Mar 07 Page 46 of 110
(d) Delete ALL loads except wave loads (since these are the only cyclic loads we are considering for the fatigue analysis). To avoid any inadvertent specification of any loads, the best way to to delete ALL loads (including the wave loads, since it also includes specifications of current etc) and create the WAVE load cards anew.
(e) To create a new wave card first type a LOAD command on a new line.
(f) After this type LOADCN and then a number to identify the wave card
(g) Next use Edit>Insert Line>WAVE(Wave Generation)>Select (for the header) and then input appropriate data. This data is entered using the bid document (For the RS-2 project, Table 8:Environmental Parameters for Fatigue Analysis of the Structural Design Criteria). This table specifies the Wave Heights and their periods for 4 directions viz. S, SW, W and NW (a total of 26 values). Thus waves in this table approach the jacket from the geographic(true) S, SW, W and NW directions. We have to ensure that the wave loads are imposed in the correct directions in our model (which may have been modelled arbitrarily).
(h) In SACS wave approach directions are always specified (using the right hand rule) with respect to the Global (SACS) +X axis going in an anticlockwise sense with the +Z (SACS) axis along the thumb. For example if the wave direction (in degrees) in the WAVE card is specified as 0 degrees, it implies that the wave is moving along the +X (SACS) axis (i.e. the wave approaches from –X (SACS) axis and moves towards +X (SACS)
axis). If the wave
direction (in degrees) is specified as 270 degrees, it implies that the wave is moving along the –Y (SACS) axis (i.e. approaches from +Y (SACS) axis and moves towards the –Y (SACS) axis)
Mar 07 Page 47 of 110
For the RS-2 jacket, platform north and geographic (true) North coincide (see bid document). If the RS-2 sacinp.* file is opened in Precede it can be seen that the platform North face (i.e. the face where the jacket legs have 0 batter) has been modelled to the left, perpendicular to the –X (SACS) axis. Now a wave approaching from the geographic South (and for the RS-2 jacket, the platform South, since they coincide) should be moving from the +X (SACS) axis towards the –X (SACS) axis (i.e along the –X (SACS) axis). The wave direction for all the waves in column 1 of Table 8 (i.e. under ‘S’) should be specified as 180 degrees since the –X (SACS) axis is oriented 180 degrees away from the +X (SACS) axis in the anticlockwise sense.
Waves approaching the platform from the W direction would be moving along the +Y (SACS) axis (i.e from –Y(SACS) axis towards +Y (SACS) axis). Hence the wave direction is specified as 90 degrees, since the +Y (SACS) axis is oriented 90 degrees w.r.t the +X (SACS) axis. The wave direction angles for the other two approach directions i.e. NW and SW are therefore 45 degrees and 135 degrees respectively.
For arguments sake let us suppose that the bid document specifies that the platform north is oriented in the direction of the geographic (true) West direction. If we create the SACS input model in the usual way then the north face of the platform would still be perpendicular to the SACS –X axis. Now let us suppose that the bid document says we need to provide a wave load for the geographic West direction i.e. in the actual platform, the wave would approach the jacket from the true West and hit the North face of the platform first. In this case the wave would be moving along the +X axis of SACS and hence we would have to specify a wave direction of 0 degrees for this wave. Other wave directions may be specified accordingly so that the correct face of the jacket is loaded.
Mar 07 Page 48 of 110
th
(i) Under Wave Parameters enter Wave Type = STOK (Stokes 5 Order Wave). In the Wave Height field, change the wave height to Average wave height of the specific wave. For example, for the RS-2 project, Table 8 in the bid (structural design criteria) specifies the first wave height as 0-1.523. Compute the average of this range i.e
0 + 1.523 2
= 0.76 m. Enter a Kinematic factor to
1.0(see bid) in the WAVE card. This is an API-RP2A requirement for fatigue waves. The first entry to be made for the RS-2 project is for a wave approaching from the true S, we enter a wave direction of 180 degrees and a wave period of 8.7 seconds.
(j) Under Crest Position check an input mode of degrees (DEG).An established procedure is to advance the wave in steps of 4 degrees. For sake of simplicity if we have a perfectly sinusoidal wave a crest position may be taken to represent 0 degrees. The next crest to hit the platform represents a wave advancement through the structure of 360 degrees. A crest position of the wave will create some wave load, the intermediate wave positions will impose different wave loads after which the cycle will be repeated causing cyclic loading of the structure. To have sufficient representation of all positions (and including) the two crest positions we advance the wave in steps of 4 degrees so that the total number of static steps is 90. More steps may be created by creating smaller wave advancement steps, however a step of 4 degrees is usually sufficient.
(k) For the Critical Position enter MS for obtaining maximum base shear value. This simply means that we will pick up the critical wave position which creates the maximum base shear.
(l) Under the Miscallaneous tab enter a maximum member segmentation value of 10. This represents the maximum number of load segments that can be used to describe the non-linear load distribution on the member
Mar 07 Page 49 of 110
(m)Repeat steps (e) through (j) for the same wave height and period but with a critical position of NS i.e. to indentify the wave position which gives minimum base shear. The wave positions giving maximum base shear and minimum base shear will most likely create the largest stress range and contribute to fatigue failure.
The following extract is for the first entry in Table 8 of the RS-2 bid document for a wave height range of 0-1.523 m approaching the platform from the geographic South having a period of 8.7 seconds. Line 3 will identify the wave position creating Maximum base shear while line 6 will identify the wave position creating Minimum base shear.
LOADCN 1 WAVE WAVE1.00STOK 0.76
8.70
180.00
D 0.00 4.00 90MS10 1 0
8.70
180.00
D 0.00 4.00 90NS10 1 0
LOADCN 2 WAVE WAVE1.00STOK 0.76
(n) Repeat steps (e) through (l) to create the WAV cards for all the 26 entries in Table 8 of bid document with one entry for maximum base shear and one entry for minimum base shear i.e. a total of 52 WAV cards.
(o) The KPGI fatigue sacinp file contains an additional WAV card with steps of 30 degrees and an entry of ‘AL’ for Critical Position. We will not follow this procedure. For our initial run we create only the MS and NS WAV cards (52 in all).
(p) Delete all Load combinations from the sacinp.fatigue file.
3. Run this file in Static Linear static analysis
Select
sacinp.fatigue.
Mar 07 Page 50 of 110
4. In the output listing a saclst.fatigue file is obtained. Open this file and search for the string “Maximum Shear” by using the F2 key. Note down for each wave height/wave period and each wave direction (26 in all) the crest positions for maximum and minimum shear at mudline. For example for a wave of height 0.76m having a period of 8.7 seconds and approaching the jacket from the South, the crest positions giving maximum and minimum shear at mudline are 276 degrees and 92 degrees respectively. Let us designate these angles as X and Y.
5. Obtain three angles between maximum shear crest position (X) and minimum shear crest position (Y) as (Y-X)/4 and remaining three angles as ((360-X)+Y)/4. In the second stage of analysis we re-run the analysis with the angles X, Y and the 6 additional angles and find out the maximum stress range. For example for the wave of height 0.76m having a period of 8.7 seconds and approaching the jacket from the South, the crest positions giving maximum and minimum shear at mudline are 276 degrees and 92 degrees respectively. The intermediate angles considered are therefore 138,184,230 between 92 and 276 degrees and 320,4 and 48 degrees between 276 and 92 degrees.
6. Using the WAV card enter the 2 peak and 6 intermediate angles in the sacinp file and change the number of steps to 1 in increments of 1degree. This has to be done for each wave height and wave period for each direction (a total of 208 WAV cards for the RS-2 jacket). The extract below (in step 7) represents the 8 WAV cards for the wave height 0.76 m with a period of 8.7 seconds and approaching the jacket from the geographic South direction. Here we directly specify the crest position in terms of the angles calculated in the previous step. For example LOADCN 2 below (in step 7) represents the wave where the wave crest position has advanced through the structure by 138 degrees. The diagram below illustrates this (note that the wave shape below is only for illustrative purposes and does not represent the shape of a real wave)
Mar 07 Page 51 of 110
Crest at 0 degrees (origin)
Crest at 138 degrees from origin
7. Note that we are starting with the angle giving Minimum Base Shear and specifying the remaining angles calculated in the previous step in an anticlockwise direction. Also note that we specify MS as the criteria for critical position. Since we are specifying a specific crest position, a 1 degree increment and only 1 static step, resultant loads, member forces will be calculated only for the specified wave height and the wave will not be advanced through the structure as in step 2(j)-2(k) above.
Recall that each of the load cases below will contribute to a certain damage level. The damage level will be calculated for all the sea-states (i.e each wave height and wave period and different crest positions of the waves through the structure) summed up and the resulting total damage level will be used in computing the fatigue life of the structure.
Mar 07 Page 52 of 110
○
92
min. base shear
○
48
○
138
○
4 ○
184
○
320 ○
230
○
276
max. base shear
Angular distance between origin and wave crest position for wave height 0.76m and period 8.7 seconds LOAD LOADCN 1 WAVE WAVE1.00STOK 0.76
8.70
180.00
D 92.00
1. 1MS10 1 0
8.70
180.00
D 138.00 1. 1MS10 1 0
8.70
180.00
D 184.00 1. 1MS10 1 0
8.70
180.00
D 230.00 1. 1MS10 1 0
8.70
180.00
D 276.00
1. 1MS10 1 0
8.70
180.00
D 320.00
1. 1MS10 1 0
8.70
180.00
D 4.00
LOADCN 2 WAVE WAVE1.00STOK 0.76 LOADCN 3 WAVE WAVE1.00STOK 0.76 LOADCN 4 WAVE WAVE1.00STOK 0.76 LOADCN 5 WAVE WAVE 1.0STOK 0.76 LOADCN 6 WAVE WAVE 1.0STOK 0.76 LOADCN 7 WAVE WAVE 1.0STOK 0.76 LOADCN 8
1. 1MS10 1 0
Mar 07 Page 53 of 110
WAVE WAVE 1.0STOK 0.76
8.70
180.00
D 48.00
1. 1MS10 1 0
The WAV cards are input as described earlier.
8. Run the sacinp file in Linear Static Analysis as before.
9. A saccsf.* file is obtained which will be used for fatigue analysis.
10. Create a ftginp.* file (Data File>Create New Data File>Post>Fatigue>Select). A Title may be entered. In the next window (Fatigue Analysis Option) check the ‘Direct Deterministic Fatigue’ option.
11. In the next window i.e. Fatigue Options (FTOPT card) enter design life (25 years for the RS-2 jacket;see bid), a life safety factor (see bid, usually 2.0). Also enter the fatigue time period as the Wave Period (generally 1 year). Check the following: Skip Non-tubular elements, Skip all Plates, No Input Echo, Use Load Case Dependent SCF’s.
12. Choose the API X Prime for i.e. APP for Source of S-N curve.
13. For the SCF Option use Kuang & Wordsworth (KAW) for all joints initially. Also check the following: Prescribe Max SCF and Prescribed Min SCF.
14. In the next Fatigue Option 2 window (i.e. the FTOPT2 card) check the following check boxes: PT (so that damages are printed along with the member stresses), Export Fatigue data to PSVDB, Tubular Inline Check (this ensures that a fatigue check is done at section changes along tubular members), Inline Tubular SCFAWS and Effective Thickness Ratio-2WAL (this ensures that the moment of inertia of the two walls is used for the grouted joints). Click Next.
Mar 07 Page 54 of 110
15. Skip the ‘Input Weld Classification Factors (PCLASS)’ and ‘Override SCF’s for individual joints (JNTSCF)’.
16. Click Yes in the ‘Override Fatigue Parameters for Individual Joints (JNTOVR)’ window. Initially we do not know which joints will have fatigue lives less than the design fatigue life. Hence at this stage we will use this joint over-ride option only for the grouted joints. Note that in the sacinp.* file we have specified the sizes and cross sections of the members. An override is used to specify a size that will give a satisfactory fatigue life by overriding the cross section/size specified in the sacinp.* file.
17. In the ‘Joint Override’ window enter the name of the grouted joint. At this stage of the analysis we do not know if the grouted joints will fail so we are not sure about the ‘Chord Thickness Override’. The only change to be made in this window is to specify the source of the S-N curve as the AXP curve and the SCF Option with Marshalls Method. The remaining options can be accepted with default values. To add more grouted joints click the ‘+More’ button and add more joint overrides. Click ‘Next’ after all the grouted joint overrides have been specified. In case of the RS-2 jacket, skirt piles are being used. The extract below shows the grouted joints being over-ridden. To understand which joints these are open the sacinp.* file in Precede. Use Joint>Find Joint and then enter these Joint names to find them in the model. SACS will show these joints with a circle. Here JB7, JB8, JB37 and JB5 are the PILHD joints and SK9, SK15, SK24 and SK3 are the skirt joints.
JNTOVR JB7 AXP
MSH
JNTOVR JB8 AXP
MSH
JNTOVR SK9 AXP
MSH
JNTOVR JB37 AXP
MSH
JNTOVR SK15 AXP
MSH
JNTOVR SK24 AXP
MSH
JNTOVR JB5 AXP
MSH
JNTOVR SK3 AXP
MSH
Mar 07 Page 55 of 110
18. Skip the ‘Override Fatigue Parameters for Plate groups (PGROVR)’, ‘Override Fatigue Parameters for Plates (PLTOVR)’, ‘Override SCF’s for Member Groups (GRPSCF)’.
19. Click ‘Yes’ on the ‘Remove Groups from Analysis (GRPSEL)’. Here we specify dummy structures, appurtenances, Risers and conductors which are not part of the main structural Member but attract Wave Load. Check the ‘Exclude Members of these groups from Analysis’ and specify these groups in the boxes provided under ‘GRUP ID’.
20. Skip the ‘Override SCF’s for individual members (MEMSCF)’, ‘Override SCF’s for specific brace-chord connections (CONSCF)’, ‘Override SCF’s for specific Wide-Flange members (CONSWF)’.
21. Specify the SCF limits in the next (SCFLM) card as a maximum of 6.0 and a minimum of 1.6
22. Skip the ’Select specific joints to be analyzed (JSLC)’. Click ‘Yes’ in the next ‘Relief’ window so that brace stresses are calculated at the surface of the chord. Click ‘Finish’.
23. The next step is to input the wave occurrence data. This has to be done for each direction. Using Edit>Insert Line>FTCASE(Deterministic Fatigue Case)>Select enter the Fatigue Environment Number (in the case of the RS-2 jacket, use 1 for South, 2 for SW, 3 for W and 4 for NW), MMN for stress calculation (which specifies that stresses are determined by a max/min search on all the load cases) and the first wave height from Table 8 i.e.0.76m.
24. The next step is to specify the contribution of each fatigue case. Recall that in the linear static analysis of step 6 and 7 we have used a sacinp.* file where for each
Mar 07 Page 56 of 110
wave height, wave period and wave direction we have specified 8 WAV cards, two for the crest positions giving the maximum and minimum base shear and 6 intermediate positions. Each of these 8 load cases contributes as a fatigue case. Hence we enter its contribution as follows. Use Edit>Insert Line>FTCOMB (Fatigue Case Contribution)>Select to insert the FTCOMB line in DataGen. For each of the 8 cases for a given wave height and wave period enter a factor of 1.0. The extract below shows Fatigue case contributions for the geographic South direction waves for the RS-2 jacket.
FTCASE 1
MMN
FTCOMB 1
1.0 2
FTCOMB 8
1.0
FTCASE 1
1.0 3
MMN
FTCOMB 9
1.0 10
FTCOMB 16
1.0
FTCASE 1 FTCOMB 17
1.0 18
FTCOMB 24
1.0
FTCASE 1 FTCOMB 25
1.0 26
FTCOMB 32
1.0
FTCASE 1 FTCOMB 33
1.0 34
FTCOMB 40
1.0
1.0 11
0.76 1.0 4
1.0 5
1.0 6
1.0 7
1.0
2.29 1.0 12
MMN
3.81
1.0 19
1.0 20
MMN
5.33
1.0 27
1.0 28
MMN
6.86
1.0 35
1.0 36
1.0 13
1.0 14 1.0 15
1.0
1.0 21
1.0 22
1.0 23
1.0
1.0 29
1.0 30
1.0 31
1.0
1.0 37
1.0 38
1.0 39
1.0
25. The next step is to enter the wave occurrence data. Table 9 gives the wave exceedance data i.e. the number of waves observed to have a specified height. For ex in the RS-2 Jacket, the first entry in Table 9 is for the number of waves approaching the platform from the geographic South and having a height more than 0 meters. The table lists this value as 1,276,045. The next entry shows that 61,704 waves were observed to have a height more than 1.524 meters. This means that (1,276,045-61,704 = ) 1,214,341 waves had a height between 0 and 1.524 meters.
Mar 07 Page 57 of 110
26. This interval is further broken down using the spreadsheet “octa fatigue wave occurance cal.xls” to interpolate on the wave heights and number of occurrences in any interval. The spreadsheet extract below shows the actual exceedance observations from the South for different wave heights (from Table 9 of the RS-2 bid document) entered in the columns on the left. The output in the columns on the right is the interpolation values. For example the spreadsheet output shows that the number of waves having a height of 0.095 meters was 220094 and so on. The spreadsheet output data from the columns on the right is what is entered in the ftginp.* file.
27.
1 2 3 U4 s5 6 e7 8 9 E0 0 d 0 i0 0 t 0 >0 0 I0 n0 0 s0 e0 0 r 0 0 t 0
L i n
Height
Exceedance
0.000 1.524 3.048 4.572 6.096 7.620 9.144 10.668 12.192 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
1276045 61704 3132 167 11 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Height
Exceedance
Occurrence
0.095 0.286 0.476 0.667 0.857 1.048 1.238 1.429 1.619 1.810 2.000 2.191 2.381 2.572 2.762 2.953 3.143 3.334 3.524 3.715 3.905 4.096 4.286 4.477 4.667 4.858 5.048 5.239 5.429 5.620 5.810 6.001
1055951 723103 495172 339087 232203 159010 108888 74565 51216 35285 24310 16748 11539 7950 5477 3773 2608 1808 1253 869 602 417 289 201 141 100 71 51 36 26 18 13
220094 332849 227931 156084 106885 73193 50122 34323 23349 15931 10976 7562 5210 3589 2473 1704 1166 800 555 384 266 185 128 89 60 41 29 21 15 10 7 5
Mar 07 Page 58 of 110
e>WVFREQ(Wave Frequency Interpolation)>Select to insert the wave occurrence data determined above. Since we are entering data for the South direction, enter a fatigue environment number of 1, the wave heights and occurrence data from step 26.
28. Repeat steps 23-27 to enter the remaining fatigue case combinations and wave occurrence data.
29. The last card is EXTRAC HEAD AE card (Edit>Insert Line>EXTRAC HEAD (Joint Extraction Head Line)>Select). Check “Automatic Extraction” and enter a “damage level” cut off of 0.5 to extract joints which have a fatigue life of less than 50 years.
30. Now, in SACS Executive enter an appropriate label, click on Runfile Wizard>Post>Fatigue Damage (under Post Processing). Label ensures that all the output files in this analysis will have the same label. For ex. in the analysis below will have a label oeg_fatigue (ex. ftgext.oeg_fatigue) aiding identification later.
Mar 07 Page 59 of 110
Label 31. Click on Start Wizard. Select the ftginp.* file created in the previous few steps, click ‘OK’ in the Analysis options window, select the saccsf.* file and click ‘Open’
32. Click on Run to execute the fatigue analysis.
33. In SACS Executive, an ftgext.* has been created under output files. This is the fatigue extraction file which stores results of the EXTRAC command specified in the ftginp.* file. Recall we had specified that all joints having a fatigue life less than 50 years were to be extracted. Double click on the ftgext.* file. This opens up the file in the ‘Interactive Fatigue’ module.
Mar 07 Page 60 of 110
34. In Interactive fatigue fatigue life of the joints failing is displayed. Using Members>Details/Modify we can make changes in the Chord Thickness, or S-N curve to be used (SCF/S-N>Select S-N) etc. to find the optimum change that has to be made to a chord/brace to get a satisfactory fatigue life. The fatigue life due to the change is displayed in the Interactive Fatigue window. Note down the specific changes that need to be made for each failing joint. Open the ftginp.* file in DataGen and enter the JNTOVR cards (Edit>Insert Line> JNTOVR (Joint Override)>Select and enter the overrides.
NOTE: In the case of grouted joints, if the S-N curve needs to be changed, use the API X curve with effective thickness option i.e. AXP option.
35. For example the extract below from the ftginp.* file for the RS-2 jacket represents the overrides to the joints that were failing
*LEG JOINT JNTOVR JB1 AXX
7.5
JNTOVR JB2 AXX
7.5
JNTOVR JB3 AXX
6.5
JNTOVR JB4 AXP
7.5
JNTOVR JF2 AXX
7.5
JNTOVR JF3 AXP
7.5
JNTOVR JF4 AXP
7.5
JNTOVR SK21 AXP
8.3
JNTOVR SK6 AXP
8.3
* JNTOVR JE1 AXP
6.5
JNTOVR JE2 AXX
5.7
JNTOVR JF1 AXP
6.5
JNTOVR SK12 AXP
6.5
JNTOVR JD2 AXP
5.7
JNTOVR JE3 AXP
5.7
*CONDUCTOR GUIDES JNTOVR CF19 AXP
3.6
JNTOVR CF20 AXP
3.6
Mar 07 Page 61 of 110
JNTOVR CF24 AXP
3.6
JNTOVR CF25 AXP
4.2
JNTOVR CF26 AXP
4.2
JNTOVR CF27 AXP
4.2
JNTOVR CF28 AXP
3.6
JNTOVR CF29 AXP
4.2
JNTOVR CF30 AXP
4.2
JNTOVR CF63 AXP
3.6
JNTOVR CF64 AXP
3.6
JNTOVR CF65 AXP
3.6
JNTOVR CF23 AXP
3.6
JNTOVR CF21 AXP
3.6
JNTOVR CF32 AXP
3.6
JNTOVR CF35 AXP
3.6
JNTOVR CF31 AXP
3.6
JNTOVR CF33 AXP
3.6
JNTOVR CF34 AXP
3.6
JNTOVR CF36 AXP
3.6
*GROUTED JOINTS JNTOVR JB7 AXP
MSH
8.3
JNTOVR JB8 AXP
MSH
8.3
JNTOVR SK9 AXP
MSH
8.3
JNTOVR JB37 AXP
MSH
8.3
JNTOVR SK15 AXP
MSH
8.3
JNTOVR SK24 AXP
MSH
8.3
JNTOVR JB5 AXP
MSH
JNTOVR SK3 AXP
MSH
8.3
*BRACES JNTOVR J481 AXP
3.0
JNTOVR J482 AXP
2.3
JNTOVR SKB1 AXP
2.0
23. Run the Final ftginp file with the saccsf file till no joint has a service life less than 50years.
Mar 07 Page 62 of 110
24. The output files that are obtained from the run are the ftglst and the ftgext files.
1.0 SEISMIC ANALYSIS (BY THE SINGLE PILE ANALYSIS METHOD) Seismic analysis is carried out examine the response of the structure to earthquakes. Here the only dynamic force input is from ground motion as API recommends that environmental loads need not be considered, except for still water effects of hydrostatic 1
pressure and buoyancy (static loads) . Reference Files: Z:\MECHANICAL ENGINEERING \ OEG \ OEG_SACS_Training \SACS\RS-2 \seismic_2 The methodology followed for Seismic analysis using the single pile method is as follows4: a) All the vertical loads (dead weight, operational loads and live loads with appropriate load contingency and reduction factors) are set up as load cases in the static analysis input file for the structure which results in static stresses in the members. The effects of buoyancy and hydrostatic pressure are also included in the static analysis. b) A second input file will all the vertical loads (including contingency and reduction factors) is prepared for dynamic analysis c) Using DYNPAC the natural frequencies and mode shapes are extracted.
Mar 07 Page 63 of 110
d) Using the DYNAMIC RESPONSE program, the CQC responses of the structure due to the component earthquakes along X, Y & Z directions and the resultant response due to the 3 components is found out. e) The static and dynamic (seismic) stresses are combined to get the total stresses in the members. 4 different load cases are generated that represent 4 possible combinations of static and seismic response. The first 2 cases are the member check due to tension and compression in the members with a seismic load factor of 1.0. The next 2 response combinations are for the joint check due to tension and compression in the members with a load factor of 2.0 f) The members and the joints are then subjected to code checks by use of POST and JOINT CAN modules of SACS. The detailed procedure is outlined below.
1.
Create 5 directories under seismic analysis folder – Static, Pile, NatFreq, Earthqk and Post.
2.
The first step is to estimate the natural frequencies and mode shapes of the structure accounting for the non-linear pile soil interaction. For this we have to run a single pile analysis to estimate equivalent pile stub lengths. The detailed procedure of the single pile analysis is outlined in section 3.0
3.
In the Static folder, run a Linear Static analysis with PSI option for the In-place model (steps 1 through 11 of section 3.0). From the ‘psilst’ file obtain the max. axial, lateral forces and moment for any pile group and the axial displacement from operating storm load cases only. For example, the psilist.earthqk_1 file in the Static folder at the reference location mentioned above lists the following results for the 4 pilehead joints. Note that only the cases giving the maximum lateral forces have been listed below Pile Jt. L.C
Axial(KN) Lateral
B.M
daxial
JB5
J206 -17895.07
1591.75
9844.5
2.35
JB8
J204 -18068.19
1565.89
9663.0
2.38
Mar 07 Page 64 of 110
4.
JB37
J204
8952.04
1957.07
10454.3 -0.88
JB7
J206
8732.56
1965.69
10458.7 -0.85
In the Pile folder, Make a copy of the psiinp.* file and rename it as pilinp.*. Carry out the remaining steps (12 through 23) of Section 3.0. For the RS-2 jacket two piles (JB37 and JB7) use Soil Table 1 and the remaining two (JB5 and JB8) use Soil Table 2 for psi. Hence two pilinp.* files have been created in the Pile folder viz. pilinp.rs2_sol1 and pilinp.rs2_sol2. If both soil tables are provided in the same file then during the single pile analysis make sure that the table not being used is commented to avoid a SACS error.
5.
The pillst.jb5, pillst.jb8, pillst.jb37 and pillst.jb7 files in the Pile folder list the required pile stub dimensions for the 4 piles.
6.
Pile
Joint to Joint Length(cm)
Axial Offset(cm)
JB5
1506.826
133.784
JB8
1501.829
131.364
JB37
1549.35
146.895
JB7
1513.053
182.0.38
Update the Inplace model by removing the pile group members and pilehead joints and adding the pile stubs to it. Also paste the pile stub section property line to the input file. (Refer to Section 4.0 on modal analysis for procedural details).
7.
Copy the input (sacinp) file to the NatFreq folder.Make all corner nodes of the various levels as ‘222000’ thereby specifying them as retained degrees of freedom. Delete all load cases except one combination for operating loads with suitable contingencies. This load case should not contain any environmental load data. Name it ‘SLE’ or ‘DLE’ for Strength Level and Ductility Level Earthquake runs respectively. Include DYM option in LDOPT card and DY in LCSEL card. In case of the RS-2 jacket this Load combination has been given a ‘601’ label at the end of the sacinp.* file.
Mar 07 Page 65 of 110
8.
Create a dyninp.* file (as specified in Section 4.0) with the number of mode shapes to be extracted. Specify SA option in DYNOPT card and run a Dynpac analysis. Open the output file (dynlst.*) and check the period and total mass participation factors. For example the output frequencies for the first 10 modes of the RS-2 jacket with equivalent pile stubs modelled are
9.
MODE
FREQ.(CPS)
GEN. MASS
EIGENVALUE
PERIOD(SECS)
1
0.487909
2.0375832E+03
1.0640532E-01
2.0495643
2
0.594632
1.1043473E+03
7.1638032E-02
1.6817123
3
0.848458
5.5122671E+02
3.5186813E-02
1.1786092
4
0.983272
2.5653819E+03
2.6199514E-02
1.0170130
5
1.080732
2.9881244E+03
2.1687239E-02
0.9252988
6
1.324675
7.2876359E+02
1.4435157E-02
0.7549021
7
1.985413
2.0073924E+02
6.4259662E-03
0.5036735
8
2.045198
4.1118429E+03
6.0557740E-03
0.4889503
9
2.446901
6.2300432E+01
4.2306521E-03
0.4086801
10
2.631432
2.8111308E+02
3.6581037E-03
0.3800213
Copy the updated input file (in the Static folder this updated sacinp file has been named sacinp.earthqk1) generated in Step 7 above to the Static folder. Add LCSEL card with load case ‘SLE’ or ‘DLE’ (in sacinp.earthqk1 this load case has been labelled 601) and specify ‘CMB’ option in the LDOPT card. Run a Linear static analysis. Note that the model file contains the pile stubs. Therefore PSI is not carried out. The run is a simple Linear Static analysis. A common solution file saccsf.* file will be created. This step will induce stresses in the members due to static conditions.
10. The next step is to run a response analysis to estimate behavior of the structure to the forced external excitation due to the earthquake loading. Copy the updated model, the ‘dynmas’ and ‘dynmod’ files generated in Step 8 (from the Natfreq
Mar 07 Page 66 of 110
folder) above to the Earthqk directory. Also copy the ‘saccsf.*’ file generated in Step 9 above to the Earthqk directory.
11. Create a ‘dyrinp.*’ file (Data File>Create New Data File>Dynam>Spectral Earthquake>Fatigue>Select). Choose SPEC as the Analysis option, specify the number of modes to be included, the vertical co-ordinate as +Z and the mudline elevation. Click Next.
12. In the Structural Damping card enter an overal modal damping of 5%. In the Simulated Earthquake Output loads card specify Load Type as ‘BS’ (Base Shear) and ‘Both’ under the reverse option. Under the Static plus Dynamic Spectral Combination enter Element Check and Joint Check load case factors of 1.0 and 2.0 respectively. Also specify the SLE load case (in RS-2 this is labelled 601) with a factor of 1.0.
13. Check the General Spectral Response Analysis in the Analysis Type. In the Spectral load card specify the Spectrum Source as CARD, Spectrum Type as ‘R’, Spectrum form as ACEL, Damping Type as SDO, Modal Combination type as st
nd
rd
CQC. Under Joint Print specify A, V and D respectively for the 1 , 2 and 3 Joint Data Print Option.
14. The Response factor is 0.06, and directionality factors of 1.0,1.0 and 0.5 in the X, Y and Z directions respectively. In the Spectral Response Header, enter Number of Damping values as 1. Next we have to specify the Spectral Response Graph. This is obtained from IS1893 and is reproduced below. In the Spectral Response Function card enter 12 as the number of periods (meaning 12 values will be entered) with a damping ratio of 5%. Then continue entering the Period and Response value from the IS1893 graph depending upon the type of Soil until DataGen generates an END card. Save this file.
Mar 07 Page 67 of 110
15. In SACS Executive click ‘Dyn’ and select ‘Earthquake’ under ‘Dynamic Analysis’.Run an earthquake analysis with the above files and generate the ‘dyrcsf’ file.
16. In the ‘dyrlst.*’ file check the max. axial, lateral forces and moment under the ‘CQC SUMMATION FROM ALL DIRECTIONS’ column in the ‘dyrlst’ file.
17. These should be compared with the values added in Step 4. In case it is different, update the ‘pilinp’ file with these values.
18. Repeat steps 4 to 17 till the values converge to a reasonable limit.
Mar 07 Page 68 of 110
19. Create a ‘pstinp’ file in the Post folder. Select Load Case 1 and 2 only i.e., {Earthquake + Static (Tension)} and {Earthquake + Static (Compression)} with a AMOD of 1.7 on the allowables.
20. Perform Element Code check and generate a Postvue Database file using the ‘dyrcsf’ file generated in Step 11 above.
Create a ‘jcninp’ file with AMOD as 1.7 for Load Cases 3 and 4 only i.e., {Earthquake + Static (Tension)} and {Earthquake + Static (Compression)} for joint check case. Run Joint Can analysis. 6.0
MODAL ANALYSIS
Modal analysis is carried out to determine the various mode shapes of the platform. Recall the equation of motion
&& + C M x x& + Kx = P(t ) where M represents the structural mass, C the structural damping, K the structural stiffness and P(t) the external forcing function. If the structure has n degrees of freedom, then it has n natural frequencies. Free vibration at one of these natural frequencies results in a particular relationship between the amplitudes and phases of the n d.o.f’s. This shape of vibration is called a normal mode. Generally, free vibration will result in all the normal mode vibrations occurring simultaneously. Forced vibration will result in large oscillations when the forcing frequency equals any natural frequency3.
Free vibration implies that the external forcing function be set to zero, in other words, it means that the structure is displaced from its normal equilibrium position and released so that it oscillates. At this stage we also set the damping to zero and essentially solve the equation
&& + Kx = 0 M x What this means for our modal analysis is that only those loads which are of a permanent nature (ex. Dead loads etc) be retained and all those loads which vary with time be removed from the sacinp.* file. The steps are outlined below and reference is made to the Bunduq Platform sacinp.* file located in the following folder
Mar 07 Page 69 of 110
(Z:\MECHANICAL ENGINEERING \ OEG \ JB's Data\Disc4 – Offshore \ Bunduq (Part 3) \ VA SACS Analysis\Rev 5 (For Report)): -
1. Make a copy of the sacinp file and rename it. In case of the Bunduq platform in the above folder, this file has been named sacinp.dyn.gip_va. This copy will be the sacinp file to be used for modal analysis. Make the following changes in the new sacinp.* file by opening it in DataGen. Right click and Modify LDOPT line. Change the ‘Seastate Analysis Option’ from NSM to DYN.
2. For the modal analysis we do not use the full model. Instead we specify a few joints as having master or retained degrees of freedom. The other joints will be automatically treated as having slave degrees of freedom. Normally the corner joints of the jacket and deck are specified as retained DOF’S. By doing this during the solution process the slave degrees of freedom will be statically condensed out and a solution obtained for the retained d.o.f’s only thus speeding up the solution. 3. The retained d.o.f’s can be specified in two ways. One way is to open the sacinp.* file in Precede, click on Joint>Details/Modify and then double click on the corner joints of the jacket and deck and then specify the fixity as 222000. The other way is to open the sacinp.* file in DataGen, go to the joints which are to be specified as having retained d.o.f’s, right click and modify the Joint fixity to 222000.
4. Retain only permanent loads on the structure. Remove all other Load Data. Convert all NGDL weights from Buoyant weights to Dry Weights.
5. As mentioned above to carry out a modal analysis we apply only the permanent loads. Delete all the LCSEL cards. Keep only one LCSEL card which selects only one load combination of these permanent loads. For example, for the Bunduq platform this load combination has been named DLE. Also note that in the LCSEL card under ‘Function‘ select DYNA instead of ALL or STND.
Mar 07 Page 70 of 110
6. At the end of the sacinp.* file DLE is defined as a combination of the permanent loads as shown below LCOMB LCOMB DLE 1 1.100 2 1.120 3 1.120 5 1.100 6 1.150 7 1.100 LCOMB DLE
8 1.000 9 1.000 100.8250 110.8625 120.8250 130.8250
LCOMB DLE 160.8250
7. Before running the modal analysis we need to account for pile-soil interaction. This is important because the soil is not rigid and hence its effect is to increase the natural time period of the jacket structure.
8. To account for PSI we have to run a single pile analysis. The procedure to do so is outlined in section 3.0. The output of this procedure is a pillst.* file which gives the equivalent pile stub joint to joint length, axial offset and cross section details.
9. Open the pillst.* file and search for the string ‘Stub Properties’. The pile stub length and axial offset required are reported under as ‘Joint to Joint Length’ and ‘Axial Offset’. To find the pile cross sectional properties search for the string (using F2) ‘Sect Pilstub’. For example, the file pillst.single_pile in the same folder lists the cross section properties of the pile associated to pilehead joint 099P as shown below
SECT PILSTUB PRI395.99880690.0880690.0880690.0 10.0
10.0
10. Copy this line from the pillst.* file and paste it in the sacinp.* file as a section definition. In the file sacinp.dyn.gip_va this section has been added just before the groups have been defined. Note that the cross sectional properties defined by the SECT PILSTUB card are slightly different from the properties obtained from the pillst.single_pile file. This difference may have arisen because KPGI’s single pile analysis used ‘deflection’ as a criteria while we use ‘force’ as the criteria during step 18 of the single pile analysis procedure. Save and close the sacinp.* file before the next step.
Mar 07 Page 71 of 110
11. Also define a group which uses the section properties of the pilestub. For example in the sacinp.dyn.gip_va file this GRUP card is as shown below
GRUP PL2 PILSTUB
20.00 8.0024.80 1
1.001.00
N 7.849
12. There is one final step before doing modal analysis. This is to add the equivalent pile stubs having the length and cross sectional properties obtained from the single pile analysis.
13. To understand the addition to be made, open the sacinp.sta.gip_va in Precede. This is the sacinp.* file that has been used to carry out the single pile analysis. The PileHead joints are labelled 001P, 009P, 019P, 081P,089P and 009P. The Precede view in the screen shot below has been obtained using Display>Plane>3 Joints and selecting 3 joints on the front face of the model. Then use Joint>Find and enter
Mar 07 Page 72 of 110
001P to find the first pilehead joint. In the view below joint 001P appears circled.
14. Click Joint>Details/Modify and double click on joint 001P. Note that the Fixity for this joint has been specified as PILEHD.
15. Open the sacinp.* file to be used for modal analysis in Precede. Click Joint>Details/Modify and double click on the PileHead joints one at a time and delete
the
PILEHD
specification
under
‘Fixity’.
For
example,
open
sacinp.dyn.gip_va in Precede and see that the fixity of the pilehead joints has now been changed from PILEHD and now they function as any other regular joint.
16. Now create the equivalent pilestubs. Click on Joint>Add>Relative. In the pop-up window click in the space in front of the ‘Reference Joint’ and then double click on one of the PileHead joints. Enter the distance noted for the member ‘Joint to Joint length’ from the single pile analysis with a negative value so that the new joint is created below the pilehead joint. For example in the sacinp.dyn.gip_va joint R564
Mar 07 Page 73 of 110
has been created relative to the original Pilehead joint 001P by a distance of -3.218 m. This distance has been taken from the pillst.gip_va file as the ‘Joint to Joint Length’ for the pile associated with PileHead Joint 001P after the single pile analysis has been run.
17. Use Joint>Details/Modify and specify the fixity at the new joint created as ‘111111’ creating complete fixity.
18. Next use Member>Add and add a member between the new joint created and the pilehead joint. For the Group Label scroll down the drop down list and select the group label created in step 11 (in the sacinp.dyn.gip_va file the new group created was labelled PL2). Once this new member has been created we have to specify the axial offset. Use Member>Details/Modify and double click on the new member. In the Offset type use Local. While creating the member if the new joint has been picked first then it is Joint A for this member while the pilehead joint is Joint B. In the sacinp.dyn.gip_va file the axial offset of 30.1 cm has been specified at the pilehead joint i.e. at Joint B. Click Apply to accept.
19. Now repeat steps 16 through 18 of this section to create the remaining equivalent pilestubs. Save and close Precede. This completes modification of the SACS model file to account for the non-linear pile soil interaction.
20. Next,
in
SACS
Executive
use
Data
File>Create
New
Data
File>OK>Dynam>Dynpac>Select to create a dyninp.* file. A title may be entered. Click ‘Next’ to go to the DYNPAC options. Use +Z as the vertical coordinate. MN as the Units, the number of modes to be extracted (in the dyninp.dyn.gip_va file 3
this has been specified as 160), Structural density as 7.85 MT/m . Ignore the other options. Under the ‘Mass’ tab select ‘Cons’ under the ‘Mass Calculation’ and ‘SA’ under the ‘Masses from SACS loads options’. Specify ‘-Z’ as the ‘SACS loads direction for masses’ Click ‘OK’ and continue clicking ‘Next’ until an END statement is created as the last line of the dyninp.* file. Save and close this file.
Mar 07 Page 74 of 110
21. In SACS Executive click on ‘Dyn’. The default analysis option is ‘Extract Mode Shapes’. Start the analysis and pick the modified sacinp and the dyninp file under the ‘Mode Shape’ tab. Under the Postvue tab check the ‘Create Postvue DB’ in case a visualization of the mode shapes is required. Run the analysis.
22. The output of the analysis is the dynlst file. This file lists the modal frequencies and mass participation factors. Scroll down to the end of the file to ensure that the cumulative mass participation factor is at least 95%. In the dynlst.gip_va file note th
that for the 160 mode the cumulative mass participation factor has exceeded 95% so we are fine.
** MASS PARTICIPATION FACTORS ** ** CUMULATIVE FACTORS ** MODE
160
X
Y
0.0000007
0.0000026
Z
0.0008027
X
Y
Z
1.002432 1.001249 0.959747
23. To view the mode shapes, double click on Postvue in SACS Executive. This opens up the modal analysis results in Postvue. To animate the mode shape click Display>Shape. In the Deflected Shape Display Options window, check Animation of shape and click OK
24. The default mode shape displayed is for mode 1 along with modal frequency. To view other mode shapes click Load>Display Single LC and then enter the mode number to be displayed. Alternatively click Load>Display next LC in List. This will display the next mode and so on.
25. This concludes the modal analysis.
Mar 07 Page 75 of 110
2.0
VIBRATION ANALYSIS
1.
The steps 1 to 9 specified in the Seismic Analysis section 5.0 above have to be first performed in the vibration analysis run.
2. Note : Retain relevant degrees of freedom including nodes present on equipment.
3.
Create a dyrinp file and specify the run speeds of the reciprocating machines in the RSPEED card. Change the number of modes in the ENGVB card. Specify the unbalanced force and moments acting at various joints of the structure using the UNBAL card. The damping factor is 2% .
4.
Run a dynamic response analysis using the dyrinp and the dynmas and dynmod files generated from the modal analysis steps.
5.
Check the displacement levels in the joints versus the allowable specified in the dyrinp file.
6.
In case the displacement is more than the specified value for some joint provide minor plate stiffening in the sacinp file and re-run all steps.
7.
Make sure that enough mode shapes are extracted to cover the engine running speeds by 10% extra at least so that any resonance is picked up in the analysis.
Mar 07 Page 76 of 110
3.0
LIFT ANALYSIS (OFFSHORE AND ONSHORE)
Lifting is a necessary operation for offshore structures. For example a Module which has been fabricated in the yard rests on its legpots. Before it can be loaded out onto the barge, loadout beams must be inserted between the module and the legpots. For this purpose the module has to be lifted by a crane and the load out beams introduced in between. In the offshore scenario the deck has to be lifted by the offshore crane and placed on the jacket. Reference Files: Z:\MECHANICAL ENGINEERING \ OEG \ OEG_SACS_Training \SACS\RS-2 \Lift
The difference between Offshore and Onshore lifts lies only in the DAF’s provided and
the absence of rigging platform loads in the Onshore lift.
The DAF’s are as follows:
For Padeye and Padeye Connected members For all Other members
Onshore Lift
Offshore Lift
1.5
2
1.15
1.35
1) Create a copy of the sacinp file and rename it.
2) Remove all deck members (for a jacket lift) by opening the sacinp.* file in Precede and using Joint>Delete and Member>Delete. Rotate the model such that the jacket is
lying
flat
on
the
face
with
no
batter.
This
is
done
by
using
Joint>Translate/Rotate>General and rotating about the appropriate axis to obtain the correct orientation. .
Mar 07 Page 77 of 110
3) Delete the Boat Landing, Riser Guard, risers and all conductors except the preinstalled curved conductors.
4) Remove all the fixities at the end of conductors (Joint>Details/Modify. Enter fixity of joint as 000000.
5) Remove all the loads except dead load and specific NGDL’s. Change the buoyant weights to dry weights. Add the rigging loads in case of offshore lift.
6) In the actual lift calculations are done to ensure that the hook point lies directly above the COG of the structure. The first step is therefore to identify the COG of the structure under the lift condition. In the sacinp.* file create a load case DJMX which combines the computer generated structural dead load, the non-generated dead load, the riser and clamp loads and the curved conductor elastic forces.
7) Run a seastate analysis (SACS Executive>Utils>Sea state Analysis) to determine the COG of the structure for the load case DJMX.
8) Add a lift point, slings depending upon the position of COG and trunnion members. The thumb rule is that the angle that the slings make with the horizontal ○
should roughly be 60 . For the jacket the slings generally attach to trunnions on the rd
th
3 and 4 bay. The correct lift point will be given by the Installation contractor at a later stage of the project.
9) Create two coincident joints at the lift point. A fixity of 110111 is given to one node and a fixity of 111111 is given to the other node. The total lift weight is lets say W. If 50% of the lift weight i.e. W/2 is applied at the joint with the 110111 fixity then by equilibrium the load carried by the other joint ( with 111111 fixity) will be W/2. If this is done then all 4 slings will carry equal weight and this represents what is termed as a 50:50 lift.
Mar 07 Page 78 of 110
10) In some cases due to sling length inaccuracies it is possible that the slings do not carry an equal amount of weight. Hence we also have to analyse the structure for the condition where one pair of diagonally opposite pair of slings carries 75% of the weight with the other diagonally opposite pair carrying the balance 25% of the weight. This is is termed as a 75:25 lift.
11) The slings are given a tension only property.
12) Attach one pair of diagonally opposite slings to one node and the other pair of slings to the other node.
13) As modelled, SACS will not be able to solve the static problem. This is because in case there is even a slight difference in the x and y co-ordinates of the hook point with respect to the structure COG then the corresponding moment arm created by the reaction force on the hook point may cause the structure to have large displacements (i.e. it may cause the jacket/module to swing/twist about. To prevent this add springs at 2 diagonally opposite ends to avoid the rotation of the jacket like a pendulum.
14) Springs will be such that one has stiffness in X & Y and other has stiffness only in Y direction. Spring stiffness = 1 x 10^(4) kN/m
15) Run a Linear Static Analysis to obtain the reaction in the springs. The reactions at the springs should be very less i.e. less than 10kN. Also note the total vertical reaction at the hook point. This value will be used to create the 50:50 lift and the 75:25 lift conditions.
16) Now, the following changes are made in the sacinp file:
Mar 07 Page 79 of 110
(a)
In the LCSEL card only load cases which are to lifted are specified, i.e.
the NGDL’s and dead loads, curved conductor elastic forces, risers and clamp loads, rigging loads etc. (b)
Create a LCOMB DJMX which includes the factors for all the load cases
specified above. LCOMB LCOMB DJMX
(c)
1 1.030 2C 1.133 6A 1.000 4 1.080 3 1.000
Create load conditions say LOADCN 91 representing half the self
weight. This is done by using the reaction obtained at the hook point from the first linear static run This load is applied at the hook point joint which has a fixity of 110111. This creates the 50:50 load case. Similarly create LOADCN’s 92 and 93 representing a 75:25 lift and a 25:75 lift. The extract below is for these load cases. LOADCN 91 LOADLB 91 50-50 DISTRIBUTION LOAD 9998
8850.00
GLOB JOIN 50-50
LOADCN 92 LOADLB 92 75-25 DISTRIBUTION LOAD 9998
13275.0
GLOB JOIN 75-25
LOADCN 93 LOADLB 93 25-75 DISTRIBUTION LOAD 9998 (d)
4425.00
GLOB JOIN 25-75
The next step is to create load combinations between DJMX and load
case 90 along with the appropriate DAF for offshore and onshore lifts. For ex. for an offshore lift:
LCOMB 101 DJMX 1.350 91 1.350 LCOMB 102 DJMX 2.000 91 2.000 LCOMB 103 DJMX 1.350 92 1.350 LCOMB 104 DJMX 1.350 93 1.350
17) Create a jcninp.* file as mentioned in the Inplace analysis. Note that in this jcninp file in the
LCSEL card only the 50-50 case is entered as the other cases are not
Mar 07 Page 80 of 110
mentioned in the API code. The following is an extract from a jcninp.* file used for the RS-2 jacket. JCNOPT API MN TCHORD LCSEL IN
5.1100.00B2 C NID
MAMX
PT
PT
SK8610.0 SK8710.0 JE310.0 JE410.0 102
JSLC SK86 JE4 JE3SK87 RELIEF END
Here LCSEL IN 102 represents the 50:50 lift with check carried out with the DAF’s for padeye connected members.
18) Run a linear Static Analysis. In PSVDB
Reports Joints Reactions,
check
that the reactions in the springs are small (<10KN).
19) Note that since the Joint can check is carried out using load case 102 which uses the DAF’s for padeye connected members, it is possible that some members showing failure may be members that are not connected to padeyes. Check if these members are failing with the lower DAF of 1.35 for non-padeye connected members.
Mar 07 Page 81 of 110
4.0
TRAILER LOAD-OUT ANALYSIS
1. Copy the Lift analysis model and paste the sacinp.* file in the Loadout folder. Open this model in Precede. Remove slings and add load-out beams between the legpots. Ensure that the load out beam is provided offsets at its ends so that it correctly represents its actual length between the two legpots and not the length between the centerlines of the leg pots.
2.
The load out is done with the help of trailers. Under the weight of the structure the soil under the trailers in contact with the tyres is displaced. Hence these trailer support locations can be represented with the help of springs. These springs are added to the model at points falling on the central longitudinal axis of the trailers.
3.
The stiffness of the springs representing the trailers is calculated based on the Subgrade modulus of soil (subgrade modulus of soil at Hazira Yard =20kg/cm2), the number of tyres in contact (or taking the load) and the tyre contact area per tyre 2
(which is usually taken as 0.2x0.3m ). A sample calculation is shown below.
Width: 2.43 m No. of tyres per axle: 8 Capacity per axle: 25T/axle Contact area of one tyre: 0.2 x 0.3 m2 3
Subgrade modulus of soil: 20kg/cm
Maximum reaction at trailer support location: 81.3 T (This reaction can be obtained by fixing the points where we intend to add the trailer springs, running a linear static analysis and computing maximum the reaction).
No. of axles required to share the load below load out beam: 81.3/25 = 4 Total no. of tyres carrying this load: 8 x 4 = 32 Total area of tyre =0.2 x 0.3 x 32 = 1.92 m2 Spring stiffness: 1.92 x 20 x 10 3 = 38400 kg/mm
Mar 07 Page 82 of 110
Stiffness in X and Z direction (10%)= 3840 kg/mm
4.
Apply 10% lateral load in the direction of movement. Apply concentrated loads at the end of the load out beam to consider the leg pots acting at the ends.
5.
Create load cases to account for loss of support conditions by providing support displacements below each trailer location turn by turn.
6.
Run a Linear Static analysis and check the output to make sure that the lateral spring reactions are small.
7.
Provide stub bracings back to the leg in case of failures. Also make sure that offsets are added to the load-out beam before checking the UC ratios.
Mar 07 Page 83 of 110
SEA TRANSPORTATION ANALYSIS Once the jacket/deck has been fabricated and loaded on to the barge it is ready for transportation to the offshore site. During this transportation the barge and consequently is subjected to rolling (rotation about the barge longitudinal axis), pitching (rotation about the barge transverse axis) and heaving (movement in the direction of the vertical). To prevent movement of the cargo during transportation sea fastening is employed. The sea fasteners are generally tubular sections. A transportation analysis is carried out to check adequacy of structural members and sea fasteners subjected to gravity and inertial forces during the transportation process. Reference Files: Z:\MECHANICAL ENGINEERING \ OEG \ OEG_SACS_Training \SACS\RS-2 \Tow
1.
The sea fastening is designed to resist only the environmental loads during the transportation phase. Hence the transportation analysis has to be performed in two stages.
2.
The first analysis considers the structural dead weight alone in the configuration obtained at the end of the load-out operation. Here the structure stands on the barge in a specified position on its legpots (which are pinned for this analysis).
3.
The second analysis considers only environmental loads with all the seafastening members and legpots pinned. The results of the two analyses are then combined prior to members and joint checking.
4.
To carry out the first analysis, open the sacinp.* file and rename it sacinp.static.
5.
The SACS model file for the inplace analysis of the jacket would have been created with the jacket in the vertical standing position as it would be in the in-place onsite condition. However the jacket will be transported in a horizontal position with the face without the batter towards the barge deck and supported on legpots and fastened with seafasteners. Additionally note that the inplace file contains parts of
Mar 07 Page 84 of 110
the structure that will not be transported along with the jacket viz. deck, boat landing, risers and straight conductors.
6.
The first step is therefore to remove the deck, boat landing, risers and straight conductors from the inplace model. Given the difficulty of welding the curved conductors at the offshore location these are not removed from the jacket but are transported along with the jacket.
7.
The next step is to rotate the model so that the correct face (the one without the batter) is at the bottom. This rotation is achieved by opening the sacinp.static file in Precede, clicking on Joint>Translate/Rotate>General. For the RS-2 jacket a 270
○
about the Y axis will result in the batterless North face being the face closest to the barge which is what we want. Click on the Rotation axis as the Y axis and enter a ○
rotation of 270 .
8.
The next step is to add sea fastener members in the load-out model at appropriate locations based on the cargo layout on the barge. Normally 3 fasteners are provided per leg and at any point of time any 2 fasteners must be active in pitch and any two in roll.
9.
To add the sea fastener members create joints using Joint>Add>Relative. Then create a member using Member>Add. Seafasteners are generally tubulars. The section dimensions are normally known from past project experience.
10. By using Members>Details/Modify section properties are assigned to the seafasteners. For example in the RS-2 jacket sea fasteners with an OD of 60.0 cm and a wall thickness of 1.2 have been used.
11. The first run represents the condition where the structure is standing on its leg pots on the barge and take up the entire structural weight. Hence these are pinned.
Mar 07 Page 85 of 110
12. This is done by clicking Joint>Details/Modify, clicking on the joint which connects to the barge and specifying a pinned connection at this joint by providing a Joint Fixity of 111000.
13. Sea fasteners, even if welded on to the structure, must not take up any dead load and hence the joints at which they connect to the barge are not provided with any fixity.
14. Combine all the gravity loads into a single load case, for ex. GRAV by using the LCOMB card at the end of the load definitions.
15. Use an LCSEL card immediately after the OPTIONS card in the sacinp.static file to select this load combination.
16. Also make sure the LDOPT line has a CMB i.e. combine option specified.
17. In SACS Executive run a linear static analysis for GRAV dead load case only. This will generate a saccsf.static file which will be used subsequently. This run concludes the first step of the transportation analysis.
18. For the second stage copy the sacinp.static file, paste it in the same folder and rename this file as sacinp.inertia. Open this file in Datagen, remove the CMB option from the LDOPT line and also remove the LCSEL card which was used for the gravity run.
19. The sacinp.inertia file will be used for the second stage of the transportation analysis in which the sea-fasteners will also come into play to resist the environmental forces.
Mar 07 Page 86 of 110
20. To do this open the sacinp.inertia file in Precede, click on Joint>Details/Modify, select those sea fastener joints which connect to the barge and modify their fixity so as to make them pinned connections (i.e. specify joint fixity as 111000).
21. The next step is to specify the accelerations that the barge will be subjected to. These accelerations cause the inertial loading of the structural members as well as the seafasteners.
22. In the absence of any bid data on barge accelerations, transportation loads may be evaluated based on the criteria published by Noble Denton in their report “General Guidelines for Marine Transportations”. The figure below specifies these critera along with a picture showing the different motions of a floating vessel. Note that in the figure below the heave axis is positive down. 23. Before understanding how to input the correct accelerations in SACS let us first understand the type of accelerations that will be imposed on the cargo in the Barge co-ordinate system as shown in the figure above.
24. The figure below is an illustration of how to compute accelerations due to the roll, pitch and heave motions. Note that in these figures the roll axis is the longitudinal X axis and it goes into the plane of the paper (implying that we are viewing it from the aft portion to fore potion of the barge; the heave axis is the Z axis and is positive down and the sway axis (Y) to the right. The barge can have the following motions ±Roll ± Heave and ±Pitch ± Heave.
Mar 07 Page 87 of 110
Mar 07 Page 88 of 110
○
25. Let us first consider a positive roll of 20 about the X axis and a heave downwards of 0.2g i.e. a positive heave in the barge co-ordinate system. 26. In stillwater acceleration due to gravity is 1.0g down. If the barge were to have a downward heave of 0.2g then the total acceleration on the cargo is 1.2g down. In ○
addition if the barge rolls 20 , then the 1.2g acceleration down creates a component ○
○
1.2gcos(20 ) = 1.128g in the downward direction and a 1.2gsin(20 ) = 0.410g to the right.
27. Now recall that we are doing a two stage analysis. One for gravity and one for environmental loading alone. If we consider the heave direction then the gravity case would have already imposed a loading of 1.0g. This means the positive roll and heave downwards is causing an additional 0.128g DOWNWARDS so that when the gravity and inertia runs are combined the total acceleration applied to the structure in the downward direction would be 1.128g and in the sway direction would be 0.410g.
○
28. Next, let us consider a positive roll of 20 about the X axis and a heave upwards of 0.2g i.e. a negative heave in the barge co-ordinate system.
29. In stillwater acceleration due to gravity is 1.0g down. If the barge were to have a upward heave of 0.2g then the total acceleration on the cargo is 0.8g down. In ○
addition if the barge rolls 20 , then the 0.8g acceleration down creates a component ○
○
0.8gcos(20 ) = 0.752g in the downward direction and a 0.8gsin(20 ) = 0.274g to the right. 30. Again, since we are doing a two stage analysis then in the heave direction the gravity case would have already imposed a loading of 1.0g. This means the positive roll and heave downwards is causing an additional -0.248g UPWARDS so that when the gravity and inertia runs are combined the total acceleration applied to the structure in the downward direction would be 0.752g and in the sway direction would be 0.274g.
Mar 07 Page 89 of 110
31. Similar computations are carried out for the translational accelerations.
○
32. The roll and pitch motions also cause angular accelerations. For example a 20 roll with a period of 10 seconds will cause a angular acceleration of (2π T ) T = roll period,
α
2
α
, (where
2
= roll angle) i.e. 7.896 deg/s . A similar computation is done for
the angular acceleration due to pitching.
33. Now that we have understood the computation of the translational and angular accelerations we proceed to creating the file necessary to specify these values and carry out a tow analysis.
34. The file mentioned above is called a towinp file. In SACS Executive use DataFile>Create New data file>Model>Tow>Select to start creating this towinp file.
35. Note that the acceleration computed above are at the barge center of rotation (COR). The COG of the cargo placed on the barge will, in general, not match with this COR. Hence we have to first specify where the cargo is placed on the barge w.r.t COR. In the ‘Tow Analysis Option’ window, enter the distance (in terms of x, y and z distances) of the barge COR to the SACS origin and not the structure COG. SACS will calculate resultant inertial forces at the structure COG. Click finish. 36. The next step is to specify the roll, sway and heave axes. By default the SACS X axis will be the roll axis, the SACS Y axis will be the sway axis and the SACS Z axis will be the heave axes. Right click>Modify on the TOWOPT line and enter the correct roll, sway and heave axes.
37. Note that the translational accelerations computed above are for the barge coordinate system where the heave axis is positive down. We have to make sure that the correct accelerations are imposed for each type of motion.
Mar 07 Page 90 of 110
38. In case any load factors have to be provided they can be done in the next line. Use Edit>Insert Line>LCFAC(Tow Load Case Factor/Selection)>Select. Enter the load case labels and the relevant factors. For example in the RS-2 jacket the nongenerated structural load (load 2C) and riser/clamp loads (load 4) are provided with a factor of 1.13 to reflect a mill tolerance and contingency of 13%. The other two loads i.e. computer generated structural dead load (load 1) and curved conductor elastic forces (load 6) are provided with a load factor of 1.00.
39. The load cases resulting from barge motion that we will consider for transportation analysis are: 4 load cases from (±Roll ± Heave) and 4 from (±Pitch ± Heave).In the subsequent steps we will designate these load cases with the labels : +R+H, +RH,-R+H, -R-H, +P+H, +P-H, -P+H, -P-H. These labels will be used in a subsequent combine analysis.
40. Next we specify the accelerations computed for the given roll, pitch and heave motions. SACS uses these acceleration values to impose inertial loads (i.e. D’Alemberts forces). Before we do this, let us first understand the concept of a D’Alemberts force in dynamic analysis.
Recall from dynamic analysis methods that to account for accelerations we impose a D’Alembert force in a direction opposite to motion as shown in the figures below.
Mar 07 Page 91 of 110
Here the mass ‘m’ moves by a displacement ‘u’ under the action of external force ‘p(t)’. ‘z(t)’ represents possible base movement in addition to displacement of the mass m. Equilibrium is achieved between the external force with the help of the spring force (f s), damping force (f d) and most importantly the D’Alembert inertial
& & ). This D’Alembert force acts in a direction opposite to motion. force ( mu ○
41. In step 27, we computed that a positive roll of 20 and 0.2g heave downwards result in an additional 0.128g DOWNWARDS so that when the gravity and inertia runs are combined the total acceleration applied to the structure in the downward direction would be 1.128g and in the sway direction would be 0.410g. Given that a)
we are using the SACS positive Z axis as our heave axis
b) SACS converts specified accelerations into D’Alemberts inertial forces in the opposite direction ○
we need to specify the accelerations in SACS correctly so that the roll of 20 and 0.2g heave downwards will cause resultant downward acceleration of 1.128g and an acceleration of 0.410g in the sway direction.
42. Hence if we specify a 0.128g acceleration in the SACS heave (+Z) direction which points upwards, SACS will convert this acceleration into a D’Alembert inertial force downward (i.e. a 0.128g in the downward SACS –Z direction) and when this is combined with the 1.0g from the gravity run, we will get a resultant downward acceleration of 1.128g on the structure.
43. Also notice that the SACS Y axis which we are using as our sway axis is in the opposite direction to the barge sway axis. We computed a 0.410g acceleration acting in the sway direction. If we apply a 0.410g acceleration in the SACS +Y direction then SACS will convert this acceleration into a D’Alembert inertial force so that it is applied on the structure consistent with the direction in which it should ○
act for a a positive roll of 20 and 0.2g heave downwards
Mar 07 Page 92 of 110
○
44. The angular acceleration for a 20 roll with a period of 10 seconds was computed to be 7.896 deg/s2. Let us understand the sign to be input for this angular acceleration.
45. As the barge moves from its mean roll position to its extreme rolled position it moves from a finite angular velocity at mean position to a zero angular velocity at the extreme rolled position. This means it undergoes a negative angular acceleration. If this angular acceleration were positive, then the angular velocity of the barge in the rolled position would be higher than that in the mean position. Since it is not it only means that the barge is decelerating. Hence a positive roll about the roll (X) axis will cause a negative angular acceleration about the X axis. ○
i.e. a postive 20 roll with a period of 10 seconds will cause an angular acceleration 2
of -7.896 deg/s about the SACS X axis.
46. During a negative roll, the barge is still decelerating from its mean position to the other rolled position. However this deceleration is in the opposite direction (i.e. it is a deceleration about the –X axis) when compared to the positive roll deceleration. This is nothing but a positive acceleration about the X axis. Hence a negative roll about the roll (X) axis will cause a positive angular acceleration about the X axis. ○
i.e. a postive 20 roll with a period of 10 seconds will cause an angular acceleration 2
of 7.896 deg/s about the SACS X axis.
47. The translational and angular accelerations for the remaining motions as specified likewise.
48. The extract below is from a towinp.* file used for the RS-2 transportation analysis.
TOWOPT MN
MP OR
51. -2.12
XYZ
ACCL +R+H
-7.89
0.410
0.128
ACCL +R-H
-7.89
0.274
- 0.248
ACCL -R+H
7.89
-0.410
0.128
ACCL -R-H
7.89
-0.274
-0.248
Mar 07 Page 93 of 110
ACCL +P+H
-4.935
-0.259
0.172
ACCL +P-H
-4.935
-0.173
- 0.219
ACCL -P+H
4.935
0.259
0.172
ACCL -P-H
4.935
0.173
-0.219
49. The computed accelerations have been specified using the ACCL card (Edit>Insert Line>ACCL>Select.
50. Go through the 8 ACCL cards above and make sure that you understand the magnitudes of all these accelerations, the direction (hence the sign) in which they have been applied and given that SACS will generate the D’Alembert inertia forces for each, whether they will reflect actual accelerations imposed on the cargo after they have been combined with the gravity run.
51. Once all the 8 inertia load cases have been entered in the towinp.* file, type END and save this file.
52. Enter ‘inertia’ as the label in SACS Executive and run a Tow Analysis (Runfile wizard > Load > Tow / Transportation Inertia Loads > Start Wizard) by choosing the towinp.* file and the sacinp.inertia. The output files are towlst.inertia and saccsf.inertia. This concludes the second stage of the transportation analysis.
53. The next step is to combine the results from the gravity run and the inertia run. This is done by creating a cmbinp.* file. In SACS Executive click Data File>Create New Data File>Utils>Select. Skip the title window. Make sure that the units are set to ‘MN’ and then click Finish.
54. The next step is to create load cases which combine the primary (inertia) and secondary (gravity) runs. For example the first load case is created with the +Roll+Heave as a primary case combined with the gravity as secondary case. In the cmbinp.* file (use Edit>Insert Line>LCOND). In the ‘Output load combination
Mar 07 Page 94 of 110
window’ enter a load combination label (101), with a ‘LIN’ – linear algebraic sum load combination type and a stress modifier of 1.33 (i.e. 33% increase).
55. Next, use Edit>Insert Line>COMP(Input Load Component)>Select to specify the primary and secondary loads in this load. In the subsequent ‘Input Load Component window’ enter the label of the contributing load case name i.e. +R+H (this was the load label given to the +Roll+Heave case in a previous step), click ‘PRIM’ as the ‘Load Case Source’ to specify this as a primary load and finally enter a load case factor of 1.0.
56. Again use >Insert Line>COMP(Input Load Component)>Select. This time enter the first load which contributes as the secondary load case. Here the ‘computer generated structural dead weight’ was given a label ‘1’. Enter this as the label and click ‘SECD’ to specify this as a secondary load. The load factor for this is 1.0. In a similar manner specify the other secondary loads i.e. non-generated structural load (load 2C) and riser/clamp loads (load 4) factor of 1.13 to reflect a mill tolerance and contingency of 13% and curved conductor elastic forces (load 6) with a factor of 1.0
57. All these steps are repeated to create the primary and secondary combinations for the +R-H,-R+H, -R-H, +P+H, +P-H, -P+H, -P-H. The extract from the cmbinp.* file is reproduced below LCOND 101 LIN 1.333 COMP P+R+H COMP S 1 COMP S 2C
1.00 1.00 1.13
COMP S 4
1.13
COMP S 6
1.00
LCOND 102 LIN 1.333 COMP P+R-H COMP S 1 COMP S 2C COMP S 4
1.00 1.00 1.13 1.13
Mar 07 Page 95 of 110
COMP S 6
1.00
LCOND 103 LIN 1.333 COMP P-R+H COMP S 1 COMP S 2C
1.00 1.00 1.13
COMP S 4
1.13
COMP S 6
1.00
LCOND 104 LIN 1.333 COMP P-R-H COMP S 1 COMP S 2C
1.00 1.00 1.13
COMP S 4
1.13
COMP S 6
1.00
LCOND 105 LIN 1.333 COMP P+P+H COMP S 1 COMP S 2C
1.00 1.00 1.13
COMP S 4
1.13
COMP S 6
1.00
LCOND 106 LIN 1.333 COMP P+P-H COMP S 1 COMP S 2C
1.00 1.00 1.13
COMP S 4
1.13
COMP S 6
1.00
LCOND 107 LIN 1.333 COMP P-P+H COMP S 1 COMP S 2C
1.00 1.00 1.13
COMP S 4
1.13
COMP S 6
1.00
LCOND 108 LIN 1.333 COMP P-P-H COMP S 1 COMP S 2C
1.00 1.00 1.13
COMP S 4
1.13
COMP S 6
1.00
Mar 07 Page 96 of 110
58. Once the 8 combine cases are created type the ‘END’ card and save the cmbinp.* file.
59. Run
a
combine
Solution
file
Analysis
(SACS
Executive>Runfile
Wizard>Utils>Combine Solution File) and give the cmbinp.* file, the saccsf.inertia (Specified as the Primary) and the saccsf.static (specified as the Secondary) as the input files and saccsf.combined as the output file name.
60. Create a pstinp.* file (In SACS Executive, Data File>Create New Data file>Post>Post>Select. In the Post Options window specify the following options: Modification or Extraction Option-MOD, Local Buckling, Skip Member Sort and Execute. In the ‘Load Case Selection’ window specify the load cases include those load cases which were given in the cmbinp.* file.
61. In SACS Executive run a post analysis using Runfile Wizard>Post>Element Stress & Code Check. Select the saccsf.combined file and generate the pstlst.* file (to check the UC ratios) and the pstcsf.* file. Alternatively run the Generate postvue analysis on the saccsf.combined file to create the psvdb folder to view the UC ratios graphically (Runfile Wizard>Post>Generate Postvue Database. Check Create PostvueDB and Use Post Input file. Select the pstcsf.* file and the sacinp.inertia as the model file and Run.
62. Run joint can check analysis using the saccsf.combined file to complete the analysis.
63. If any changes are made in the input, changes must be made in both the gravity as well as the inertia input file. Then create a Multirun file. In the menu bar File Run Multiple Select sequence as (a) Stat.run (b) Tow.run (c) Cmb.run
Mar 07 Page 97 of 110
(d) Pstinp.run (e) Pvi.run (f) Jcn.run
14. Run the multirun file and then check results as before.
Mar 07 Page 98 of 110
5.0
FLOATATION AND UPENDING ANALYSIS
1.
The above analysis is performed to check the reserve buoyancy and the hook loads.
2.
Use the Lift analysis model and remove all sling data from the file.
3.
Create a fltinp file and specify sling location, sling parameters, buoyancy tank data (if required), leg definitions and the various elevations of the hooks for which the analysis is required.
4.
Run the Floation & Upending analysis and open the fltnpf file to check graphically the reserve buoyancy and the sling loads / hook loads for the various positions analysed.
6.0
REFERENCES i.
ii.
iii. iv. 5.
Octa Engineers (2003), Technical Notes for Structural Design of Offshore Topsides and Jackets (pp 81-82) Recommended Practice for Planning, Design and Constructing Fixed Offshore Platforms – Workind Stress Design (RP-2A, Dec. 2000), Section 2.3.1.b.7, pp15 Dynamics of Fixed Marine Structures – Barltrop N.D.P, rd Adams A. J, 3 Edn, Section 2.6.1, pp 32 NPP2 Jacket Earthquake Analysis Report – Report 6-21 (Ocean Engineering Group) SACS IV Manual, Release 5 Revision 10, Section 2.7.2.2, pp 41
Mar 07 Page 99 of 110
Appendix A : Figures
Mar 07 Page 100 of 110
Figure 1. Typical Jacket Structures
Mar 07 Page 101 of 110
Figure. 2 : Components of a Jacket platform
Mar 07 Page 102 of 110
Figure. 3: Typical Bracing Configurations
Figure 4. Piling arrangements
Mar 07 Page 103 of 110
Figure. 5 Typical Deck Structure
Mar 07 Page 104 of 110
Figure. 6: Selection of wave theory
Mar 07 Page 105 of 110
Figure. 7: Module being lifted offshore
Mar 07 Page 106 of 110
Figure. 8: Trailer load out
Mar 07 Page 107 of 110
Figure 9. Skidded Loadout
Mar 07 Page 108 of 110
Figure 10: Vessel motions
Figure. 11 : Launching and Upending of Jacket
Mar 07 Page 109 of 110
Appendix B – Tables