Field of the invention

The invention relates to a method and an associated system of estimating loads on a subsea component, such as a wellhead system, a subsea tree, an emergency disconnect package and or a lower riser package, based on measurements performed in at least two positions/sections in the lower part of the riser. The invention is specifically applicable for offshore applications.

Background of the invention

Small bored and thick walled open sea completion and workover (C/WO) riser systems are currently extensively used to provide well access for performing C/WO operations on subsea oil and gas wells. During operations, the riser connects the surface vessel, or any other floating arrangement, to the wellhead system (WH) and subsea tree (XT) at the sea bottom. At the lower end of the riser there is a lower riser package (LRP) and an emergency disconnect package (EDP) that allow the operator to seal off the well and disconnect in the unlikely event of an emergency. The weight of the LRP and EDP are substantial.

During operations, waves may cause significant rig motion. As the rig moves, dynamic loads are applied to the riser, and loads may be transferred to the WH. The riser is also exposed to loads both from waves and sea current. Even though open sea risers are flexible and can be exposed to large elastic deformations, loads applied on the EDP, LRP, XT and WH can be high. In the event of an accidental drive-off or drift-off of the rig, the equipment may be damaged.

An important part of the preparation for a C/WO operation is the evaluation of structural loading of the riser and WH system. Operating limitations must be established in order to prevent overloading of the WH, XT and riser system. At the same time it must be verified that dynamic loads do not cause excessive permanent damage to the equipment, particularly the part placed at the bottom of the sea.

According to ISO 13628-7 a structural analysis shall be performed in order to verify the adequacy of the riser design. The structural analysis is based on a number of assumptions regarding particularly the environmental conditions. In order to ensure safe operations, assumptions are generally conservative in nature, and as a consequence, operating windows tend to be quite narrow. Furthermore, estimates of fatigue damage may also be very high, and as a result riser system components or XTs may have to be replaced more frequently than would have been necessary if more accurate load estimates were available. Wells may also have to be abandoned much earlier if conservative assumptions are adopted. Prior art solutions include riser monitoring systems (RMS) that provide more accurate information on the loads applied to the riser. One such system measures loads on the lower tapered stress joint (LTSJ) as well as on the tension joint above and below the rucker tension wire (RTW) ring. Loads are then estimated for the rest of the riser by using a numerical finite element model simulating the behavior of the physical system. Another riser monitoring system measures the displacement and inclination at a number of positions along the riser. For this system a numerical model is used to provide load estimates. A third system includes strain gauge sensors mounted on a special spool piece positioned in the riser stack and a method for estimating loads in the riser using measurement data and a finite element model. While existing riser monitoring systems can be expected to be quite successful in estimating the loads on the riser, their capability to estimate loads on the WH, XT, LRP and EDP is limited. Measurement positions are arranged quite far from these components, and extrapolation of results is difficult using finite element models only. Furthermore, the use of a complex numerical model for estimating loads introduces uncertainty and may be the cause of errors. Numerical calculations may also be very time-consuming and difficult to use for real time operations.

For new wells and completions it may be very time-consuming, or sometimes even impossible, to integrate load sensors in the existing design. Furthermore, for the large number of wells with subsea completions from the last 30 years, integration of sensor systems prove to be very difficult. It is of great importance that loads on these old wells are monitored during operations because they have already been exposed to significant fatigue. Additionally, their resistance to fatigue may be low. The method for measuring loads on the lower stack and WH constitute an important part of a riser monitoring system. For some old WH systems the weakest link may be the WH connector or XT connector. In some cases the well itself may also have been weakened through its history. Hence, measured and estimated bending loads shall be compared to capacities of sections and connectors in an on-line system. Furthermore, as loads are cyclic, the measured or estimated loads must also be carefully monitored and compared to estimates of fatigue life.

It is an objective of the present invention to establish a method and system for providing estimates of the loads on the WH, XT, EDP, LRP and possible other components of the subsea system.

Furthermore, it is also an objective of the present invention to use the estimates obtained by this method and system to assist in positioning a surface vessel in an optimal way in order to minimize loads on the WH and thereby optimize the fatigue life of the WH.

Summary of the invention A riser is extending from a surface vessel, or any other floating arrangement, to the wellhead (WH). A lower tapered stress joint (LTSJ) may be arranged in the lower part of the riser. The LTSJ may optionally be connected to the WH via an EDP, LRP and XT. The LTSJ is provided with at least two measuring devices in two positions/sections A and B measuring section forces (tension/bending moment) and/ or inclination/ displacement. The EDP/ LRP is provided with an inclination/ acceleration measurement device in position/section C. The terms 'position' and 'section' shall be understood as a location where measurements are made, i.e. a cross-section of the riser/subsea component of interest. The relationship between measurements performed by the measuring devices in positions A and B at the LTSJ and in position E on the EDP/LRP, and the forces acting on any given point on the subsea component (EDP, LRP, XT, WH) can be established using a model that includes both the effect of section forces in positions A or B and, if necessary, EDP/ LRP inertia and gravity forces. The calculations may in a first embodiment be made by assuming that the subsea component is in equilibrium. Alternatively, in a second embodiment, in case inertia forces cannot be disregarded, a more complex dynamic solution is found.

It shall be understood that the term "measuring device" shall be understood in a broad sense throughout the disclosure, and that other terms may be used in the same meaning, including strain gauges or similar sensors for measuring section forces or inclination, inclinometer, accelerometer, velocity measurements, displacement measurement or temperature sensors. Similarly, the term "section force" may include bending and torsion moments, tension and shear forces.

The strain gauges or similar measuring devices may be located in at least at two positions A and B at the lower part of the riser, preferably at the lower tapered stress joint that forms part of the riser. The strain gauges or similar sensors are constructed to produce estimates of tension, bending moments and or inclination at the respective positions. The strain measurements can be used to obtain estimates of tension forces, shear forces and bending moments at respective locations. In a typical configuration strain sensors are used in three positions, e.g. position A, B and C, and inclinometers and/ or accelerometers is used in two other positions, e.g. positions D and E. In a typical configuration for measuring section forces, four strain gauges are arranged in a spaced-apart relationship around the outer circumference of the LTSJ with a 90 degrees spacing. Bending moments along two perpendicular axes can be obtained in addition to tension load. In the case of the tension load, replicate estimates will be obtained for the said sensor configuration. Generally, the number of independent measurements shall be greater than the number of sought responses in order to allow for redundancy in sensor design. By positioning strain gauges at an inclination relative to the main axis of the riser, information about additional load components can be obtained. Internal pressure and torsion loading are assessed by placing sensors fully or partly in the

circumferential direction.

Thermal effects must also be considered, especially if the riser contains a fluid at a higher temperature than the surroundings. This may involve the additional use of dedicated temperature sensors such as thermocouples positioned at the location of strain measurement or at one or several locations directly in conjunction with the measurement or with the fluid contained inside the riser. The fluid temperature may also be monitored using existing temperature sensors in the control system. The effect of temperature on strain/tension measurement may be simulated and removed using thermo-mechanic models evaluating thermal expansion and measured results.

Inclinometers can be used to measure the inclination of the LTSJ or EDP/LRP at any point on the stack. Some inclinometers are also capable of measuring velocities and accelerations at the very same points. Translational and rotational velocities and accelerations may be considered. Velocities and accelerations are of importance when considering the dynamics of the subsea components. Dynamic loading may have to be considered when assessing loads if the subsea component has a large mass and accelerates at a sufficiently high rate. The inertia forces are evaluated by calculating the mass times the acceleration of the component.

Gravitational loads must be considered when the axis of the lower stack is inclined relative to the vertical as gravitation caused shear forces and bending moments in sections of the riser.

First embodiment: Assuming subsea stack in equilibrium

By assuming that inertia forces can be neglected, equilibrium equations can be used to calculate forces in the stack from section A-B in the lower part of the riser, e.g. a section in the LTSJ, downwards through the equipment connected to the LTSJ, such as EDP, LRP, XT, and to the WH. Independent estimates may be obtained from one or more sensor/measuring devices arrangements, including a sufficient number of independent sensor measurements. One possible configuration/arrangement involves two set of sensors for measuring longitudinal strains at locations A and B. Estimates of bending moments for sections A and B may be used to estimate corresponding shear forces. It has been found that accurate estimates of the shear forces can be obtained by disregarding dynamic and bending effects entirely and assessing only equilibrium for the relatively short section A-B. In this case the shear force at position A, VA, can be expressed as: equat ion (1) LAB

Here MA is the bending moment in measuring position A, MB is the bending moment in measuring position B. LAB is the effective distance between measuring positions A and B . When assessing L _AB , consideration must be given to the actual point of application of forces on the section for which equilibrium is assessed. In most cases L _AB would be the actual distance between e.g. strain gauge sensors A and B. It should be pointed out that in the fully three-dimensional case the shear force V _A , as well as bending moments M _A and M _B , can be expressed as vectors since bending may occur relative to two perpendicular bending axes. However, the system for measuring strain in a cross-section distinguishes the two components and produces two independent components of bending moment and shear force. If measuring devices A and B are placed too far apart, errors are introduced in the assessment of the shear force V _A - These errors are due to the curvature of the riser as well as to dynamic mass forces introduced as a result of sensor motion. Hence, the distance L _AB should be minimized. However, if the distance is made too small errors in bending force estimates M _A and M _B may be significant compared to the difference in bending moment M _A -M _B . The optimal distance L _AB can only be determined if the accuracy of measuring devices are known as well as the load range. The distance L _AB can normally be between 0.2 - 2 meters depending on the dimensions of the risers and external load.

In order to allow for more accurate estimates of shear forces over an extended range of bending moments, more sensors/ measuring devices and measurement points along the LTSJ may be introduced. If the bending moments applied to the LTSJ are high, errors in M _A and M _B are small compared to the difference M _A -M _B . The distance L _AB can be made small. However, if load levels are generally low, the distance L _AB should be made large in order to make the difference M _A -M _B large. In order to achieve this, a measuring position C is introduced to allow for accurate measurements when the load levels are low. For low loading levels, sensors A and C are activated, and for high load levels, sensors A and B are activated. The measurement system will determine which set of sensors/measuring devices to use by evaluating load levels and errors in the measurements. It is of course possible to introduce more sensors/ measuring devices in order to be able to perform

measurements for a larger range of bending moments.

Estimates of shear forces can also be obtained from sensors/ measuring devices positioned at locations A, B or C by positioning a sufficient number of strain gauges to allow unique determination of all load components such as tension, bending, torsion, shear and internal pressure. For each position there are six independent section force components to consider in addition to the internal pressure, i.e.

altogether 18 variables for the three cross sections plus pressure. However, the tension can be regarded to differ only slightly for all sections, and changes in pressure can be estimated by considering the mass of the section between sensor/ measuring positions. Torsion loads can also be regarded as equal for all sections. As a result, there are only 15 independent variables which can be assessed by performing a minimum of 15 independent measurements. Strain sensor sets in positions A and B may be replaced by sensor sets in positions D and E. If so, the sensor sets in positions D and E should include inclinometers and tension sensors measuring the deformation of the LTSJ. Measurement position E acts as a reference for bending of the LTSJ. In the special case where it is assumed that the stress joint has a constant elastic stiffness I along a length (z), that the rotation of the stack, ΘΕ, is small and that mass forces can be neglected for the component, it is possible to develop an exact expression for the deflection along the length: equation (2)

At the same time the bending moment M(z) can be expressed as: equation (3)

Here, E is Young's modulus, I is the stiffness of the section, θο is the inclination of the riser at position D relative to the LTSJ axis at the lowermost section or the EDP. In a fully three-dimensional case where any inclination is best described by vectors or Euler angles, the directionality of the LTSJ inclination must also be considered, λ is a characteristic length which can be expressed as:

equation (4) T is the riser tension force which may either be measured by a set of strain gauges (A or B) or deduced from readings from the rig riser tensioner system and mass calculations. Since it is known that the shear force in any section is the derivative of the bending moment, the shear force V(z) can also be found to be: equation (5) A ² A ^J V(z) =— exp (- -] Estimates of bending moments for any section of the LTSJ can also be obtained in the case that the LTSJ has a non-uniform or tapered cross section. In this case, the load distribution can be calculated numerically.

The estimates of shear forces at position A, or any other lower position on the LTSJ, may be used in combination with measurement of tension forces and bending moments to assess tension forces, shear forces and bending moments at different sections in the stack and wellhead. It has been demonstrated that in most cases inertia forces can be disregarded, and that estimates of forces at any given cross- section of the stack can be obtained by considering only the equilibrium for the part between the cross section and measurement point A on the LTSJ.

Errors introduced by the described method disregarding inertia and external loads tend to be less than 2-3 %. It should also be noted that external loads on the subsea stack due to current and wave motion are usually small compared to the loads applied by the riser. However, by using both load and inclination measurements it is possible to estimate external loading (current and wave loading). With data from more than two independent sensors it is possible to introduce the external load on the LTSJ as an additional unknown and to determine this unknown using

measurement data. In this case the system becomes a tool also for assessing hydrodynamic loads.

Assuming equilibrium for the stack and disregarding geometric changes due to loading, the shear forces and bending moments applied on the wellhead, Vw and M _w , can be estimated using the following formulas: equation (6) Vw = ¾

equation (7) M _w = M _A + ¾ - L _AW

Here LAW is the length of the section extending from the lowermost measurement point on the LTSJ to the wellhead top (WH top). Likewise, applied in any other cross-section or position, the shear forces, V _z , and bending moments, M _z , at any position Z can be calculated using the formulas: equation (8) ¾ = YA

equation (9) A*z = M _A + V _A - L _AZ

Tension forces at cross-sections W or Z (T _w or T _z ) can be obtained assessing tension in measurement point A and subtracting the net down weight of the intermediate section. For this simplified case tension does not affect bending moments at the wellhead.

A more complete and accurate calculation of loads on the wellhead can be obtained by considering the deformation of the LTSJ and stack. A finite element model for the stack from the wellhead up to the section A on the LTSJ can be established. The model can be solved using actual section forces in cross-section A. The results will be the shape of the deformed stack and LTSJ and the section forces in any cross- section of the stack including the WH. As an alternative to a numerical model, a closed form solution can be developed by considering force equilibrium with tension forces in section A acting in a direction deviating from the vertical due to the deformation of the stack and, if applicable, wellhead inclination. For a purely two-dimensional evaluation section forces in the wellhead can be estimated assuming equilibrium for the lower part of the riser and the lower stack as follows: equation (10) T _w = T _A costf) - V _A sin ) equation (1 1) equation (12)

M _w = M _A — (T _A £OS( ) — ¾stn(^)) ^■ L _A + (τ _ά Είτι(β}— ¾cos(^)) ^■ (L _} + L _s ^' cos( ^' )

+ G _s ^■ L _G sin(a)

Here, Tw, Vw and Mw are the effective tension, shear force and bending moment at the wellhead. T _A , V _A and M _A are the effective tension, shear force and bending moment at the tension at cross-section A. a is the average inclination of the lower stack, and β is the inclination of the riser at section A, both with respect to the vertical. Gs is the gravitational force of the stack acting in the vertical direction. L _A is the displacement of cross-section A from the vertical line going through the wellhead. L _G is the distance from the WH datum to the center of gravity of the entire lower stack along the axis of the lower stack. Ls is the distance from the wellhead to the top of the lower stack along the axis of the lower stack. Lj is the distance from the top of the lower stack to cross-section A along the axis of the lower stack. Distances L _A and Lj must be estimated by assessing the deformation of the LTSJ exposed to a set of section forces in A, for instance using equation (2).

Verification or more accurate estimates of forces can be obtained by assessing the measured data from the inclinometer positioned on the stack, position E. The inclinometer will reveal whether the stack and wellhead are tilted and whether a net bending moment is applied as a result of the weight of the stack acting in a direction non-parallel to the wellhead axis. Furthermore, the inclinometer also reveals whether dynamic effects should be considered, and gives input to the on-line calculation of these effects.

Mass forces acting on the stack can be calculated by multiplying the mass and acceleration of the stack. By directly measuring the acceleration of the stack the loads can be established. In the case that the movement of the wellhead is one of rotation around a reference point in the well, the moment of inertia must be considered (this is described in the second embodiment below). In most cases movement of the stack is limited, and inertia forces are of lesser importance. This also applies for hydrodynamic loading.

Second embodiment: Assuming subsea stack in no n- equilibrium As an alternative to the assumption that the subsea stack is in equilibrium, inertia of the subsea stack may be considered, in which case inertia terms must be introduced in equations 6 and 7 (or alternatively equations 8, 9 and 10).

In this case of equation 7 the governing equation controlling the bending moment at a position W can be modified: equation (13) M _w = M _A + V _A ^■ L _AW + (l _w m) where the last term expresses the inertia forces due to a rotation about the point W. Here, Iw is the moment of inertia about the point W and ω is the angular velocity around the same point. The time derivative is calculated. Angular velocities and accelerations can be measured and input into the equation making assessment of resulting forces and bending moments simpler.

The equation is deduced exclusively for the two-dimensional plane. Establishment of fully three-dimensional equations can be performed in a similar manner as demonstrated for the two-dimensional case. In most cases inertia forces of the subsea stack may be disregarded allowing only the first embodiment to be implemented.

For both embodiments discussed above, as an alternative to either only section force measurements or only inclination measurements, both inclination and section force measurements may be combined to give estimates of shear force continuously along the LTSJ with greater accuracy than by the simplified equilibrium evaluation.

Calculations then take into consideration the actual shape of the bent section of the LTSJ. Given that the accuracy of force and inclination measurements are of similar relative accuracy, improved estimates of forces acting on the system will be obtained by calibrating the system by means of both sets of independent

measurements. Increased redundancy will also be obtained. Mass forces for the

LTSJ may also be taken into account as the accelerations of the riser are measured directly, but the effects of such forces on results are small.

The invention is set forth and characterized in the independent claims, while the dependent claims describe other characteristics of the invention. The invention describes a system and method of measuring loads and deformation on the lower part of the riser, e.g. at the lower tapered stress joint (LTSJ), and, based on these measurements, calculate an estimate for the loads at any position on a subsea component. The subsea component can be any component typically in place subsea, such as WH, EDP, LRP and or XT. It is a need that the content of the riser is known. If the riser is filled with a fluid, the density and temperature of the fluid must be known. This is necessary in order to properly estimate the tension in the system. If the riser contains a wireline or drill string, mass balance calculations are also performed in order to make sure that the level of tension is correct. The stiffness of the riser joint may also have to be re-evaluated.

Knowledge about the forces applied on the wellhead and the dynamics of the well makes it possible to assess the stiffness of the well. The stiffness of the well does not need be known in order to estimate the forces applied on the wellhead from measurement of section forces and inclinations on the LTSJ. However, the stiffness of the well is an important parameter in the global riser model which can be used to check results and to estimate forces at cross-sections further up in the riser. Hence, estimation of the WH stiffness is important for the purposes of verification. The measurement set-up can be used for estimating wellhead stiffness. The wellhead stiffness depends on the quality of the cementing, the soil properties and interactions with the template structure. It is very difficult to measure the stiffness of the well directly in other ways due to limited access.

Independent verification of the method for estimating loads on the stack and WH is performed by introducing for example strain gauges at sections of the stack. If new XTs are installed these can be instrumented. Sensors or measuring devices are then preferably permanent. If special adapters are used to connect LRP and XT, such adapters may constitute beneficial positions for positioning of sensors. There may be several possible positions for mounting a sensor package for the purposes of verification. It should be pointed out that such a measurement is not a requirement for the invention. The method is known to be of high accuracy, and there are other methods also for validating results.

The invention relates to a method for estimating section forces including bending moments on a subsea component, the subsea component being connected to a riser and a well, wherein the method comprises the steps of;

measuring the section forces and/or deformations in a lower section (A) and a first upper section (B) at the lower part of the riser by the use of at least two independent measuring devices,

measuring the inclination and acceleration of the subsea component using a set of sensors placed in a location of the subsea component (section E), calculating the section forces for any section (Z) of the subsea component by the use of a processing device adapted for receiving data from said measuring devices and performing calculations, the calculations involving:

o an assessment of dynamic equilibrium for the subsea component and the part of the riser below the lower section (A) including load terms

o considering the magnitude and directionality of section forces in the section (A), and o gravitational and inertia forces for the subsea component based on said measurements.

In an aspect, the distance between the lower (A) and the upper (B) sections is between 0.2 and 2.0 meters.

In an aspect, the method further comprises the step of measuring the inclination in at least one position (A, B, C, E) in the lower part of the riser. In an aspect, the method is performed using the assumptions that inertia forces can be neglected.

In an aspect the method is performed using the assumptions that gravitational forces can be neglected.

In an aspect the method is performed using the assumptions that both gravitational and inertia forces can be neglected.

In an aspect of the method, the estimation of shear force components in a lower section A of the lower riser by assuming equilibrium for the part of the riser between the first upper section B and the lower section A, said shear force components set equal to the difference between the bending moment components for said sections A, B divided by the distance between the sections. In an aspect of the method, the estimation of shear force components in a lower section A of the lower riser s performed by assuming equilibrium for the part of the riser between a second upper section C, a first upper section B and the lower section A, said shear force components set equal to either the difference between the bending moment components for said sections A and C divided by the distance between the sections or the difference between the bending moment components for said sections A and B divided by the distance between the sections.

In an aspect, the distance between the closest sensor sections may be within 0.5 m and 2.0 meters.

In an aspect, the method comprises the calculation of the force components acting in any section Z of the subsea component by assuming that the components of shear force acting in a lower section A of the riser, and that the bending moment components for any section Z of the subsea component equals the measured bending moment component at a section A of the riser plus the shear force at section A times a distance between the said section A of the riser and the said section Z of the subsea component.

In an aspect, a riser deformation angle is measured in two sections A, B, C or D and the shape and load distribution of the riser is calculated assuming that the lower part of the riser deforms elastically. The riser may be exposed to a known tension, which tension may be known because it is either measured or estimated.

In an aspect of the system, the subsea component may comprise a wellhead, a subsea tree, an emergency disconnect package and or a lower riser package. The first measuring device may in an aspect be a strain gauge and the second measuring device may be a strain gauge.

The first measuring device may in one aspect be a strain gauge and the second measuring device may be an inclinometer.

In an aspect, the first measuring device and the second measuring device are adapted to measure strain, bending moment and or shear force.

The strain gauges or similar sensors for measuring section forces can either be attached permanently to the riser joint or mounted in a non-permanent way. In latter case sensors may be attached to the riser prior to operations or they may be attached during operations, for example by a remote operated vehicle (ROV) operating subsea.

The estimates of shear forces can also be obtained from sensors positioned at sections A, B or C by positioning a sufficient number of strain gauges to allow unique determination of all load components (tension, bending, torsion, shear and internal pressure). There are seven independent section force components including bending moments in total to be measured and at least seven independent

measurements are needed to evaluate the shear force.

In an aspect, the inclinometers may be attached to the riser and a part of the subsea package (EDP/LRP) or to the well completion (XT/WH). Inclinometers may be placed at the same sections as strain sensors or between strain sensors. Furthermore, several inclinometers may also be attached to the lower part of the riser. This will increase the cost and complexity of the system, but will give both greater accuracy and redundancy. In addition it will allow a more exact measurement of external forces such as drag from the sea current. Sensors may be mounted both in a permanent or a non-permanent manner. In the latter case, the systems can be mounted prior to operation or during operations.

In an aspect, the inclinometers may measure the movement/rotation with the corresponding velocities along three axes. By comparing the inclination of the riser at one or several positions with the inclination of the lower stack, it is possible to obtain an independent estimate of the bending loads on the LTSJ and stack. In an aspect, the inclinometers may be placed in a section E at the subsea

component or at the lowermost part of the riser and at a position D at the top of the LTSJ of the lower riser. This allows for calculating estimates of the dynamic movement of the subsea component and compensation for dynamic effects through the inclusion of an inertia term. In combination with loads, this allows for evaluation of the lateral stiffness of the well. Brief description of the drawings

These and other characteristics of the invention will be clear from the following description of a preferential form of embodiment, given as a non-restrictive example, with reference to the attached drawings wherein;

Figure 1 shows schematically a typical configuration of a lower stack and wellhead. Figure 2 shows possible positions of sensors such as inclinometers and strain gauge sensors, on the LTSJ and wellhead stack.

Figure 3 shows possible positions of inclinometers and strain gauge sensors on the lower tapered stress joint and stack.

Figure 4 shows the LTSJ and stack and a typical deformation of the LTSJ. Figure 5 shows typical shear stress distributions and bending moment distribution in the lower tapered stress joint and stack.

Figure 6 shows the load equilibrium of a section of the LTSJ between bending moment sensor distributions.

Figure 7 shows the deformation of the LTSJ and the system for measuring the angle. Figure 8 shows the equilibrium of the entire stack (including XT) and the main parameters included in the estimation of loads Tw, Vw and Mw on the wellhead.

Figure 9 shows the equilibrium of the stack above the XT and the main parameters included in the estimation of the loads on the XT re-entry hub.

Figure 10 shows a system for evaluating utilization of the lower stack taking into consideration rig position and motion, waves, wind and current.

Figure 1 1 shows the positioning of the rig as a part of a square pattern test to determine the system response and to establish an operating window.

Figure 12 shows a system for assessing and using input data from a multitude of sensors to estimate loads on the stack and to evaluate errors in estimates. Detailed description of a preferential embodiment

Figure 1 shows schematically a typical configuration of a lower stack and wellhead. A riser, or riser string, 1 extends from a surface vessel (not shown) to a subsea wellhead 6, the wellhead 6 forming the entry to a well in the subsea formation 7. In the shown embodiment, the riser 1 is connected to the wellhead 6 via a wellhead stack, comprising an emergency disconnect package (EDP) 3, a lower riser package (LRP) 4, a subsea tree adapter (XT adapter) and a subsea tree (XT) 5. A Lower Tapered Stress Joint (LTSJ) 2 forms the lower part of the riser string 1 and connects the riser to the EDP 3. The reference sign 'T' in the figure refers to the riser tension applied on the LTSJ 2.

Figure 2 shows possible positions of inclinometers and strain gauge sensors on the lower tapered stress joint and stack. In the figure two inclinometers are arranged at positions D and E, while strain gauges are arranged at positions A, B and V. Three sensors (A, B, D) have been placed on the LTSJ 2. Positions E and V are at the stack. Position D need not necessarily be above positions A and B. Inclinometers may be placed at the same positions as the strain sensors or between the strain sensors. Position V also functions as a validation point for the measurements in positions A and B. L _AB refers to the length L between the two neighboring sensors A and B.

Figure 3 shows an embodiment of the system of figure 2 where there are arranged four sensors on the LTSJ 2 at positions A, B, C and D. L _AB refers to the length L between the two neighboring sensors A and B, while L _B c refers to the length L between positions B and C .

Figure 4 shows the system of figure 3, but in this figure the riser string 1 , LTSJ 2 and stack are deformed, shown by the bent riser string and riser tension T. An additional strain sensor is arranged at position W at the stack (on the wellhead).

Figure 5 shows typical shear stress distributions V and bending moment distribution M in the LTSJ 2 and stack at different locations along the stack, LTSJ 2 and riser string 1.

Figure 6 shows the load equilibrium of a section of the LTSJ between bending moment sensor distributions A and B. The riser is assumed to be un-deformed. T _A is the tension force in position A, V _A is here a corresponding shear force component and M _A is a corresponding bending moment component. Similar forces act in section B. W _AB is the weight of section of the pipe between positions A and B. L _AB is the distance between positions A and B.

Figure 7 shows the deformation of the LTSJ and the system for measuring the angle. Two inclinometers D and E measure the inclination of the system at positions D, θο, and E, Θ _Ε , respectively. T _A refers to the tension at position A, V _A refers to a shear force component at position A and M _A refers to a bending moment component at position A. Similarly, M _B and T _B refers to a bending moment component and the tension at position B, respectively. W _AB is the weight of the LTSJ section between positions A and B.

Figure 8 shows the equilibrium of the entire stack (including XT) and the main parameters included in the estimation of section force components Tw, Vw and Mw on the wellhead. M _DLTSJ is the mass of the lower part of the LTSJ. The weights of the EDP, LRP, adapter and XT are given as W _EDP , W _LRP , W _ADPT , W _XT , respectively.

Figure 9 shows an embodiment of Figure 8, but in this embodiment is not including the loads on the XT. Figure 9 shows the equilibrium of the stack above the XT and the main parameters included in the estimation of the loads on the XT re-entry hub. Figure 10 shows a system for evaluating utilization of the lower stack (shown above wellhead 6) taking into consideration the position of the rig 1 1 , motion, waves and wind (collective term 12), and current. The riser 1 is extending from the rig 1 1 to the wellhead 6 and is shown being influenced by wave, current, wind motions etc. Data (via lines 13) may be transferred directly on-shore (not shown)(10). Figure 1 1 shows the positioning of the rig as a part of a calibration pattern test to determine the system response and to establish an operating window. In this case it is assumed that the rig is anchored by anchors (dotted lines 15). A _z indicates an acceptable operating area, P _A z indicates a possibly acceptable operating area, while N _AZ indicates a not acceptable operating area of the position of the rig 1 1 relative the position of the wellhead 6.

Figure 12 shows a system for assessing and using input data from a multitude of sensors to estimate loads on the stack and to evaluate errors in estimates. The figure is an example and only considers a simplified system with a LTSJ with constant stiffness. The invention is herein described in non-limiting embodiments. A person skilled in the art will understand that there may be made alterations and modifications to the embodiments that are within the scope of the invention as described in the attached claims. For example, if used in an advisory mode it is important that the output produced by the RMS is quality assured. The proposed invention can, if properly qualified and within certain limits, be regarded as a safety critical system. First, it is designed with a sufficient redundancy. Sensors on the LTSJ assess a number of strain components and do not only allow for the estimation of bending loads, but also the assessment of errors in estimation. The tension in the riser system is not a completely independent variable. Tension is measured by sensors placed higher on the risers (usually by strain gauges close to the tension joint) and by the tensioner system of the rig. The tension at the LTSJ is equal to the tension at a higher point in the system minus the net weight for the part of the riser separating the measurement points. Furthermore, sensors measure tension at two or more closely spaced positions on the LTSJ, and differences in tension measurements can be expected to be very small. If tension measurements can be verified to be correct, a quality check has also been performed for the bending moment measurements since estimates are determined from the same set of sensors. A further verification of bending moment readings can be obtained by comparing the results from several sensors. However, bending moment measurements can also be checked and calibrated by independent means. First, prior to use measurements can be independently checked and calibrated by controlled bending of the LTSJ under a known load. Second, during use a function test and calibration can be performed by moving the rig in a predefined pattern in order to bend the LTSJ with sensors.