Wellhead fatigue damage estimation using probabilistic generative ai models

Abstract

Different probabilistic generative AI models are used to estimate wellhead fatigue damage. A load probabilistic generative AI model is used to determine the load for the wellhead based on the design of the wellhead and the metocean conditions for the wellhead in conjunction with the riser and facility information. A stress probabilistic generative AI model is used to determine the stress on the wellhead based on the design of the wellhead and the load for the wellhead. The fatigue damage on the wellhead is determined based on the stress on the wellhead and fatigue resistance parameters of the wellhead materials. The fatigue damage on the wellhead is used to make operational decisions for the wellhead.

Description

FIELD

[0001] The present disclosure relates generally to the field of estimating wellhead fatigue damage using probabilistic generative AI models.BACKGROUND

[0002] Operation of an underwater well (e.g., subsea well) may result in wellhead fatigue damage. Accurate estimation of wellhead fatigue damage accumulation is required to make informed decisions about the underwater well.SUMMARY

[0003] This disclosure relates to estimating wellhead fatigue damage using probabilistic generative AI models. Wellhead design information, metocean information, and / or other information may be obtained. The wellhead design information may define the design of a wellhead. The metocean information may characterize metocean conditions for the wellhead. Via a load probabilistic generative AI model, load for the wellhead may be determined based on the design of the wellhead, the metocean conditions for the wellhead, and / or other information. The load probabilistic generative AI model may serve as a surrogate for simulation of fluid-structure interaction dynamics for the wellhead. Via a stress probabilistic generative AI model, stress on the wellhead may be determined based on the design of the wellhead, the load for the wellhead, and / or other information. The stress probabilistic generative AI model may serve as a surrogate for finite element analysis for the wellhead. Fatigue damage for the wellhead may be determined based on the stress on the wellhead and / or other information. One or more operations for the wellhead may be facilitated based on the fatigue damage for the wellhead and / or other information.

[0004] A system for estimating wellhead fatigue damage using probabilistic generative AI models may include one or more electronic storage, one or more processors and / or other components. The electronic storage may store information relating to a wellhead, wellhead design information, information relating to a design of the wellhead, metocean information, information relating to metocean conditions for the wellhead, information relating to a load probabilistic generative AI model, information relating to load for the wellhead, information relating to a stress probabilistic generative AI model, information relating to stress on the wellhead, information relating to fatigue damage for the wellhead, information relating to operations for the wellhead, and / or other information.

[0005] The processor(s) may be configured by machine-readable instructions. Executing the machine-readable instructions may cause the processor(s) to facilitate estimating wellhead fatigue damage using probabilistic generative AI models. The machine-readable instructions may include one or more computer program components. The computer program components may include one or more of a wellhead component, a metocean component, a load component, a stress component, a fatigue damage component, an operation component, and / or other computer program components.

[0006] The wellhead component may be configured to obtain wellhead design information and / or other information. The wellhead design information may define the design of a wellhead.

[0007] The metocean component may be configured to obtain metocean information and / or other information. The metocean information may characterize metocean conditions for the wellhead. In some implementations, the metocean conditions for the wellhead may include current profile and wave characteristics for the wellhead. In some implementations, the wave characteristics for the wellhead may include peak wave period, significant wave height, and wave direction for the wellhead.

[0008] The load component may be configured to determine, via a load probabilistic generative AI model, load for the wellhead based on the design of the wellhead, the metocean conditions for the wellhead, and / or other information. The load probabilistic generative AI model may serve as a surrogate for simulation of fluid-structure interaction dynamics for the wellhead.

[0009] In some implementations, physical constraints of the wellhead may be incorporated into one or more loss functions for training of the load probabilistic generative AI model and the stress probabilistic generative AI model.

[0010] In some implementations, a riser may be configured to be connected to the wellhead. The load for the wellhead determined via the load probabilistic generative AI model may include load along the riser. The load for the wellhead determined via the load probabilistic generative AI model may include load at a connection between the riser and the wellhead.

[0011] The stress component may be configured to determine, via a stress probabilistic generative AI model, stress on the wellhead based on the design of the wellhead, the load for the wellhead, and / or other information. The stress probabilistic generative AI model may serve as a surrogate for finite element analysis for the wellhead. In some implementations, the stress probabilistic generative AI model may output one or more images depicting stress field throughout the wellhead.

[0012] The fatigue damage component may be configured to determine fatigue damage for the wellhead. The fatigue damage for the wellhead may be determined based on the stress on the wellhead and / or other information. In some implementations, determination of the fatigue damage for the wellhead may include determination of fatigue damage rate for the wellhead.

[0013] The operation component may be configured to facilitate one or more operations for the wellhead. The operation(s) for the wellhead may be facilitated based on the fatigue damage for the wellhead and / or other information. In some implementations, the operation(s) for the wellhead may include monitoring the fatigue damage for the wellhead. In some implementations, the operation(s) for the wellhead may include modifying the design of the wellhead.

[0014] These and other objects, features, and characteristics of the system and / or method disclosed herein, as well as the methods of operation and functions of the related elements of structure and the combination of parts and economies of manufacture, will become more apparent upon consideration of the following description and the appended claims with reference to the accompanying drawings, all of which form a part of this specification, wherein like reference numerals designate corresponding parts in the various figures. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the invention. As used in the specification and in the claims, the singular form of “a,”“an,” and “the” include plural referents unless the context clearly dictates otherwise.BRIEF DESCRIPTION OF THE DRAWINGS

[0015] FIG. 1 illustrates an example system for estimating wellhead fatigue damage using probabilistic generative AI models.

[0016] FIG. 2 illustrates an example method for estimating wellhead fatigue damage using probabilistic generative AI models.

[0017] FIG. 3 illustrates an example underwater well.

[0018] FIG. 4 illustrates an example process for estimating wellhead fatigue damage using probabilistic generative AI models.

[0019] FIG. 5 illustrates an example table of load for a wellhead.

[0020] FIG. 6 illustrates an example image depicting stress field throughout a wellhead.DETAILED DESCRIPTION

[0021] The present disclosure relates to estimating wellhead fatigue damage using probabilistic generative AI models. Different probabilistic generative AI models are used to estimate wellhead fatigue damage. A load probabilistic generative AI model is used to determine the load for the wellhead based on the design of the wellhead and the metocean conditions for the wellhead in conjunction with the riser and facility (e.g., rig or vessel) information. A stress probabilistic generative AI model is used to determine the stress on the wellhead based on the design of the wellhead and the load for the wellhead. The fatigue damage on the wellhead is determined based on the stress on the wellhead and fatigue resistance parameters of the wellhead materials. The fatigue damage on the wellhead is used to make operational decisions for the wellhead.

[0022] The methods and systems of the present disclosure may be implemented by a system and / or in a system, such as a system 10 shown in FIG. 1. The system 10 may include one or more of a processor 11, an interface 12 (e.g., bus, wireless interface), an electronic storage 13, an electronic display 14, and / or other components. Wellhead design information, metocean information, and / or other information may be obtained by the processor 11. The wellhead design information may define the design of a wellhead. The metocean information may characterize metocean conditions for the wellhead. Via a load probabilistic generative AI model, load for the wellhead may be determined by the processor 11 based on the design of the wellhead, the metocean conditions for the wellhead, and / or other information. The load probabilistic generative AI model may serve as a surrogate for simulation of fluid-structure interaction dynamics for the wellhead. Via a stress probabilistic generative AI model, stress on the wellhead may be determined by the processor 11 based on the design of the wellhead, the load for the wellhead, and / or other information. The stress probabilistic generative AI model may serve as a surrogate for finite element analysis for the wellhead. Fatigue damage for the wellhead may be determined by the processor 11 based on the stress on the wellhead and / or other information. One or more operations for the wellhead may be facilitated by the processor 11 based on the fatigue damage for the wellhead and / or other information.

[0023] The electronic storage 13 may include one or more electronic storage media configured to electronically store information. The electronic storage 13 may store software algorithms, information determined by the processor 11, information received remotely, and / or other information that enables the system 10 to function properly. For example, the electronic storage 13 may store information relating to a wellhead, wellhead design information, information relating to a design of the wellhead, metocean information, information relating to metocean conditions for the wellhead, information relating to a load probabilistic generative AI model, information relating to load for the wellhead, information relating to a stress probabilistic generative AI model, information relating to stress on the wellhead, information relating to fatigue damage for the wellhead, information relating to operations for the wellhead, and / or other information.

[0024] The electronic display 14 may refer to an electronic device that provides visual presentation of information. The electronic display 14 may include a color display and / or a non-color display. The electronic display 14 may be configured to visually present information. The electronic display 14 may present information using / within one or more graphical user interfaces. For example, the electronic display 14 may present information relating to a wellhead, wellhead design information, information relating to a design of the wellhead, metocean information, information relating to metocean conditions for the wellhead, information relating to a load probabilistic generative AI model, information relating to load for the wellhead, information relating to a stress probabilistic generative AI model, information relating to stress on the wellhead, information relating to fatigue damage for the wellhead, information relating to operations for the wellhead, and / or other information.

[0025] A well may refer to a hole or a tunnel in the ground. A well may be drilled in one or more directions. For example, a well may include a vertical well, a horizontal well, a deviated well, and / or other type of well. A well may be drilled in the ground for exploration and / or recovery of natural resources in the ground. For example, a well may be drilled in the ground to aid in extraction of petrochemical fluid (e.g., oil, gas, petroleum, fossil fuel). Application of the present disclosure to other types of wells and wells drilled for other purposes are contemplated.

[0026] Equipment may be installed at the well to facilitate well operations. For example, a wellhead may be installed at the top of the well. A wellhead may refer to one or more components at the top / surface of the well that provides structural and / or pressure-containing interface for drilling and production equipment. For example, a wellhead may include spools, valves, and / or adapters that provide pressure control of a production well. A wellhead may allow for connection of various equipment to the well for production. For example, for an underwater well (e.g., a subsea well), one end of a riser may be connected to a wellhead of a well and the other end of the riser may be connected to a surface facility, such as a platform, floating production storage, and / or offloading vessels. A riser may include one or more pipes that transport fluid between the well / wellhead and the surface facility. A riser may include one or more flexible components, floatation components, and / or components to facilitate use of the riser in an underwater environment.

[0027] Accurately evaluating wellhead fatigue damage is both challenging and essential to ensure proper design and / or operation of wellheads. For example, high pressure and high temperature conditions of wellhead operation may significantly increase the stress on the wellhead, making it susceptible to fatigue failure. Cyclic loading on the wellhead from environmental effects, such as waves, currents, and rig movement, contribute to wellhead fatigue. For example, movement of water around the riser may place force on the riser to push the riser out of its neutral position above the wellhead. Movement of water around the riser may cause shifting, vibration, and / or other movement of the riser. Vortex-induced vibration (VIV) may be a key factor in wellhead fatigue. Shifting, vibration, and / or other movement of the riser may fatigue / weaken the wellhead connected to the riser. Wellhead fatigue damage may accumulate over a period of time. Lateral loads induced by ocean currents may increase significantly with water depth. High pressures in subsea environments may further exacerbate the fatigue damage on the wellhead. Too much wellhead fatigue damage may result in failure or breakage of the wellhead.

[0028] Wellhead fatigue analysis is a complex multi-physics problem due to the inherent structural and fluid nonlinearities, involving a wide range of spatial and temporal scales. Computational modeling is crucial for predicating and controlling these complex fluid-structure interaction dynamics, governed by coupled partial differential equations and ordinary differential equations exhibiting spatiotemporal nonlinearity. Traditional computational methods based on classical numerical techniques, such as finite element analysis (FEA), face significant challenges in efficiency.

[0029] Finite element analysis may consist of numerous (e.g., millions) of elements, requiring sufficient mesh refinement to accurately capture stresses in critical areas of the structure. Both elastic and elastic-plastic material behaviors may need to be considered during the FEA-based design and / or monitoring of the wellhead. However, with increasing geometric complexity, nonlinear material characteristics, and high mesh resolution, conventional FEA-based simulations may become impractical, particularly when dealing with numerous parameters, in terms of computational costs.

[0030] The present disclosure addresses three technical challenges for assessing wellhead fatigue damage. First, the present disclosure reduces the computational cost of wellhead fatigue analysis considering vortex-induced vibration by finite element analysis. Second, the present disclosure lowers the computation cost of 3D finite element analysis for wellheads, a task that becomes particularly challenging when multiple 3D finite element analysis simulations are required to explore various wellhead design configurations. Third, the present disclosure facilitates downstream tasks, such as design optimization and uncertainty quantification of wellhead systems leveraging the inherent modeling capability in the probability space introduced by the incorporation of probabilistic generative AI model.

[0031] To address these challenges, the present disclosure utilizes different probabilistic generative AI models to replace computationally intensive solver used in finite element and fluid-structure interaction analyses. A load probabilistic generative AI model is used as a surrogate for simulating fluid-structure interaction dynamics for the wellhead and a stress probabilistic generative AI model is used as a surrogate for finite element analysis for the wellhead. The dynamic analysis of riser-current interactions and wellhead is tied to the prediction of stress fields, which correspond to probabilistic data distribution in three-dimensional spaces. This physics-data connection allows the use of probabilistic generative AI models that excel in approximating high-dimensional probabilistic distributions. For instance, the probabilistic generative AI models used as surrogates of simulation of fluid-structure interaction dynamics and finite element analysis include denoising diffusion probabilistic models. The wellhead fatigue physics are incorporated into the architecture of denoising diffusion probabilistic models. Physical constraints of the wellhead and connected components (e.g., the riser) are incorporated into the loss functions. The partial differential equations and ordinary differential equations for the wellhead and fluid-structure interaction dynamics are integrated into the loss functions during the forward training processes of the denoising diffusion probabilistic models. In practice, a limited number of sensors are installed for monitoring purposes. The availabilities of these sensor data are incorporated into the loss functions to ensure that the predictions from the probabilistic generative AI models are calibrated based on the sensor data., or in some cases can be used as boundary or initial stations to facilitate the model training process.

[0032] The present disclosure transforms a stress field simulation into an image generation process. Rather than attempting to simulate stress on the wellhead due to load on the wellhead, the present disclosure enables the utilization of a probabilistic generative AI model to generate images representing the stress field throughout the wellhead. The probabilistic generative AI model may generate slices of stress field through the wellhead and / or a 3D stress field through the wellhead.

[0033] FIG. 3 illustrates an example well 302. The well 302 may be located under the water. For example, the well 302 may be located under the ocean / sea. A wellhead 304 may be installed on the well 302, and a riser 306 may be connected to the wellhead 304. The riser 306 may provide a connection through which fluid may flow between the well 302 / wellhead 304 and a facility 308 at / above the water surface. The well 302 and / or the riser 306 may include other components not shown in FIG. 3 (e.g., blowout preventer, lower marine riser package, flexible joint, slick joint, buoyancy joint, water current sensor). For example, water current sensors may be placed along the riser 306 or a structure attached to facility 308 to measure the speed and direction of water around the riser 306.

[0034] The speed of water movement along the riser may not be uniform. For example, FIG. 3 shows an example current profile 310 for the riser 306. The current profile 310 may show the maximum direction and speed of water movement projected onto a two-dimensional plane. The current profile 310 may represent the speed and direction of water along the water column along the riser 306, from the water surface to the seabed. The current profile 310 may characterize the speed of water movement across the water column along the riser 306. The current profile 310 may characterize the speed of water movement (current speed) as a function of water depth (e.g., depth below the water surface). As shown in the current profile 310, the speed of water movement may change with changing depth.

[0035] Wave may move across the water surface. Physical features of the wave may be defined by peak wave period (Tp), significant wave height (Hs), and wave direction. The peak wave period (Tp) may refer to the period associated with the most energetic waves in the wave spectrum in a specific time period. The significant wave height (Hs) may refer to the average wave height, from crest to trough, of the highest one-third of the waves in a specific time period. Wave direction may refer to the direction in which the wave is moving. Movement of water may cause the riser 306 to move (vibrate, oscillate, displace, deform). Movement of the riser 306 may refer to the movement of the riser stack. The wellhead 304 may experience fatigue damage due to the movement of water (in the form of current and wave) which then causes the wellhead 304 movement. Wave above the wellhead 304 may cause movement of the riser 306, which may lead to fatigue damage of the wellhead 304. Movement of water may cause vortex-induced vibration (VIV) of the riser 306. The wellhead 304 may experience fatigue damage due to the VIV of the riser 306.

[0036] FIG. 4 illustrates an example process 400 for estimating wellhead fatigue damage using probabilistic generative AI models. The process 400 may be used to convert historical metocean conditions for a wellhead into historical fatigue damage experienced by the wellhead from past metocean conditions. The process 400 may be used to convert present metocean conditions for a wellhead into fatigue damage being experienced by the wellhead. The process 400 may be used to convert forecasted metocean conditions for a wellhead into fatigue damage to be experienced by the wellhead.

[0037] The process 400 may be used to evaluate and monitor the metocean conditions on fatigue damage for an existing wellhead. For example, the process 400 may be used to monitor fatigue damage of an existing wellhead. The process 400 may be used to evaluate the metocean conditions on fatigue damage for a wellhead to be produced, built, and / or installed. For example, the process 400 may be used to predict fatigue damage on different wellhead designs for a wellhead impacted by different metocean conditions.

[0038] Inputs for estimating fatigue damage for a wellhead may include wellhead design 402 and metocean conditions 404. Inputs for estimating fatigue damage for a wellhead may further include information on the riser connected to the wellhead (e.g., the riser 306) and / or the facility above the wellhead (e.g., the facility 308). The wellhead design 402 may include information on the design of the wellhead. The wellhead design may include information on the well on which the wellhead is to be installed and / or information on soil around the wellhead / well. The wellhead design 402 may include information on geometry / shape, size, structure, configuration, components, materials, and / or specification of the wellhead and / or associated components. The wellhead design 402 may include design of an existing wellhead. The wellhead design 402 may include design of a wellhead to be produced, built, and / or installed. The metocean conditions 404 may include current profile and wave characteristics (peak wave period, significant wave height, wave direction) for the wellhead. The metocean conditions 404 may correspond to (cover) a time duration. The metocean conditions 404 may include historical, present, and / or predicted metocean conditions for a wellhead. The metocean conditions 404 may include metocean conditions during the wellhead's service / operation of the wellhead.

[0039] The wellhead design 402 and the metocean conditions 404 may be input into a load probabilistic generative AI model 406. The load probabilistic generative AI model 406 may serve as a surrogate for simulation of fluid-structure interaction dynamics for the wellhead. The load probabilistic generative AI model 406 may have been trained to output load 408 for a wellhead based on the design of the wellhead and the metocean conditions for the wellhead. The load probabilistic generative AI model 406 may have been trained using real field data and / or synthetic / simulated data. The load probabilistic generative AI model 406 may have been trained using input-output pairs, where the input includes the design of a wellhead and the metocean conditions for the wellhead and the output includes the load for the wellhead. The load 408 output by the load probabilistic generative AI model 406 may include the load for the wellhead for the time duration corresponding to the metocean conditions 404. The load 408 output by the load probabilistic generative AI model 406 may include the load along the riser connected to the wellhead (e.g., actual riser connected to an existing wellhead, a riser to be connected to a wellhead), the load at the connection between the riser and the wellhead, and / or other load for the wellhead.

[0040] The wellhead design 402 and the load 408 for the wellhead may be input into a stress probabilistic generative AI model 410. The stress probabilistic generative AI model 410 may serve as a surrogate for finite element analysis for the wellhead. The stress probabilistic generative AI model 410 may have been trained to output stress 412 on a wellhead based on the design of the wellhead and the load for the wellhead. The stress probabilistic generative AI model 410 may have been trained using real field data and / or synthetic / simulated data. The stress probabilistic generative AI model 410 may have been trained using input-output pairs, where the input includes the design of a wellhead and the load for the wellhead and the output includes the stress on the wellhead. The stress 412 output by the stress probabilistic generative AI model 410 may include the stress on the wellhead for the time duration corresponding to the metocean conditions 404 / the load 408. The stress 412 output by the stress probabilistic generative AI model 410 may include the stress field throughout the wellhead. The stress 412 may be output by the stress probabilistic generative AI model 410 in the form of one or more images depicting the stress field throughout the wellhead. The stress 412 may be output by the stress probabilistic generative AI model 410 in the form of images depicting slices of stress field through the wellhead and / or a 3D image depicting 3D stress field through the wellhead.

[0041] Wellhead fatigue damage 414 may be determined (e.g., computed, calculated, estimated) based on the stress 412. The wellhead fatigue damage 414 may include the fatigue damage on the wellhead for the time duration corresponding to the metocean conditions 404 / the load 408 / the stress 412. The wellhead fatigue damage 414 may include fatigue damage rate for the wellhead. The fatigue damage rate for the wellhead may be used to determine wellhead fatigue damage accumulation. The wellhead fatigue damage accumulation may refer to an amount of fatigue damage accumulated at / experienced by the wellhead. The wellhead fatigue damage accumulation may refer to an amount of fatigue damage accumulated at / experienced by the wellhead for the entire time since the wellhead has been installed at the well (total accumulated wellhead fatigue damage). The wellhead fatigue damage accumulation may refer to an amount of fatigue damage accumulated at / experienced by the wellhead for a particular duration of time / particular operation (operational accumulated wellhead fatigue damage). The fatigue damage rate for the wellhead for a duration may be multiplied by the duration to determine how much wellhead fatigue damage is accumulated during the duration. Total wellhead fatigue damage accumulation may be tracked to prevent failure of the wellhead during operation.

[0042] Allowable fatigue damage limit for a wellhead may refer to how much additional fatigue damage may be accumulated at / experienced by the wellhead before failure or breakage of the wellhead is expected. The allowable fatigue damage limit may indicate the remaining “life” of the wellhead for the life cycle of the wellhead (total allowable fatigue damage limit). The allowable fatigue damage limit may indicate the remaining “life” of the wellhead for a particular duration of time / particular operation (operational allowable fatigue damage limit).

[0043] One or more operations 416 for the wellhead may be facilitated based on the wellhead fatigue damage 414. One or more operational decisions for the wellhead may be made based on the wellhead fatigue damage 414. For example, a design of a wellhead may have been tested using the process 400. The load, the stress, and the wellhead fatigue damage for the wellhead design may be predicted (e.g., under a certain or various metocean conditions) using the process 400. The wellhead fatigue damage for the wellhead design may indicate weak / problematic parts of the wellhead design. Based on the wellhead fatigue damage for the wellhead design, the wellhead design may be modified. The process 400 may be used to test numerous wellhead designs quickly and at low cost.

[0044] As another example, the fatigue damage for the wellhead may be monitored (in real time, in near real time). The fatigue damage for the wellhead may be monitored to change operations at the wellhead. For example, if the total wellhead fatigue damage accumulation rises above a threshold level, operations at the wellhead may be stopped (e.g., disconnect riser from the wellhead). If the wellhead fatigue damage accumulation for a particular operation rises above a threshold level, the particular operation may be stopped. If predicted wellhead fatigue damage accumulation for a future operation is less than a threshold level, the future operation may be performed. If predicted wellhead fatigue damage accumulation for a future operation is greater than a threshold level (fatigue allowance for the future operation), the future operation may not be performed unless appropriate mitigations are in place.

[0045] The estimation of wellhead fatigue damage via the probabilistic generative AI models may assist well operators to make informed decisions from wellhead fatigue perspective as to whether one or more operations should be suspended or performed. The process 400 may be performed / repeated with real-time monitoring of the metocean conditions for a wellhead to make informed decisions for well operations. The process 400 may be performed / repeated with real-time monitoring of the metocean conditions for a wellhead to increase (e.g., optimize) use of the well and avoid unnecessary non-use of the well (e.g., avoid disconnection from the well due to inaccurate forecast of wellhead fatigue damage rate)

[0046] FIG. 5 illustrates an example table 500 of load for a wellhead. The table 500 and / or the values in the table 500 may be output by a load probabilistic generative AI model. The table 500 may include values of load histograms at top of the wellhead. The table 500 may include values of effective tension, shear, and bending moment. The table 500 may provide values for different bins, and may provide the number of cycles of load for a duration of time (e.g., year). Other forms of load output by the load probabilistic generative AI model are contemplated.

[0047] FIG. 6 illustrates an example image 600 depicting stress field throughout a wellhead. The image 600 may include a stress contour plot. The image 600 may be output by a stress probabilistic generative AI model. The image 600 may depict a wellhead and values of stress field throughout the wellhead. The values of the stress field may be indicated by pixel values of the image 600. The image 600 may depict other information relating to the stress field, such as direction of stress and / or type of stress.

[0048] Referring back to FIG. 1, the processor 11 may be configured to provide information processing capabilities in the system 10. As such, the processor 11 may comprise one or more of a digital processor, an analog processor, a digital circuit designed to process information, a central processing unit, a graphics processing unit, a microcontroller, an analog circuit designed to process information, a state machine, and / or other mechanisms for electronically processing information. The processor 11 may be configured to execute one or more machine-readable instructions 100 to facilitate estimating wellhead fatigue damage using probabilistic generative AI models. The machine-readable instructions 100 may include one or more computer program components. The machine-readable instructions 100 may include one or more of a wellhead component 102, a metocean component 104, a load component 106, a stress component 108, a fatigue damage component 110, an operation component 112, and / or other computer program components.

[0049] The wellhead component 102 may be configured to obtain wellhead design information and / or other information. Obtaining wellhead design information may include one or more of accessing, acquiring, analyzing, determining, examining, generating, identifying, loading, locating, opening, receiving, retrieving, reviewing, selecting, storing, and / or otherwise obtaining the wellhead design information. The wellhead component 102 may obtain wellhead design information from one or more locations. For example, the wellhead component 102 may obtain wellhead design information from a storage location, such as the electronic storage 13, electronic storage of a device accessible via a network, and / or other locations. The wellhead component 102 may obtain wellhead design information from one or more hardware components (e.g., a computing device) and / or one or more software components (e.g., software running on a computing device). In some implementations, the wellhead design information may be obtained from one or more users. For example, a user may interact with a computing device to input the wellhead design information (e.g., upload the wellhead design information, specify design of a wellhead).

[0050] The wellhead design information may define the design of a wellhead. The design of a wellhead may refer to a plan and / or drawing of how the wellhead is built and / or operated. The design of a wellhead may include information on geometry / shape, size, structure, configuration, components, materials, and / or specification of the wellhead and / or associated components. The design of a wellhead may be in two dimensions, three dimensions, and / or other dimensions.

[0051] The wellhead design information may define a design of a wellhead by characterizing, describing, identifying, quantifying, reflecting, setting forth, and / or otherwise defining the design of the wellhead. The wellhead design information may define a design of a wellhead by including information that defines one or more content, qualities, attributes, features, and / or other aspects of the design of the wellhead. For example, the wellhead design information may define the design of a wellhead by including documentation on the structural configuration of the wellhead. Other types of wellhead design information are contemplated.

[0052] The design of a wellhead may include information on geometry / shape, size, structure, configuration, components, materials, and / or specification of equipment to be connected to the wellhead (e.g., a well, a riser, a subsea stack, blowout preventer, tree, tubing head spool, connectors). For example, information on components / equipment directly and / or indirectly connected to the wellhead may be obtained. Information on the environment of the wellhead may be obtained. Such information may be part of the wellhead design information or separate from the wellhead design information.

[0053] For example, riser information, equipment information, well information, soil information, facility information and / or other information may be obtained with or separately from the wellhead design information. The riser information may define the design of a riser for the wellhead (e.g., riser to which the wellhead is connected / to be connected). The equipment information may define the design of one or more pieces of equipment between the riser and the wellhead. The well information may define the design of a well for the wellhead. The soil information may define the characteristics of the soil around the wellhead / the well for the wellhead (e.g., information on types of soil and / or soil characteristics around the wellhead / the well). The facility information may define the design of the facility (e.g., rig, vessel) for the wellhead.

[0054] The metocean component 104 may be configured to obtain metocean information and / or other information. Obtaining metocean information may include one or more of accessing, acquiring, analyzing, determining, examining, generating, identifying, loading, locating, measuring, opening, receiving, retrieving, reviewing, selecting, storing, and / or otherwise obtaining the metocean information. The metocean component 104 may obtain metocean information from one or more locations. For example, the metocean component 104 may obtain metocean information from a storage location, such as the electronic storage 13, electronic storage of a device accessible via a network, and / or other locations. The metocean component 104 may obtain metocean information from one or more hardware components (e.g., a computing device, sensors) and / or one or more software components (e.g., software running on a computing device). In some implementations, the metocean information may be obtained from one or more users. For example, a user may interact with a computing device to input the metocean information (e.g., upload the metocean information, specify the metocean conditions for a well / wellhead).

[0055] The metocean information may characterize metocean conditions for the wellhead. The metocean information may characterize metocean conditions for the wellhead during the wellhead's service / operation of the wellhead. The metocean information may characterize metocean conditions for a wellhead by defining, describing, identifying, quantifying, reflecting, setting forth, and / or otherwise characterizing the metocean conditions for the wellhead. The metocean information may include information relating to metocean conditions within the whole water depth including at or near the wellhead. The metocean information may include information relating to water depth, waves, currents, tide and surge variations, and / or other environmental conditions around / surrounding / above the wellhead. In some implementations, the metocean information may include time series information. The metocean information may characterize metocean conditions at different moments in time. The metocean information may characterize changes in metocean conditions through time.

[0056] The metocean information may characterize metocean conditions for a wellhead by including information that defines one or more content, qualities, attributes, features, and / or other aspects of the metocean conditions for the wellhead. For example, the metocean information may characterize metocean conditions for a wellhead by including information that specifies types and values of metocean conditions for the wellhead. Other types of metocean information are contemplated.

[0057] Metocean conditions for a wellhead may refer to wind, wave, climate, and / or other environmental conditions that affect wellhead and / or water around / surrounding / above the wellhead. The metocean conditions for the wellhead may include current profile and wave characteristics for the wellhead. The wave characteristics for the wellhead may include peak wave period, and significant wave height, and / or wave direction for the wellhead. Other wave characteristics are contemplated.

[0058] The metocean conditions for the wellhead may correspond to a time. The time corresponding to the metocean conditions for the wellhead may include a historical time, a present time, or a future time. Historical metocean conditions may be used to determine historical wellhead fatigued damage / historical wellhead fatigued damage rate. Present metocean conditions may be used to determine present wellhead fatigued damage / present wellhead fatigued damage rate. Future metocean conditions may be used to determine future wellhead fatigued damage / future wellhead fatigued damage rate. The time corresponding to the metocean conditions for the wellhead may include the time of the wellhead's service / time during which the wellhead will be in operation.

[0059] The load component 106 may be configured to determine, via a load probabilistic generative AI model, load for the wellhead. Determining load for a wellhead may include ascertaining, approximating, calculating, establishing, estimating, finding, identifying, obtaining, quantifying, selecting, setting, and / or otherwise determining the load for the wellhead. The load for the wellhead may refer to force on the wellhead. The load for the wellhead may include force applied to the wellhead. The load for the wellhead may include the load along the riser transferred to the wellhead through the equipment between the riser and the wellhead. The load for the wellhead may include the load at one or more connections between the riser and the wellhead.

[0060] The load for the wellhead may be determined based on the design of the wellhead, the metocean conditions for the wellhead, and / or other information. The load for the wellhead may be determined by providing the design of the wellhead, the metocean conditions for the wellhead, and / or other information to the load probabilistic generative AI model. Additional information about installation / operation of the wellhead, such as information on the rise connected to the wellhead and / or the facility above the wellhead (e.g., a rig, a vessel) may be provided to the load probabilistic generative AI model. For example, the load probabilistic generative AI model may be additionally provided with riser information, equipment information, well information, soil information, facility information, and / or other information.

[0061] The load probabilistic generative AI model may utilize the design of the wellhead, the metocean conditions for the wellhead, and / or other information as input. The load probabilistic generative AI model may output the load for the wellhead. The load output by the load probabilistic generative AI model may include the load for the wellhead for the time duration corresponding to the metocean conditions. For example, FIG. 5 illustrates the example table 500 of load for a wellhead. Other forms of load output by the load probabilistic generative AI model are contemplated.

[0062] The load probabilistic generative AI model may refer to a probabilistic generative AI model that has been trained to output load for a wellhead. The load probabilistic generative AI model may serve as a surrogate for simulation of fluid-structure interaction dynamics for the wellhead. A probabilistic generative AI model may refer to a machine learning model that creates new data by learning patterns in existing / training data. A probabilistic generative AI model may refer to a machine learning model that learns the underlying probability distribution of data to generate new, similar data. The load probabilistic generative AI model may have been trained to output the load for the wellhead based on the design of the wellhead and the metocean conditions for the wellhead. The load probabilistic generative AI model may have been trained using real field data and / or synthetic / simulated data. The load probabilistic generative AI model may have been trained using input-output pairs, where the input includes the design of a wellhead and the metocean conditions (measured, calculated, simulated) for the wellhead and the output includes the load (e.g., measured, calculated, simulated) for the wellhead.

[0063] The stress component 108 may be configured to determine, via a stress probabilistic generative AI model, stress on the wellhead. Determining stress on the wellhead may include ascertaining, approximating, calculating, establishing, estimating, finding, identifying, obtaining, quantifying, selecting, setting, and / or otherwise determining the stress on the wellhead. The stress on the wellhead may refer to the force acting on a unit area of the wellhead. The stress on the wellhead may include the intensity of force distributed across / through the wellhead. The stress on the wellhead may include stress field throughout the wellhead.

[0064] The stress on the wellhead may be determined based on the design of the wellhead, the load for the wellhead, and / or other information. The stress on the wellhead may be determined by providing the design of the wellhead, the load for the wellhead, and / or other information to the stress probabilistic generative AI model. The stress probabilistic generative AI model may utilize the design of the wellhead, the load for the wellhead, and / or other information as input. The stress probabilistic generative AI model may output the load for the wellhead. The stress output by the stress probabilistic generative AI model may include the stress on the wellhead for the time duration corresponding to the metocean conditions. In some implementations, the stress probabilistic generative AI model may output one or more images depicting stress field throughout the wellhead. For example, FIG. 6 illustrates the example image 600 depicting stress field throughout a wellhead. Other graphical / visual forms of stress output by the stress probabilistic generative AI model are contemplated.

[0065] The stress probabilistic generative AI model may refer to a probabilistic generative AI model that has been trained to output stress on a wellhead. The stress probabilistic generative AI model may serve as a surrogate for finite element analysis for the wellhead. The stress probabilistic generative AI model may have been trained to output stress on the wellhead based on the design of the wellhead and the load for the wellhead. The stress probabilistic generative AI model may have been trained using real field data and / or synthetic / simulated data. The stress probabilistic generative AI model may have been trained using input-output pairs, where the input includes the design of a wellhead and the load for the wellhead (measured, calculated, simulated) and the output includes the stress (measured, calculated, simulated) on the wellhead.

[0066] In some implementations, physical constraints of the wellhead (and associated components) may be incorporated into one or more loss functions for training of the probabilistic generative AI models (the load probabilistic generative AI model, the stress probabilistic generative AI model). Example physical constraints of the wellhead may include deformation criteria, stress and material relationships (e.g., the membrane or average stress across a wellhead high pressure housing is typically below its yield strength by design). In some implementations, physical constraints may be categorized into two main types: data-driven discrepancies and physical laws. Data-driven discrepancies may capture the prediction error against the ground truth sensor data, while physical laws may be represented as the residuals of differential equations. Weights for these two factors may be integrated and adaptively controlled during the training process. During the early stages of training, more weight may be assigned to the residuals of the physical equations. This approach helps the learning space converge towards the desired physical space more quickly. As training progresses, the weight may be gradually shifted towards data-driven discrepancies to further refine the learned dynamics. The loss function may include an entropy regularization term. Th entropy regularization term may ensure that the learned parameter space does not converge to a specific point but instead adheres to a high-dimensional manifold that better reflects the underlying physical laws.

[0067] Physical constraints of the wellhead may be used as parameters in the loss function(s) for training of the probabilistic generative AI model. The loss function(s) for the probabilistic generative AI models may employ a hybrid formulation that integrates both data-driven and physics-informed components. The data-driven losses may capture discrepancies between predicted and measured (or ground-truth) data. Meanwhile, the physics-informed components may include residuals computed from the partial differential equations governing stress fields, boundary conditions, and initial conditions. Incorporating physics-informed loss components may ensure that the probabilistic generative AI models learn patterns that fully comply with underlying physical laws.

[0068] The fatigue damage component 110 may be configured to determine fatigue damage for the wellhead. Determining fatigue damage for a wellhead may include ascertaining, approximating, calculating, establishing, estimating, finding, identifying, obtaining, quantifying, selecting, setting, and / or otherwise determining the fatigue damage for the wellhead. The fatigue damage for the wellhead determined by the fatigue damage component may include a rate at which the wellhead is experiencing fatigue damage, the fatigue damage experienced by the wellhead over a duration of time (time duration corresponding to the metocean conditions), and / or the total fatigue damage experienced by the wellhead (since installation, since start of wellhead service / operation).

[0069] The fatigue damage for the wellhead may be determined based on the stress on the wellhead and / or other information. The fatigue damage for the wellhead may be determined further based on information relating to the material of the wellhead, such as fatigue resistance parameters of the wellhead material. Fatigue resistance parameters of the wellhead material may refer to parameters that quantify the wellhead material's ability to withstand repeated cycles of stress or strain without filing. In some implementations, determination of the fatigue damage for the wellhead may include determination of fatigue damage rate for the wellhead. In some implementations, the fatigue damage on the wellhead / fatigue damage rate may be determined as a percentage value and / or other values. For example, the fatigue damage may be determined as a percentage of damage over a time period. For example, an undamaged wellhead may have started with 100% allowable damage throughout its entire life cycle. The fatigue may be determined as how much of damage will be accumulated at the wellhead for a particular duration of time (e.g., 3% for 48-hour period).

[0070] The operation component 112 may be configured to facilitate one or more operations for the wellhead. An operation for a wellhead may include an operation performed using the wellhead, an operation directed at the wellhead, and / or other operation relating to the wellhead. An operation for a wellhead may include a well operation / an operation relating to a well. Facilitating an operation for a wellhead may include making the operation easier. Facilitating an operation for a wellhead may include enabling / assisting in preparation, planning, and / or performance of the operation for the wellhead. Facilitating an operation for a wellhead may include controlling the operation for the wellhead. Facilitating an operation for a wellhead may include starting, stopping, changing, preventing, and / or otherwise controlling the operation for the wellhead. Other facilitations of operations for the wellhead are contemplated.

[0071] For example, an operation for an existing wellhead may include using the wellhead for production from the well or disconnecting the riser from the wellhead. For instance, whether a production operation for a well with the wellhead will proceed or not may be determined based on the fatigue damage for the wellhead (e.g., proceeding based on the fatigued damage for the wellhead being below a threshold value, not proceeding based on the fatigued damage for the wellhead exceeding a threshold value). As another example, an operation for a future wellhead may include designing the wellhead, producing the wellhead, and / or installing the wellhead. For instance, the design of the wellhead may be modified based on the expected fatigue damage for the wellhead exceeding a threshold value.

[0072] The operation(s) for the wellhead may be facilitated based on the fatigue damage for the wellhead and / or other information. The fatigue damage for the wellhead may be used to make operational decisions for the wellhead / the well. The fatigue damage for the wellhead may be used to plan and / or perform operations for the wellhead. The fatigue damage for the wellhead may be used to determine which operations will be performed, whether operations will be performed, and / or how the operations will be performed.

[0073] Facilitation of an operation based on the fatigue damage for the wellhead may include presentation of information relating to the fatigue damage for the wellhead on one or more displays, monitoring of the operation based on information relating to the fatigue damage for the wellhead, planning of the operation based on information relating to the fatigue damage for the wellhead, automation of the operation based on information relating to the fatigue damage for the wellhead, and / or other facilitation of the operation. Information relating to the fatigue damage may include the fatigue damage, information derived from the fatigue damage, and / or information from which the fatigue damage is derived. For example, information relating to the fatigue damage may include the fatigue damage rate, the fatigue damage accumulation at the wellhead, an allowable fatigue damage limit, and / or other information.

[0074] In some implementations, the operation(s) for the wellhead may include monitoring the fatigue damage for the wellhead. The fatigue damage for the wellhead may be tracked to determine how much wellhead damage has accumulated at the wellhead (since installation, for an operation). The remaining allowable fatigue damage limit for the wellhead (for the entire life of the wellhead, for an operation) may be tracked using the fatigue damage for the wellhead. The remaining allowable fatigue damage limit for the wellhead may be monitored (e.g., regularly calculated, checked, presented)

[0075] In some implementations, the operation(s) for the wellhead may include modifying the design of the wellhead. The fatigue damage for the wellhead may be calculated for different wellhead designs and / or different metocean conditions. For example, different designs of a wellhead may be tested for one or more metocean conditions expected at a well location. Rather than using complex and compute-intensive calculations to simulate complex fluid-structure interaction dynamics, the present disclosure may be used to easily and quickly compute fatigue damage for different wellhead designs. The fatigue damage for different wellhead designs may be used to modify the design of the wellhead (to reduce predicted fatigue damage at the well location) and / or to select which wellhead design should be installed at the well.

[0076] Implementations of the disclosure may be made in hardware, firmware, software, or any suitable combination thereof. Aspects of the disclosure may be implemented as instructions stored on a machine-readable medium, which may be read and executed by one or more processors. A machine-readable medium may include any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computing device). For example, a tangible computer-readable storage medium may include read-only memory, random access memory, magnetic disk storage media, optical storage media, flash memory devices, and others, and a machine-readable transmission media may include forms of propagated signals, such as carrier waves, infrared signals, digital signals, and others. Firmware, software, routines, or instructions may be described herein in terms of specific exemplary aspects and implementations of the disclosure, and performing certain actions.

[0077] As used herein, the phrase “configured to” is intended to be interpreted broadly, as “being capable of or suitable for performing” some function or feature, without requiring any adaptations to provide said function or feature.

[0078] In some implementations, some or all of the functionalities attributed herein to the system 10 may be provided by external resources not included in the system 10. External resources may include hosts / sources of information, computing, and / or processing and / or other providers of information, computing, and / or processing outside of the system 10.

[0079] Although the processor 11, the electronic storage 13, and the electronic display 14 are shown to be connected to the interface 12 in FIG. 1, any communication medium may be used to facilitate interaction between any components of the system 10. One or more components of the system 10 may communicate with each other through hard-wired communication, wireless communication, or both. For example, one or more components of the system 10 may communicate with each other through a network. For example, the processor 11 may wirelessly communicate with the electronic storage 13. By way of non-limiting example, wireless communication may include one or more of radio communication, Bluetooth communication, Wi-Fi communication, cellular communication, infrared communication, or other wireless communication. Other types of communications are contemplated by the present disclosure.

[0080] Although the processor 11, the electronic storage 13, and the electronic display 14 are shown in FIG. 1 as single entities, this is for illustrative purposes only. One or more of the components of the system 10 may be contained within a single device or across multiple devices. For instance, the processor 11 may comprise a plurality of processing units. These processing units may be physically located within the same device, or the processor 11 may represent processing functionality of a plurality of devices operating in coordination. The processor 11 may be separate from and / or be part of one or more components of the system 10. The processor 11 may be configured to execute one or more components by software; hardware; firmware; some combination of software, hardware, and / or firmware; and / or other mechanisms for configuring processing capabilities on the processor 11.

[0081] It should be appreciated that although computer program components are illustrated in FIG. 1 as being co-located within a single processing unit, one or more of computer program components may be located remotely from the other computer program components. While computer program components are described as performing or being configured to perform operations, computer program components may comprise instructions which may program processor 11 and / or system 10 to perform the operation.

[0082] While computer program components are described herein as being implemented via processor 11 through machine-readable instructions 100, this is merely for ease of reference and is not meant to be limiting. In some implementations, one or more functions of computer program components described herein may be implemented via hardware (e.g., dedicated chip, field-programmable gate array) rather than software. One or more functions of computer program components described herein may be software-implemented, hardware-implemented, or software and hardware-implemented.

[0083] The description of the functionality provided by the different computer program components described herein is for illustrative purposes, and is not intended to be limiting, as any of computer program components may provide more or less functionality than is described. For example, one or more of computer program components may be eliminated, and some or all of its functionality may be provided by other computer program components. As another example, processor 11 may be configured to execute one or more additional computer program components that may perform some or all of the functionality attributed to one or more of computer program components described herein.

[0084] The electronic storage media of the electronic storage 13 may include non-transitory computer readable media. The electronic storage media of the electronic storage 13 may be provided integrally (i.e., substantially non-removable) with one or more components of the system 10 and / or as removable storage that is connectable to one or more components of the system 10 via, for example, a port (e.g., a USB port, a Firewire port, etc.) or a drive (e.g., a disk drive, etc.). The electronic storage 13 may include one or more of optically readable storage media (e.g., optical disks, etc.), magnetically readable storage media (e.g., magnetic tape, magnetic hard drive, floppy drive, etc.), electrical charge-based storage media (e.g., EPROM, EEPROM, RAM, etc.), solid-state storage media (e.g., flash drive, etc.), and / or other electronically readable storage media. The electronic storage 13 may be a separate component within the system 10, or the electronic storage 13 may be provided integrally with one or more other components of the system 10 (e.g., the processor 11). Although the electronic storage 13 is shown in FIG. 1 as a single entity, this is for illustrative purposes only. In some implementations, the electronic storage 13 may comprise a plurality of storage units. These storage units may be physically located within the same device, or the electronic storage 13 may represent storage functionality of a plurality of devices operating in coordination.

[0085] FIG. 2 illustrates method 200 for estimating wellhead fatigue damage using probabilistic generative AI models. The operations of method 200 presented below are intended to be illustrative. In some implementations, method 200 may be accomplished with one or more additional operations not described, and / or without one or more of the operations discussed. In some implementations, two or more of the operations may occur substantially simultaneously.

[0086] In some implementations, method 200 may be implemented in one or more processing devices (e.g., a digital processor, an analog processor, a digital circuit designed to process information, a central processing unit, a graphics processing unit, a microcontroller, an analog circuit designed to process information, a state machine, and / or other mechanisms for electronically processing information). The one or more processing devices may include one or more devices executing some or all of the operations of method 200 in response to instructions stored electronically on one or more electronic storage media. The one or more processing devices may include one or more devices configured through hardware, firmware, and / or software to be specifically designed for execution of one or more of the operations of method 200.

[0087] Referring to FIG. 2 and method 200, at operation 202, wellhead design information may be obtained. The wellhead design information may define a design of a wellhead. In some implementations, operation 202 may be performed by a processor component the same as or similar to the wellhead component 102 (Shown in FIG. 1 and described herein).

[0088] At operation 204, metocean information may be obtained. The metocean information may characterize metocean conditions for the wellhead. In some implementations, operation 204 may be performed by a processor component the same as or similar to the metocean component 104 (Shown in FIG. 1 and described herein).

[0089] At operation 206, via a load probabilistic generative AI model, load for the wellhead may be determined based on the design of the wellhead and the metocean conditions for the wellhead. The load probabilistic generative AI model may serve as a surrogate for simulation of fluid-structure interaction dynamics for the wellhead. In some implementations, operation 206 may be performed by a processor component the same as or similar to the load component 106 (Shown in FIG. 1 and described herein).

[0090] At operation 208, via a stress probabilistic generative AI model, stress on the wellhead may be determined based on the design of the wellhead and the load for the wellhead. The stress probabilistic generative AI model may serve as a surrogate for finite element analysis for the wellhead. In some implementations, operation 208 may be performed by a processor component the same as or similar to the stress component 108 (Shown in FIG. 1 and described herein).

[0091] At operation 210, fatigue damage for the wellhead may be determined based on the stress on the wellhead. In some implementations, operation 210 may be performed by a processor component the same as or similar to the fatigue damage component 110 (Shown in FIG. 1 and described herein).

[0092] At operation 212, one or more operations for the wellhead may be facilitated based on the fatigue damage for the wellhead. In some implementations, operation 212 may be performed by a processor component the same as or similar to the operation component 112 (Shown in FIG. 1 and described herein).

[0093] Although the system(s) and / or method(s) of this disclosure have been described in detail for the purpose of illustration based on what is currently considered to be the most practical and preferred implementations, it is to be understood that such detail is solely for that purpose and that the disclosure is not limited to the disclosed implementations, but, on the contrary, is intended to cover modifications and equivalent arrangements that are within the spirit and scope of the appended claims. For example, it is to be understood that the present disclosure contemplates that, to the extent possible, one or more features of any implementation can be combined with one or more features of any other implementation.

Claims

1. A system for estimating wellhead fatigue damage using probabilistic generative AI models, the system comprising:one or more physical processors configured by machine-readable instructions to:obtain wellhead design information, the wellhead design information defining a design of a wellhead;obtain metocean information, the metocean information characterizing metocean conditions for the wellhead;determine, via a load probabilistic generative AI model, load for the wellhead based on the design of the wellhead and the metocean conditions for the wellhead, wherein the load probabilistic generative AI model serves as a surrogate for simulation of fluid-structure interaction dynamics for the wellhead;determine, via a stress probabilistic generative AI model, stress on the wellhead based on the design of the wellhead and the load for the wellhead, wherein the stress probabilistic generative AI model serves as a surrogate for finite element analysis for the wellhead;determine fatigue damage for the wellhead based on the stress on the wellhead; andfacilitate one or more operations for the wellhead based on the fatigue damage for the wellhead.

2. The system of claim 1, wherein the metocean conditions for the wellhead include current profile and wave characteristics for the wellhead.

3. The system of claim 2, wherein the wave characteristics for the wellhead include peak wave period, significant wave height, and wave direction for the wellhead.

4. The system of claim 1, wherein the one or more operations for the wellhead include monitoring the fatigue damage for the wellhead.

5. The system of claim 1, wherein the one or more operations for the wellhead include modifying the design of the wellhead.

6. The system of claim 1, wherein determination of the fatigue damage for the wellhead includes determination of fatigue damage rate for the wellhead.

7. The system of claim 1, wherein the stress probabilistic generative AI model outputs one or more images depicting stress field throughout the wellhead.

8. The system of claim 1, wherein:a riser is configured to be connected to the wellhead; andthe load for the wellhead determined via the load probabilistic generative AI model includes load along the riser.

9. The system of claim 1, wherein:a riser is configured to be connected to the wellhead; andthe load for the wellhead determined via the load probabilistic generative AI model includes load at a connection between the riser and the wellhead.

10. The system of claim 1, wherein physical constraints of the wellhead are incorporated into one or more loss functions for training of the load probabilistic generative AI model and the stress probabilistic generative AI model.

11. A method for estimating wellhead fatigue damage using probabilistic generative AI models, the method comprising:obtaining wellhead design information, the wellhead design information defining a design of a wellhead;obtaining metocean information, the metocean information characterizing metocean conditions for the wellhead;determining, via a load probabilistic generative AI model, load for the wellhead based on the design of the wellhead and the metocean conditions for the wellhead, wherein the load probabilistic generative AI model serves as a surrogate for simulation of fluid-structure interaction dynamics for the wellhead;determining, via a stress probabilistic generative AI model, stress on the wellhead based on the design of the wellhead and the load for the wellhead, wherein the stress probabilistic generative AI model serves as a surrogate for finite element analysis for the wellhead;determining fatigue damage for the wellhead based on the stress on the wellhead; andfacilitating one or more operations for the wellhead based on the fatigue damage for the wellhead.

12. The method of claim 11, wherein the metocean conditions for the wellhead include current profile and wave characteristics for the wellhead.

13. The method of claim 12, wherein the wave characteristics for the wellhead include peak wave period, significant wave height, and wave direction for the wellhead.

14. The method of claim 11, wherein the one or more operations for the wellhead include monitoring the fatigue damage for the wellhead.

15. The method of claim 11, wherein the one or more operations for the wellhead include modifying the design of the wellhead.

16. The method of claim 11, wherein determining the fatigue damage for the wellhead includes determining fatigue damage rate for the wellhead.

17. The method of claim 11, wherein the stress probabilistic generative AI model outputs one or more images depicting stress field throughout the wellhead.

18. The method of claim 11, wherein:a riser is configured to be connected to the wellhead; andthe load for the wellhead determined via the load probabilistic generative AI model includes load along the riser.

19. The method of claim 11, wherein:a riser is configured to be connected to the wellhead; andthe load for the wellhead determined via the load probabilistic generative AI model includes load at a connection between the riser and the wellhead.

20. The method of claim 11, wherein physical constraints of the wellhead are incorporated into one or more loss functions for training of the load probabilistic generative AI model and the stress probabilistic generative AI model.

Images