# Subsea Foundation

A foundation is a structure that transfers loads to the earth. Foundations are generally broken into two categories: shallow foundations and deep foundations. A subsea production structure may be supported by piles, by nudmats with skirt or directly by the seabed. It could also be supported by a combination of these three structures.

## Contents

- 1 Pile- or Skirt-Supported Structures
- 2 Seabed-Supported Structures
- 3 Pile and Plate Anchor Design and Installation
- 4 Geotechnical Capacity of Suction Piles
- 5 Geotechnical Capacity of Plate Anchors
- 6 Structural Design of Suction Piles
- 7 Structural Design Criteria
- 8 Installation of Suction Piles, Suction Caissons, and Plate Anchors
- 9 Plate Anchors
- 10 Driven Pile Anchor
- 11 Fatigue Design
- 12 References

## Pile- or Skirt-Supported Structures

The foundation piles of a pile-supported structure should be designed for compression, tension, lateral loads, and also shear stress as applicable. The structure is properly connected to the pile/skirt. This can be accomplished with a mechanical device or by grouting the annulus between the pile and sleeve.

## Seabed-Supported Structures

The foundation of a seabed-supported structure is designed to have sufficient vertical and horizontal bearing capacity for the loads in question. Depending on seabed conditions, high contact stresses may develop. This should be considered in the design. Underbase grouting may have to be used to achieve the required stability and load distribution on the seabed.

## Pile and Plate Anchor Design and Installation

### Basic Considerations

The technology for the evaluation of the geotechnical capacity of a suction pile and plate anchor is still under development; therefore, specific and detailed recommendations cannot be given at this point. Instead, general statements are used to indicate that consideration should be given to some particular points and references are given. Designers are encouraged to utilize all research advances available to them. It is hoped that more specific recommendations can be issued after completion of the research in this area.

## Geotechnical Capacity of Suction Piles

### Basic Considerations

Suction pile anchors resist vertical uplift loads by various mechanisms depending on the load type:

- Storm loading:
- External skin friction;
- Reverse end bearing (REB) at the tip of the piles;
- Submerged weight of the anchor.

- Long-term loop current loading:
- External skin friction appropriately reduced for creep and cyclic effects due to long-term duration of the event;
- Reduced value of reverse end bearing;
- Submerged weight of the anchor.

- Pretension loading:
- External skin friction;
- Smaller of internal skin friction or soil plug weight;
- Submerged weight of anchor.

- Suction piles resist horizontal loads by the following mechanisms:
- Passive and active resistance of the soil;
- External skin friction on the pile wall sides (as appropriate);
- Pile tip shear.

The geotechnical load capacity of the anchor should be based on the lower bound soil strength properties. This is derived from the site-specific soil investigation and interpretation. Anchor adequacy with respect to installation should be based on the upper bound soil strength properties. Because this geometry may change the relationship between the horizontal and vertical anchor loads, the impact of the mooring line geometry in the soil on the anchor loads should be considered. For example, the inverse catenary of the mooring line in the soil may make the mooring line angle steeper at the anchor. This steeper angle could result in a reduced horizontal anchor load, but an increased vertical anchor load. Both an upper and lower bound inverse catenary should be checked to ensure that the worst case anchor loading is established. Axial safety factors take into consideration the fact that piles are primarily loaded in tension and are therefore higher than for piles loaded in compression. As with other piled foundation systems, the calculated ultimate axial soil resistance should be reduced if soil setup, which is a function of time after pile installation, will not be complete before significant loads are imposed on the anchor pile. Because the lateral failure mode for piles is considered to be less catastrophic than vertical failure, lower factors of safety have been recommended for lateral pile capacity. Use of different safety factors for vertical and lateral pile capacities may be straightforward for simple beam-column analysis of, for example, mobile moorings, but more complex methodologies do not differentiate between vertical and lateral pile resistance.

### Analysis Method

It is recommended that suction pile design for permanent moorings use advanced analysis techniques such as limit equilibrium analysis or finite element analysis of the pile and adjacent soil. For example, the resultant capacity may be calculated with limit equilibrium methods, where the circular area is transformed into a rectangle of the same area and width equal to the diameter, and with 3D effects accounted for by side shear factors. For mobile moorings, simpler beam column analysis using load transfer displacement curves (i.e., P-y, T-z, Q-z) described in API RP 2A are considered adequate if suitably modified. In areas where tropical cyclonic storms may exceed the capacity of the mobile mooring or anchoring system, such as the Gulf of Mexico, the design of suction piles should consider an anchor failure mode that reduces the chance of anchor pullout. For example, the mooring line anchor point can be located on the suction anchor such that the anchor does not tilt during soil failure, but “plows” horizontally in a vertical orientation.

## Geotechnical Capacity of Plate Anchors

### Basic Considerations

The ultimate holding capacity is usually defined as the ultimate pull-out capacity (UPC), which is the load for the soil around the anchor reaching failure mode for plate anchors. For anchors that dive with horizontal movement, the ultimate holding capacity is reached when excessive lateral drag distance has occurred. At UPC, the plate anchor starts moving through the soil in the general direction of the applied anchor load with no further increase in resistance or the resistance starts to decline. The ultimate pullout capacity of a plate anchor is a function of the soil’s undrained shear strength at the anchor fluke, the projected area of the fluke, the fluke shape, the bearing capacity factor, and the depth of penetration. The disturbance of the soil due to the soil failure mode should be considered when analyzing the plate anchor’s ultimate pull-out capacity. This mode is generally accounted for in the form of a disturbance factor or capacity reduction factor. The bearing capacity factor and disturbance factor should be based on reliable test data, studies, and references for such anchors. The plate anchor’s penetration is usually in a range of two to five times the fluke width, depending on the undrained shear strength of the soil, in order to generate a deep failure mode. If the final depth does not generate a deep failure mode, a suitable reduction in bearing capacity factor should be used. Plate anchors get their high holding capacity from being embedded into competent soil. Therefore, it is important for the anchor’s penetration depth to be established during the installation process. Furthermore, a plate anchor gets its high ultimate pull-out capacity by having its fluke oriented nearly perpendicular to the applied load. To ensure that the fluke will rotate to achieve a maximum projected bearing area, the plate anchor design and installation procedure should:

- Facilitate rotation of the fluke when loaded by environmental loads or during installation or both;
- Ensure that no significant or unpredicted penetration is lost during anchor rotation, which may move the fluke into weaker soil;
- Have the structural integrity to allow such fluke rotation to take place during installation and keying operations or while subject to the ultimate

pull-out capacity load. Depending on the type of plate anchor and its installation orientation, this item may also apply to fluke rotation about both horizontal and vertical axes. As appropriate, the anchor capacities should be reduced to account for anchor creep under long-term static loading and cyclic degradation.

### Prediction Method for a Drag Embedded Plate Anchor

Three aspects of drag embedded plate anchor performance require prediction methods:

- Anchor line mechanics;
- Installation performance;
- Holding capacity performance.

All three mechanisms are closely linked and influence one another, as explained next.

### Prediction Method for Direct Embedded Plate Anchor

Anchor capacity determination for direct embedded plate anchors is identical to that shown for drag embedded anchors with the following exceptions:

- Final penetration depth is accurately known.
- Nominal penetration loss during keying should be included (usually taken as 0.25 to 1.0 times the fluke’s vertical dimension, depending on the shank and keying flap configuration).
- Calculation of the effective fluke area should use an appropriate shape factor and projected area of the fluke with a keying flap in its set position.

Safety factors for drag embedded plate anchors are higher because overloading of the anchor normally results in the anchor pulling out, whereas a drag anchor may drag horizontally or dig in deeper, developing constant or higher holding capacity under a similar situation. For plate anchors that exhibit overloading behaviors similar to those of drag anchors, consideration may be given to using drag anchor safety factors, assuming the behavior can be verified by significant field tests and experience.

## Structural Design of Suction Piles

### Basic Considerations

The purpose of this article is to provide guidance and criteria for the structural design of suction piles. Some of the guidance and criteria are also applicable to driven piles.

### Design Conditions

The suction pile structure should be designed to withstand the maximum loads applied by the mooring line, the maximum negative pressure required for anchor embedment, the maximum internal pressure required for anchor extraction, and the maximum loads imposed on the anchor during lifting, handling, launching, lowering, and recovery. The fatigue lives of critical components and highly stressed areas of the anchor should be determined and checked against the required minimum fatigue life. Mooring Loads on Global Anchor Structure The load case that provides the maximum horizontal and vertical loads at the mooring padeye should be used for the global structural design of the anchor. The soil reactions generated by the geotechnical analysis will be used in these calculations. Sensitivity checks should be performed to ensure that a load case with less than the maximum load, but applied at a more onerous angle at the padeye, does not control the design. Mooring Loads on Anchor Attachment The mooring line attachment padeye or lug is a critical structural component. To meet fatigue resistance criteria, the padeye is often an integral cast lug and base structure. This avoids the use of heavy weldments, which can result in a lower fatigue life. The attachment padeye should be designed to satisfy both strength and fatigue requirements. The padeye should be designed for the controlling design load with an appropriate factor of safety. Designing the padeye for a maximum load equal to a factor times the break strength of the mooring line may lead to a significantly overdesigned padeye, which may not integrate well with the anchor shell and backup structures. The mooring line padeye should be designed for the controlling load case, and ensitivity checks should be performed to ensure that a load case with less than the maximum load but applied at a more onerous angle does not control the design. The orientation of the applied load at the padeye will be affected by the inverse catenary of the mooring line, vertical misalignment due to anchor tilt, and rotational misalignment due to deviation from the target orientation. These factors should be properly accounted for. Embedment Loads For anchor embedment, the estimated upper bound suction pressure required to embed the anchor to its design penetration should be used for the design of the anchor wall and anchor cap structure. However, the maximum suction pressure used should not be higher than the suction at which internal plug uplift occurs. Extraction Loads With respect to anchor extraction, two conditions require evaluation:

- Temporary condition: Extraction of a suction pile may be required for permanent moorings. For example, after all suction piles have been preinstalled along with the mooring lines, one of the mooring lines is accidentally dropped to the seabed and damaged during the hookup operation with the vessel. At this time, a decision to extract the suction pile and recover the mooring leg may be made. Typically, such situations may occur 30 to 60 days after the first suction pile has been installed. For mobile moorings, the suction piles are often extracted at the end of the current drilling or testing operation and reused in other locations.

- Terminal condition: The suction piles for a permanent mooring may be extracted at the end of their service life. The estimated maximum internal pressure required to extract the anchor for these two situations should be used for the design of the anchor wall and anchor cap structure. However, the maximum extraction pressure used should not be higher than the pressure causing overload of soil-bearing capacity at the anchor tip. The vessel removing the anchor is often capable of applying a lifting force on the anchor with the recovery line. This assistance can significantly reduce the required extraction pressure and therefore should be included in the removal analysis. Transportation and Handling of Loads The suction pile structure and its installation appurtenances should be designed for the maximum loads generated during suction pile handling, transportation, lifting, pending, lowering, and recovery. The suction pile designer should interface closely with the installation contractor when determining these load ases. Design of appurtenances for these load cases is typically performed using the installation contractor’s in-house design guidelines or other recognized codes. Nevertheless, all lifting appurtenances and their supporting structures should meet the minimum requirements of API RP 2A.

### Structural Analysis Method

Pile analysis in accordance with API RP 2A is appropriate for piles with diameter-to-thickness ratios (D/t) of less than 120. For cylindrical piles with D/t ratios exceeding 120, it is recommended that a detailed structural finite element model be developed for the global structural anchor analysis to ensure that the anchor wall structure and appurtenances have adequate strength in highly loaded areas. Supplementary manual calculations may be appropriate for members or appurtenances subjected to local loading.

### Space Frame Model

A space frame model generally consists of beam elements plus other elements needed to model specific structural characteristics. This is appropriate for piles with D/t ratios of less than 120 and for preliminary design of the top cap or padeye backup structures on large-diameter piles (i.e., D/t > 120).

### Finite Element Model

Finite element analysis is recommended for the global shell structure, top cap plate and supporting members, and the padeye backup structure for piles with D/t > 120. Complex shapes such as the padeye casting or welding should also be analyzed by finite element method.

### Manual Calculations

Manual calculations using empirical formulas and basic engineering principles can be performed when detailed finite element analysis is not needed. Stress Concentration Factors Stress concentration factors can be determined by detailed finite element analysis, physical models, and other rational methods or published formulas.

### Stability Analysis

Formulas for the calculation of the buckling strength of structural elements are presented in API Recommended Practice 2A; API Bulletin 2U, “Stability Design of Cylindrical Shells”; and API Bulletin 2V, “Design of Flat Plate Structures”. As an alternative, buckling and postbuckling analysis or model tests of specific shell or plate structures may be performed to etermine buckling and ultimate strength.

### Dynamic Response

Significant dynamic response is not expected for the anchor in its in-place condition; therefore, anchor structures are often analyzed statically. Transportation analysis, however, will typically include dynamic loads generated by harmonic motions of a simple single-degree-of-freedom model.

## Structural Design Criteria

### Design Codes

The method for structure design is the working stress design method, where stresses in all components of the structure are kept within specified values. In general, cylindrical shell elements should be designed in accordance with API RP 2A for D/t ratios of less than 300 or API Bulletin 2U when D/t exceeds 300, flat plate elements in accordance with API Bulletin 2V, and all other structural elements in accordance with API RP 2A, as applicable. In cases where the structure’s configurations or loading conditions are not specifically addressed by these codes, other accepted codes of practice can be used. In this case, the designer must ensure that the safety levels and design philosophy implied in the API Recommended Practice 2SK are adequately met. In API RP 2A, allowable stress values are expressed, in most cases, as a fraction of the yield or buckling stress. In API Bulletin 2U, allowable stress values are expressed in terms of critical buckling stresses. In API Bulletin 2V, the allowable stresses are classified in two basic limit states: ultimate limit states and serviceability limit states. Ultimate limit states are associated with the failure of the structure, whereas serviceability limit states are associated with adequacy of the design to meet its functional requirements. For the purpose of suction anchor design, only the ultimate limit state is considered. Safety Categories There are two safety categories: Category A safety criteria are intended for normal design conditions, and Category B safety criteria are intended for rarely occurring design conditions. Allowable Stresses For structural elements designed in accordance with API RP 2A, the allowable stresses recommended in these codes should be used for normal design conditions associated with Category A safety criteria. For extreme design conditions associated with Category B safety criteria, the allowable stresses may be increased by one-third if the working stress design method is utilized (e.g., API RP 2A-WSD. For shell structures designed in accordance with API Bulletin 2U, a factor of safety equal to 1.67J is recommended for buckling modes for Category A safety criteria. For Category B safety criteria, the corresponding factor of safety is equal to 1.25J. The parameter J varies with buckling stress and is defined in API Bulletin 2U. It is equal to 1.2 for elastic buckling stresses at the proportional limit and reduces linearly for inelastic buckling to 1.0 when the buckling stress is equal to the yield stress. For flat plate structures designed in accordance with API Bulletin 2V, the allowable stress is obtained by dividing the ultimate limit state stress by an appropriate factor of safety, which is 2.0 for Category A safety criteria and 1.5 for Category B safety criteria. For cylindrical elements with D/t ratios exceeding 120, it is recommended that global strength be analyzed using finite element techniques. Local buckling formulations for axial compression, bending, and hydrostatic pressure given in API RP 2A for D/t <300 and API Bulletin 2U D/t 300 are considered valid if due consideration is made for variable wall thicknesses (when it occurs) and buckling length (which may extend below the mudline when performing suction embedment analysis). The nominal Von Mises (equivalent) stress at the element’s extreme fiber should not exceed the maximum permissible stress as calculated below:

## Installation of Suction Piles, Suction Caissons, and Plate Anchors

### Suction Piles and Suction Caissons

Installation Procedure, Analysis, and Monitoring Installation procedures should be developed and installation analyses should be performed for suction pile and suction caisson anchors to verify that the anchors can penetrate to the design depth. The installation analysis should also consider anchor retrieval for the following cases:

- Mobile moorings where anchor removal is needed for reuse of the anchor or to clear the seabed. The suction pile retrieval procedures and analysis should account for the estimated maximum setup time.
- Permanent moorings where it is required by authorities that the anchors be removed after the system service life. The suction pile retrieval procedures and analysis should be based on full soil consolidation. For suction pile embedment analysis, the risk of causing uplift of the soil plug inside the anchor should be considered. The allowable underpressure to avoid uplift should exceed the required embedment pressure by a factor

of 1.5. Anchor installation tolerances should be established and considered in the suction pile anchor geotechnical, structural, and installation design process. The following typical tolerances should be considered:

- Allowable anchor tilt angle in degrees;
- Allowable deviation from target orientation of the mooring line attachment to limit padeye side loads and rotational moments on the

anchor;

- Minimum penetration required to achieve the required holding capacity. Suction pile installation analysis should provide the relevant data needed

for the suction pile design and installation procedures. The following typical information is required:

- Anchor self-weight penetration for applicable soil properties or range of properties;
- Embedment pressure versus penetration depth for applicable soil properties;
- Allowable embedment pressure to avoid plug uplift;
- Penetration rate;
- Estimated internal plug heave.

To verify that the suction pile installation is successful and in agreement with the design assumptions, the following data should be monitored and recorded during the installation of suction piles:

- Self-weight penetration;
- Embedment pressure versus penetration depth;
- Penetration rate;
- Internal plug heave (by direct or indirect means);
- Anchor heading and anchor tilt in degrees;
- Final penetration depth.

For the installation of temporary mooring suction pile anchors, measurement of the internal plug heave is not required if the anchor reaches its design embedment depth. Skirt Penetration of Suction Caissons The following points should be given attention when designing the skirt penetration capability of suction caissons:

- The skirt penetration resistance should account for the reduced shear strength along the skirt wall due to the disturbance during penetration.

Normally, the remolded shear strength is applied.

- Stiffeners (outside and inside) may influence the penetration resistance, because additional force may be required to penetrate them, and the

failure mode around internal stiffeners should be given attention. On the other hand, a gap may form above outside and inside stiffeners. This may reduce the penetration resistance and form potential flow paths. In the case of several ring stiffeners, clay from the upper part of the profile may be trapped between the stiffeners and give low resistance at larger depth. In stiff clays the soil plug may stand open, giving essentially no skin friction along the inside wall.

- The allowable underpressure for penetration should be calculated as the sum of the inverse bearing capacity at skirt tip and the internal skirt wall

friction. There is some controversy as to whether the conventional bearing capacity factor can be used to calculate the end bearing capacity below the skirt tip, but most designers tend to assume conventional bearing capacity factors.

- If the outside or inside suction caisson skirt wall is given surface treatment (e.g., painting), this may cause reduction in the skirt wall friction,

which should be taken into account in the calculations.

## Plate Anchors

### Direct Embedded Plate Anchors

Direct embedment of plate anchors can be achieved by suction, impact or vibratory hammer, propellant, or hydraulic ram. The suction embedded plate anchor (SEPLA) has been used for major offshore mooring operations. As an example, the SEPLA uses a so-called suction follower, which is essentially a reusable suction anchor with its tip slotted for insertion of a plate anchor. The suction follower is immediately retracted by reversing the pumping action once the plate anchor is brought to the design depth, and can be used to install additional plate anchors. In the SEPLA concept, the plate anchor’s fluke is embedded in the vertical position and necessary fluke rotation is achieved during a keying process by pulling on the mooring line. Installation procedures should be developed and installation analyses should be performed for direct embedded plate anchors to verify that the anchors can penetrate to the design depth. The installation analysis should also consider plate anchor retrieval if applicable. For the embedment analysis, the risk of causing uplift of the soil plug inside the suction embedment tool should be considered. The allowable underpressure to avoid uplift should exceed the required embedment pressure by a factor of 1.5. Plate anchor installation tolerances should be established and considered in the anchor’s geotechnical, structural, and installation design process. The following typical tolerances should be considered:

- Allowable deviation from target heading of the mooring line attachment to limit padeye side loads and rotational moments on the anchor padeye;
- Minimum penetration required before keying or test loading to achieve the required holding capacity;
- Allowable loss of anchor penetration during plate anchor keying or test loading. Suction embedment analysis should provide relevant data needed for the design and installation procedures, which should allow verification of the assumptions used in the anchor design. The following typical information is required:
- Suction embedment tool self-weight penetration for applicable soil properties or range of properties;
- Embedment pressure versus penetration depth for applicable soil properties;
- Allowable embedment pressure to avoid plug uplift;
- Penetration rate;
- Estimated internal plug heave. To verify that the plate installation is successful and in agreement with the assumptions in design, the following data should be monitored and recorded during the installation of the suction embedment tool to verify the assumption used in design:
- Self-weight penetration;
- Embedment pressure versus penetration depth;
- Penetration rate;
- Internal plug heave, if it is expected that plug heave could be a concern;
- Anchor orientation;
- Final penetration.

### Drag Embedded Plate Anchors

For drag embedded plate anchors used in permanent moorings, the installation process should provide adequate information to ensure that the anchor reaches the target penetration, and that the drag embedment loads are within the expected load range for the design soil conditions. The following typical information should be monitored and verified:

- Line load in drag embedment line;
- Catenary shape of embedment line based on line tension and line length to verify that uplift at the seabed during embedment is within allowable

ranges and to verify anchor position;

- Direction of anchor embedment;
- Anchor penetration.

### Test Loading of Anchors

For suction piles, suction caissons, and plate anchors, the installation records should demonstrate that the anchor penetration is within the range of upper and lower bound penetration predictions developed during the anchor geotechnical design. In addition, the installation records should confirm the installation behavior, that is, self-weight penetration, embedment pressures, and drag embedment loads and that the anchor orientation is consistent with the anchor design analysis. Under these conditions, test loading of the anchor to a full intact storm load should not be required. Plate anchors should be subjected to adequate keying loads to ensure that sufficient anchor fluke rotation will take place without further loss of anchor penetration. The keying load required and amount of estimated fluke rotation should be based on reliable geotechnical analysis and verified by prototype or scale model testing. The keying analysis used to establish the keying load should also include analysis of the anchor’s rotation when subjected to the maximum intact and one-line damage survival loads. If the calculated anchor rotation during keying differs from the anchor rotation in survival conditions, then the anchor’s structure should be designed for any resulting out-of-line loading to ensure that the anchor’s structural integrity is not compromised. In cases where the installation records show significant deviation from the predicted values and these deviations indicate that the anchor holding capacity may be compromised, test loading of the anchor to the maximum dynamic intact load may be required and may be an acceptable option to prove holding capacity for temporary moorings. However, testing anchors to the maximum intact load does not necessarily prove that required anchor holding safety factors have been met, which is of special concern for permanent mooring systems. Consequently, if the installation records show that the anchor holding capacity is significantly smaller than calculated and factors of safety are not met, then other measures to ensure adequate factors of safety should be considered:

- Additional soil investigation at the anchor location to establish and/or confirm soil properties at the anchor site;
- Retrieval of the anchor and reinstallation at a new undisturbed location;
- Retrieval of the anchor, redesign and reconstruction of the anchor to meet design requirements, and reinstallation at a new undisturbed location;
- Delay of vessel hookup to provide additional soil consolidation.

## Driven Pile Anchor

### Basic Considerations

Driven pile anchors provide a large vertical load capacity for taut catenary mooring systems. The design of driven pile anchors builds on a strong industry background in the evaluation of geotechnical properties and the axial and lateral capacity prediction for driven piles. The calculation of driven pile capacities, as developed for fixed offshore structures, is well documented in API RP 2A. The recommended criteria in API RP 2A should be applied for the design of driven anchor piles, but with some modifications to reflect the differences between mooring anchor piles and fixed platform piles. The design of a driven pile anchor should consider four potential failure modes:

1. Pull-out due to axial load;

2. Overstress of the pile and padeye due to lateral bending;

3. Lateral rotation and/or translation;

4. Fatigue due to environmental and installation loads. The safety factors of holding capacity is defined as the calculated soil resistance divided by the maximum anchor load from dynamic analysis.

### Geotechnical and Structural Strength Design

In most anchor pile designs, the mooring line is attached to the pile below the seafloor, to transfer lateral load to stronger soil layers. As a result, the design should consider the mooring line angle at the padeye connection resulting from the “reverse catenary” through the upper soil layers. Calculation of the soil resistance above the padeye location should also consider remolding effects due to this trenching of the mooring line through the upper soil layers. Driven pile anchors in soft clay typically have aspect ratios (penetration/ diameter) of 25 to 30. Piles having such an aspect ratio would be fixed in position about the pile tip and consequently would deflect laterally and fail in bending before translating laterally as a unit. Driven pile anchors are typically analyzed using a beam-column method with a lateral loaddeflection model (p-y curves) for the soil. These computations should include the axial loading in the pile, as well as the mooring line attachment point, which will influence the deflection, shear, and bending moment profiles along the pile. Pile stresses should be limited to the basic allowable values in API RP 2A under intact conditions. Basic allowable stresses may be increased by one-third for rarely occurring design conditions such as a one-line damaged condition. Because an anchor pile near its ultimate holding capacity experiencing the largest deflection always engages new soil, “static” p-y curves may be considered for calculation of lateral soil resistance. “Cyclic” p-y curves may be more appropriate for fatigue calculations, because the soil close to the pile will be more continually disturbed due to smaller, cyclic deflections. The p-y curve modifications developed by Stevens and Audibert are recommended in place of the API RP 2A p-y curves, to obtain more realistic deflections. Consideration should be given to degrading the p-y curves for deflections greater than 10% of the pile diameter. In addition, when lateral deflections associated with cyclic loads at or near the mudline are relatively large (e.g., exceeding yc as defined in API RP 2A for soft clay), consideration should be given to reducing or neglecting the soil-pile adhesion (skin friction) through this zone. The design of driven anchor piles should consider typical installation tolerances, which may affect the calculated soil resistance and the pile structure. Pile verticality affects the angle of the mooring line at the padeye, which changes the components of the horizontal and vertical mooring line loads that the pile must resist. Underdrive will affect the axial pile capacity and may result in higher bending stresses in the pile. Padeye orientation (azimuth) may affect the local stresses in the padeye and connecting shackle. Horizontal positioning may affect the mooring cope and/or angle at the vessel fair lead, and should be considered when balancing mooring line pretensions.

## Fatigue Design

### Basic Considerations

Anchor piles should be checked for fatigue caused by in-place mooring line loads. Fatigue damage due to pile driving stresses should also be calculated and combined with in-place fatigue damage. For typical mooring systems, fatigue damage due to pile driving is much higher than that caused by inplace mooring line loads. In-Place Loading A global pile response analysis accounting for the pile–soil interaction should be carried out for the mooring line reactions due to the fatigue sea states acting on the system. The local stresses that accumulate fatigue damage in the pile should be obtained by calculating a stress concentration factor (SCF), relative to the nominal stresses generated by the global analysis, at the fatigue critical locations. These locations are typically at the padeye, at the girth welds between the padeye and the pile, and between subsequent pile cans. The evaluation of SCFs for girth welds needs to account for the local thickness misalignment at the weld. Note that the calculated SCF needs to be corrected by the ratio of the nominal thickness used in the pile response analysis to the lesser of the pile wall thicknesses joining at the weld. The SCF is to be applied to the nominal pile stress range obtained at the weld location due to in-place loads, from which damage is to be calculated. Installation Loading Dynamic loads due to hammer impact during pile installation will induce fatigue damage on both padeye and pile girth welds. The evaluation of the cyclic loads involves the dynamic response of the pile–soil system due to the hammer impact. This requires a wave equation analysis per blow for a given hammer type and efficiency, pile penetration, and soil resistance. Various such analyses are to be conducted for judiciously selected pile penetrations. For each analysis, traces of stress versus time at the critical locations along the pile are to be developed, as well as the number of blows associated with the assumed penetration. For either welds or padeye, fatigue load calculations should be carried out at various pile locations using local stress range, derived from the wave equation analysis at the selected pile penetrations. The location of the girth weld should be determined by the pile makeup schedule. The local response should include the corresponding SCF effect. The number of cycles of the stress history per blow is obtained using a variable amplitude counting method, such as the reservoir methods. Fatigue Resistance Applicable SN curves (stress-number of cycles to failure) depend on the manufacturing processes and defect acceptance criteria. Typically, pile sections are welded by a two-sided SAW (Submerged ArcWelding) process and are left in the as-welded condition. For this case, the D curve may be used. Use of a higher SN curve for this application, without additional treatment of the weld, should be demonstrated by relevant data. Use of weld treatment methods, such as grinding, may support the upgrading of the SN curve, provided that

(1) the grinding process is properly implemented,

(2) weld inspection methods and defect acceptance criteria are implemented, and

(3) pertinent fatigue data are generated to qualify the weld to a performance level higher than that implied by the D curve. Total Fatigue Damage and Factor of Safety Once the fatigue loading and resistance have been determined, fatigue damage due to in-place and installation loads can be evaluated.

### Test Loading of Driven Pile Anchors

The driven pile installation records should demonstrate that the pile selfweight penetration, pile orientation, driving records, and final penetration are within the ranges established during pile design and pile driving analysis. Under these circumstances, test loading of the anchor to full intact stormload should not be required. However, the mooring and anchor design should define a minimum acceptable level of test loading. This test loading should ensure that the mooring line’s inverse catenary is sufficiently formed to prevent unacceptable mooring line slacking due to additional inverse catenary cut-in during storm conditions. Another function of the test loading is to detect severe damage to the mooring components during installation.

## References

[1] GEMS, Vessel Specification of MV Kommandor Jack, <www.gems-group.com>.

[2] K.F. Anschu¨ tz, Cutaway of Anschu¨ tz Gyrocompass, <http://en.wikipedia.org/wiki/Gyrocompass>.

[3] Navis.gr, Gyrocompass - Steaming Error http://www.navis.gr/navaids/gyro.htm.

[4] D.J. House, Seamanship Techniques: Shipboard and Marine Operations, Butterworth-Heinemann, 2004.

[5] L. Mayer, Y. Li, G. Melvin, 3D Visualization for Pelagic Fisheries Research and Assessment, ICES, Journal of Marine Science vol. 59 (2002).

[6] B.M. Isaev, Measurement Techniques, vol. 18, No 4, Plenum Publishing Co, 2007.

[7] P.H. Milne, Underwater Acoustic Positioning Systems, Gulf Publishing, Houston, 1983.

[8] R.D. Christ, R.L. Wernli, The ROV Manual, Advantages and Disadvantages of Positioning Systems, Butterworth-Heinemann 2007.

[9] Fugro Engineers B.V., Specification of Piezo-Cone Penetrometer, <http://www.fugro-singapore.com.sg>.

[10] M. Faulk, FMC ManTIS (Manifolds & Tie-in Systems), SUT Subsea Awareness Course, Houston, 2008.

[11] American Petroleum Institute, Recommended Practice for Planning, Designing and Constructing Fixed Offshore Platforms - Load and Resistance Factor Design, API RP 2A-LRFD (1993).

[12] American Petroleum Institute, Bulletin on Stability Design of Cylindrical Shells, API Bulletin 2U (2003).

[13] American Petroleum Institute, Design of Flat Plate Structures, API Bulletin 2V (2003).

[14] American Petroleum Institute, Design and Analysis of Station Keeping Systems for Floating Structures, API RP 2SK (2005).

[15] American Petroleum Institute, Recommended Practice for Planning, Designing and Constructing Fixed Offshore Platforms - Working Stress Design - Includes Supplement 2, API 2A WSD, 2000.

[16] J.B. Stevens, J.M.E. Audibert, Re-Examination of P-Y Curve Formulations, OTC 3402, Houston, 1979.