Natural gas hydrate exploitation riser mechanical response determination method and system based on multi-factor coupling

Abstract

The invention belongs to the technical field of natural gas exploitation, and relates to a natural gas hydrate exploitation riser mechanical response determination method and system based on multi-factor coupling, and the method comprises the steps: collecting data of a natural gas hydrate exploitation environment; according to the data of the natural gas hydrate exploitation environment, training a pre-established multi-factor coupling riser mechanical model; dividing model parameters in the multi-factor coupling riser mechanical model into static parameters and dynamic parameters, wherein the static parameters are parameters which remain unchanged in the mining process; the dynamic parameters are parameters which change along with time, and correcting the dynamic parameters; according to the static parameters and the corrected dynamic parameters, generating a final multi-factor coupling riser mechanical model, and inputting the to-be-tested data into the final multi-factor coupling riser mechanical model to obtain a riser mechanical response result; and carrying out potential safety hazard early warning according to the mechanical response result of the riser. The method can accurately determine the mechanical response of the riser, track and predict in real time, and effectively warn the potential safety hazard.

Description

Technical Field

[0001] This invention relates to a method and system for determining the mechanical response of riser in natural gas hydrate extraction based on multi-factor coupling, belonging to the field of deep-water natural gas hydrate extraction technology. Background Technology

[0002] The natural gas hydrate production riser is a critical pipeline system connecting the subsea natural gas hydrate reservoir to the surface production platform. Its core function is to transport decomposition products (gas, liquid, and solid multiphase flow) or drilling fluid in a high-pressure, low-temperature deep-water environment. The mechanical response of the natural gas hydrate production riser refers to its dynamic structural behavior under multi-physics coupled loads in a complex deep-sea environment, directly affecting production safety and equipment lifespan. Key influencing factors include dynamic environmental loads, multiphase flow-structure coupling effects, and extreme geological conditions.

[0003] In existing technologies, determining the mechanical response of risers used in natural gas hydrate extraction often employs relatively simplistic analytical methods. These include modal response prediction, establishing coupled seepage-thermal-deformation-phase-change models, and creating centrifuge experimental platforms to simulate the mechanical behavior of risers throughout the entire extraction cycle and assess formation risks. When considering marine environmental factors, the focus is often solely on the effects of ocean currents or waves, neglecting the coupling effects and the influence of other marine environmental parameters (such as seawater temperature and salinity). For seabed geological factors, static formation models are frequently used, overlooking the dynamic changes in formation mechanical properties caused by insufficient hydrate decomposition. Regarding the fluid within the riser, it is often treated as a single-phase flow, ignoring the complex interactions between multiphase flows involving natural gas, water, and undecomposed hydrate particles. Numerical calculations often suffer from insufficiently fine mesh generation and simplified boundary conditions, making it difficult to accurately simulate actual operating conditions. Once the model is built, the parameters are typically fixed and cannot adapt to the dynamic changes in the environment and operating conditions during extraction. Summary of the Invention

[0004] To address the aforementioned problems, the purpose of this invention is to provide a method and system for determining the mechanical response of risers in natural gas hydrate extraction based on multi-factor coupling. Through comprehensive data acquisition, construction of a multi-factor coupled mechanical model, dynamic model updating, and refined numerical calculation and verification, it achieves accurate determination and real-time tracking and prediction of the mechanical response of risers, and can effectively provide early warning of potential safety hazards.

[0005] To achieve the above objectives, the present invention proposes the following technical solution: a method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling, comprising the following steps: collecting data on the natural gas hydrate extraction environment; Based on the data of the natural gas hydrate extraction environment, a pre-established multi-factor coupled riser mechanical model is trained. The model parameters in the multi-factor coupled riser mechanical model are divided into static parameters and dynamic parameters. The static parameters are parameters that remain unchanged during the extraction process, while the dynamic parameters are parameters that change over time. The dynamic parameters are then corrected. Based on the static parameters and the corrected dynamic parameters, a final multi-factor coupled riser mechanical model is generated. The test data is input into the final multi-factor coupled riser mechanical model to obtain the riser mechanical response results. Safety hazard warnings are then issued based on the riser mechanical response results.

[0006] Furthermore, the data collected on the natural gas hydrate extraction environment includes: obtaining marine environmental data of the extraction area through sensors; obtaining seabed geological information through geophysical exploration and geological drilling; and collecting real-time operating parameters during the extraction process through pressure sensors and flow sensors on the natural gas hydrate extraction riser.

[0007] Furthermore, the data on the natural gas hydrate extraction environment is preprocessed: the data on the natural gas hydrate extraction environment is cleaned to remove outliers and noise, the processed data is divided into training set, validation set and test set, and the data is standardized or normalized. The processed data is then input into the data queue in batches.

[0008] Furthermore, based on the data from the natural gas hydrate extraction environment, the method for training the pre-established multi-factor coupled riser mechanics model is as follows: A data batch is extracted from the data queue, input into the multi-factor coupled riser mechanics model, and a forward propagation algorithm is executed to obtain the predicted riser mechanical response. The predicted riser mechanical response is compared with the actual value of the data batch, and a loss value is calculated. If the loss value is less than a threshold, the training of the multi-factor coupled riser mechanics model ends; otherwise, a backpropagation algorithm is performed based on the loss value to calculate the gradient of the model parameters, and the model parameters are updated according to the gradient. It is then checked whether the updated model has generated new monitoring data. If so, the new data is preprocessed and added to the data queue, and training continues in a loop until the model meets the convergence condition.

[0009] Furthermore, in the multi-factor coupled riser mechanical model, the fluid-structure interaction performance of the riser is calculated using the added mass method, and the vibration excitation of the riser caused by waves is calculated using the Morrison equation; the thermal conductivity performance of the riser is calculated using a three-dimensional thermal conductivity model of seawater-riser-fluid inside the riser; the geotechnical mechanical properties of the riser are calculated using a constitutive model based on elastoplastic mechanics; and the multiphase flow performance inside the riser is calculated using the Euler-Lagrange equation.

[0010] Furthermore, the added mass method involves incorporating the drag force and inertial force of the ocean current on the riser into the riser's equation of motion through an added mass term; when calculating the riser vibration excitation caused by waves using the Morrison equation, the Morrison equation is decomposed into linear wave force and nonlinear wave force for separate calculation, and the results are superimposed; when establishing a three-dimensional heat conduction model of seawater-riser-fluid inside the riser, the heat conduction coefficient in the three-dimensional heat conduction model is corrected according to the riser material properties and temperature conditions, and the heat absorbed during the hydrate dispensing process inside the riser is incorporated into the heat conduction equation through a source term; in the constitutive model, the supporting force of the formation on the bottom of the riser is used as a boundary condition, and the long-term effects of formation subsidence and deformation on the riser's mechanical response are also included; in the Euler-Lagrange equation, natural gas, water, and undecomposed hydrate particles are considered as different phases, and the interaction between these phases accurately describes the effect of the multiphase flow characteristics inside the riser on its mechanical response.

[0011] Furthermore, the dynamic parameters are corrected using Kalman filtering or particle filtering algorithms.

[0012] Furthermore, the method for correcting the dynamic parameters is as follows: inputting the data from the previous moment into the multi-factor coupled riser mechanical model to obtain the prior estimate value at the current moment; fusing the prior estimate value with the data at the current moment; calculating the optimal estimate value at the current moment based on the fusion result; updating the corresponding data parameters based on the optimal estimate value; inputting the updated data parameters into the multi-factor coupled riser mechanical model to calculate the mechanical response result of the riser at the next moment.

[0013] Furthermore, the multi-factor coupled riser mechanical model is discretized using the finite element method or finite difference method to solve the mechanical response of the riser under different mining stages and environmental conditions. The riser is divided into finite element meshes of appropriate size, and the key parts of the riser are treated with fine meshing. For the connection between the bottom of the riser and the seabed strata, the support force and constraint conditions calculated by the geotechnical mechanics model are applied.

[0014] This invention also discloses a system for determining the mechanical response of risers in natural gas hydrate extraction based on multi-factor coupling, comprising: a data acquisition module for acquiring data on the natural gas hydrate extraction environment; a model training module for training a pre-established multi-factor coupled riser mechanical model based on the data on the natural gas hydrate extraction environment; a model parameter correction module for dividing the model parameters in the multi-factor coupled riser mechanical model into static parameters and dynamic parameters, wherein the static parameters are parameters that remain unchanged during the extraction process, and the dynamic parameters are parameters that change over time, and correcting the dynamic parameters; a model output module for generating a final multi-factor coupled riser mechanical model based on the static parameters and the corrected dynamic parameters, inputting the test data into the final multi-factor coupled riser mechanical model to obtain the riser mechanical response result; and an early warning module for issuing a safety hazard warning based on the riser mechanical response result.

[0015] The technical solution of the present invention has at least the following technical effects or advantages: 1. This invention provides a rich and accurate data foundation for subsequent analysis through comprehensive data collection, ensuring the understanding of various influencing factors during the mining process. The multi-factor coupled mechanical model constructed by this invention integrates multiple physical field couplings such as fluid-structure interaction, heat conduction, and geotechnical mechanics to accurately characterize the mechanism of action of each factor on the mechanical response of the riser. It can more realistically simulate the actual mining conditions. This multi-factor coupled mechanical model has dynamic update capability, and uses Kalman filtering or particle filtering algorithms to correct dynamic parameters in real time, realizing real-time tracking and prediction of the mechanical response of the riser, greatly improving the timeliness and accuracy of the model. Combined with time series analysis and machine learning to warn of safety hazards, it can effectively ensure the safety of mining operations. This invention can accurately determine the mechanical response of the riser under the coupling of multiple factors, providing solid technical support for the safe and efficient mining of natural gas hydrates.

[0016] 2. In the dynamic mechanics model training cycle algorithm, the initialization stage establishes a reasonable starting framework for the model. Data preprocessing effectively purifies the original data, removes outliers and noise interference, and divides and standardizes the dataset, improving data quality and training efficiency. In the iterative training, the model parameters are continuously adjusted through steps such as forward computation, loss calculation, and backpropagation to update parameters. In particular, the Kalman filter or particle filter algorithm is used to update dynamic parameters, enabling the model to highly fit multi-source data such as marine environment, seabed geology, and mining operations, greatly improving the accuracy of the model in predicting the mechanical response of risers. The finite element method and finite difference method are used to discretize and solve the multi-factor coupled mechanics model. The mesh is densified in key parts to improve the calculation accuracy, and the mesh quality index is strictly controlled to avoid excessive errors.

[0017] 3. This invention accurately sets boundary conditions based on actual working conditions, rationally selects and optimizes solver and algorithm parameters, and can accurately capture the mechanical response details of the riser under different mining stages and environmental conditions, such as stress distribution, strain, displacement changes and vibration characteristics, providing accurate quantitative data support for mining operations.

[0018] 4. This invention acquires real-time data on stress, strain, acceleration, and displacement using sensors installed on the riser. Laboratory simulations construct a model of similar working conditions to measure the mechanical response, validating the numerical calculation results from both actual mining scenarios and simulated environments. Statistical analysis methods such as root mean square error (RMSE) and mean absolute error (MAE) are used to quantitatively assess the degree of agreement between the model's calculation results and actual data, providing a scientific basis for judging the model's accuracy. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the riser system in a natural gas hydrate extraction system according to one embodiment of the present invention; Figure 2 This is a flowchart illustrating the data collection process for the extraction environment of natural gas hydrates in one embodiment of the present invention; Figure 3 This is a schematic diagram of a multi-factor coupled riser mechanical model in one embodiment of the present invention; Figure 4 This is a flowchart of the training process of a multi-factor coupled riser mechanical model in one embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions of the present invention, the present invention is described in detail through specific embodiments. However, it should be understood that the specific embodiments are provided only for a better understanding of the present invention and should not be construed as limiting the present invention. In the description of the present invention, it should be understood that the terminology used is for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0021] To address the shortcomings of existing technologies, such as a lack of comprehensive understanding of multi-factor coupling, inaccurate models that fail to accurately reflect the mechanical response of risers during actual mining operations, leading to significant prediction deviations and failing to meet the demands of safe and efficient mining, this invention proposes a method and system for determining the mechanical response of risers in natural gas hydrate mining based on multi-factor coupling. This method utilizes sensors to acquire marine environmental data of the mining area; obtains seabed geological information through geophysical exploration and geological drilling; and collects real-time operating parameters during the mining process using pressure and flow sensors on the riser. These parameters are used to train a multi-factor coupled riser mechanical model, which is divided into dynamic and static parameters. The dynamic parameters are updated in real-time, and the model is solved using finite element analysis. This method achieves accurate determination and real-time tracking prediction of the riser's mechanical response, effectively providing early warning of potential safety hazards. The following detailed description of the invention, with reference to accompanying drawings and embodiments, further illustrates the solution.

[0022] The structure of the riser system for natural gas hydrate extraction in this invention is as follows: Figure 1 As shown, it includes: a riser 1 and a bend 2 connected to the riser. The riser 1 is the main pipeline for transporting hydrates. An acceleration sensor 3, a strain gauge 4, and a displacement sensor 5 are installed on the riser 1 from top to bottom to monitor acceleration, strain, and riser displacement in real time, providing crucial data for mechanical response analysis. The end of the bend 2 furthest from the riser 1 is connected to a sulfidation pump delivery section 6, which is connected to the output end of a crushing section 7. The other end of the crushing section 7 is connected to a backfilling section 8. During the extraction of natural gas hydrates, the crushing section 7 physically breaks up the rock strata and surrounding rock containing natural gas hydrates in the reservoir. For example, a high-pressure water jet equipped with a drill bit is used to impact the reservoir to form a solid fluidized mixture, which includes natural gas hydrates and rock impurities. This solid fluidized mixture is transported to the sulfidation pump delivery section 6, where it is separated through a spiral flow channel. The denser sediment is discharged, and the natural gas hydrates are transported upwards along the bend 2 and the riser 1. The discharged sediment is fed into the backfill section 8 to replenish the formation material deficit caused by mining and maintain reservoir stability. The sulfide pump delivery section 6 is subjected to complex mechanical forces during mining, and its stress and strain are crucial to the stability of the riser system. Actual mechanical response data of the riser is obtained through acceleration sensors 3, strain gauges 4, and displacement sensors 5. This data is then compared with the predicted mechanical response data obtained from the multi-factor coupled riser mechanical model to verify the accuracy of the model.

[0023] Example 1 This embodiment discloses a method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling, including the following steps: S1 collects data on the environment in which natural gas hydrates are extracted.

[0024] Comprehensive data collection was conducted on the natural gas hydrate extraction environment. The specific data collection process is as follows: Figure 2 As shown, marine environmental data of the mining area is acquired through sensors, including but not limited to ocean current velocity, current direction, wave height, period, and vertical distribution data of seawater temperature and salinity. Seafloor geological information is obtained through geophysical exploration and geological drilling, including but not limited to stratigraphic structure, geotechnical parameters, hydrate saturation, and distribution. Real-time operating parameters during the mining process are collected using pressure and flow sensors on the natural gas hydrate extraction riser, including but not limited to fluid pressure, velocity, and flow rate within the pipe.

[0025] Preprocessing of data on natural gas hydrate extraction environment: Clean the data on natural gas hydrate extraction environment, remove outliers and noise, divide the processed data into training set, validation set and test set, and standardize or normalize the data. Input the processed data into the data queue in batches.

[0026] S2 trains a pre-established multi-factor coupled riser mechanical model based on data from the natural gas hydrate extraction environment.

[0027] This embodiment constructs a multi-factor coupled riser mechanics model, which integrates collected marine environmental data, seabed geological information, and real-time operating parameters into the riser mechanics model, while coupling multiple physical fields such as fluid-structure interaction, heat conduction, and geotechnical mechanics. This multi-factor coupled riser mechanics model can accurately describe the drag force of ocean currents on the riser, the vibration excitation of the riser caused by waves, the heat transfer between seawater and the fluid inside the riser, and the impact of changes in the formation mechanical properties caused by hydrate decomposition on the riser bottom support conditions. In addition, by combining the flow characteristics of multiphase flow (natural gas, water, undecomposed hydrate particles, etc.) inside the riser, it can predict the riser mechanical response relatively accurately.

[0028] like Figure 3As shown, in the multi-factor coupled riser mechanical model, the effect of ocean currents on the riser is calculated using the added mass method to address the fluid-structure interaction performance of the riser. Specifically, the drag force and inertial force of the ocean current on the riser are introduced into the riser's equation of motion through the added mass term for calculation. The vibration excitation of the riser caused by waves is calculated using the Morrison equation. In the added mass method, when calculating the vibration excitation of the riser caused by waves using the Morrison equation, due to the nonlinear characteristics of wave forces, the Morrison equation is decomposed into linear wave forces and nonlinear wave forces for separate calculation, and the calculation results are superimposed.

[0029] Specifically, assuming the riser mass is m and the added mass coefficient is Ca, the equation of motion of the riser under the action of the ocean current is expressed as:

[0030] in, The density of seawater, The volume of seawater displaced by the riser. For the displacement of the riser, For drag force, It is an inertial force. For time.

[0031] Morrison's equations are used to describe wave-induced vibration excitation in risers, and their calculation formula is as follows:

[0032] in, For wave force, The diameter of the riser pipe. This is the drag coefficient. The inertial force coefficient, Let be the velocity of the wave particles. The nonlinear wave force is mainly described by the first term in the formula, while the linear wave force is mainly described by the second term in the formula.

[0033] To assess the thermal conductivity of the riser, a three-dimensional thermal conductivity model of seawater-riser-fluid inside the riser is used for calculation. This model is established based on the convective heat transfer between seawater and the outer wall of the riser, the fluid inside the riser and the inner wall of the riser, and the thermal conductivity of the riser material. The thermal conductivity coefficient in the three-dimensional model is corrected according to the riser material properties and temperature conditions. Furthermore, the heat absorbed during the hydrate dispensing process inside the riser is introduced into the thermal conductivity equation through a source term to accurately describe the impact of temperature changes on the mechanical response during the mining process.

[0034] A three-dimensional heat conduction model is established for the seawater-riser-fluid inside the pipe. In Cartesian coordinates, the heat conduction equation is:

[0035] in, For material density, For specific heat capacity, For temperature, For time, The thermal conductivity coefficient, The term represents the heat source, used for the endothermic effect during the decomposition of hydrates. x, y, and z are the x-axis, y-axis, and z-axis coordinates in the Cartesian coordinate system, respectively.

[0036] The formula for convective heat transfer between seawater and the outer wall of the riser is:

[0037] in, The convective heat transfer coefficient between seawater and the outer wall of the riser pipe is given. For seawater temperature, Temperature of the outer wall of the riser. This is the derivative of the temperature of the outer wall of the riser along the normal direction.

[0038] Similarly, the formula for convective heat transfer between the fluid inside the pipe and the inner wall of the riser is:

[0039] in, The convective heat transfer coefficient between the fluid inside the pipe and the inner wall of the riser is given. The temperature of the fluid inside the pipe. This refers to the temperature of the inner wall of the riser.

[0040] The geomechanical properties of risers are calculated using a constitutive model based on elastoplastic mechanics. This model characterizes the multiphase flow performance within the riser and employs the Euler-Lagrange equations for calculation. The constitutive model uses the soil support force at the bottom of the riser as a boundary condition, while also incorporating the long-term effects of soil settlement and deformation on the riser's mechanical response. Similar to linear elastic constitutive relations, ,in, For stress tensor, For strain tensor, The elasticity matrix. During hydrate decomposition, the elasticity matrix... It will change with factors such as hydrate saturation. Assuming the hydrate saturation is... Through experimental or theoretical analysis, The relationship.

[0041] Assuming the pressure at the contact surface between the bottom of the riser and the formation is According to the equilibrium condition of forces, that is:

[0042] Where A is the contact area between the bottom of the riser and the stratum. This is the supporting force of the stratum for the riser.

[0043] Simultaneously, the long-term effects of ground settlement and deformation on the riser's mechanical response are considered, assuming the ground settlement is... The displacement boundary condition at the bottom of the riser is: ,in This represents the displacement at the bottom of the riser.

[0044] The Euler-Lagrange equations treat natural gas, water, and undecomposed hydrate particles as different phases. Through the interactions between these phases, such as interphase friction, mass transfer, and heat transfer, the flow characteristics of multiphase flow in a pipe are accurately described in relation to the mechanical response of the riser.

[0045] Taking the natural gas phase and the water phase as an example, the equation of motion for the natural gas phase is:

[0046] The equation of motion for the water phase is:

[0047] in, These are the velocity vectors of the natural gas phase and the water phase, respectively. The densities of the natural gas phase and the water phase are respectively. These are the external forces (such as gravity, pressure gradient force, etc.) acting on the natural gas phase and the water phase, respectively. These are the interphase forces between the natural gas phase and the water phase (such as friction, forces caused by mass transfer and heat transfer).

[0048] The method for training a pre-established multi-factor coupled riser mechanical model based on data from the natural gas hydrate extraction environment is as follows: Set hyperparameters such as training epochs, learning rate, and convergence threshold, and clear the training history of the multi-factor coupled riser mechanics model. Retrieve data batches sequentially from the data queue, marking each batch as enqueued and ready for training. When the data queue is not empty, a data batch is extracted from the data queue and input into the multi-factor coupled riser mechanics model. The forward propagation algorithm is executed to obtain the predicted riser mechanics response. The predicted riser mechanics response is compared with the true value of the data batch, and loss functions such as mean squared error and mean absolute error are used to calculate the loss value. If the loss value is less than the threshold, the training of the multi-factor coupled riser mechanics model ends; otherwise, the backpropagation algorithm is performed based on the loss value. The gradient of the model parameters is calculated through optimization algorithms such as stochastic gradient descent (SGD), Adagrad, and Adam, and the model parameters are updated according to the gradient. It is checked whether the updated model has generated new monitoring data. If so, the new data is preprocessed and added to the data queue. If some data batches in the data queue do not perform well during training, their priority in the queue can be readjusted, or they can be reprocessed and re-added to the queue. The training continues in a loop until the model meets the convergence condition.

[0049] S3 divides the model parameters in the multi-factor coupled riser mechanical model into static parameters and dynamic parameters. Static parameters are parameters that remain unchanged during the mining process; dynamic parameters are parameters that change over time, and the dynamic parameters are corrected accordingly.

[0050] Static parameters include the elastic modulus and Poisson's ratio of the riser material, while dynamic parameters include marine environmental parameters, fluid parameters within the riser, and formation mechanical parameters affected by hydrate decomposition. Based on real-time monitoring data, the dynamic parameters are corrected using Kalman filtering or particle filtering algorithms.

[0051] like Figure 4 As shown, the method for correcting dynamic parameters is as follows: Based on the state and process noise of the multi-factor coupled riser mechanics model at the previous moment, the data from the previous moment is input into the multi-factor coupled riser mechanics model to obtain the prior estimate for the current moment; the prior estimate is fused with the data at the current moment, and based on the fusion result, the optimal estimate for the current moment is calculated using Kalman gain or particle weight; the corresponding data parameters are updated based on the optimal estimate, and the updated data parameters are input into the multi-factor coupled riser mechanics model to calculate the mechanical response of the riser at the next moment. By continuously repeating parameter correction, real-time tracking and prediction of the riser's mechanical response are achieved.

[0052] S4 generates the final multi-factor coupled riser mechanical model based on the static parameters and the corrected dynamic parameters. The test data is then input into the final multi-factor coupled riser mechanical model to obtain the riser mechanical response results.

[0053] S5 provides early warning of potential safety hazards based on the mechanical response results of the riser.

[0054] Based on the riser mechanical response results, combined with time series analysis and machine learning, the riser mechanical response in the future period is predicted, and potential safety hazards are warned.

[0055] The mechanical model of a multi-factor coupled riser is solved using numerical calculation methods. Specifically, the mechanical model is discretized using the finite element method or the finite difference method, and the mechanical response of the riser under different mining stages and environmental conditions is solved through computer simulation. The mechanical response includes the stress distribution, strain, displacement changes, and vibration characteristics of the riser.

[0056] The riser is divided into finite element meshes of appropriate sizes. For critical sections of the riser, a finer mesh is used, particularly for the connection between the riser bottom and the seabed strata. Critical sections include, but are not limited to, curved sections, connection points, and areas near the seabed, where finer meshing is employed to improve computational accuracy. During mesh generation, to ensure that mesh quality indicators such as aspect ratio and Jacobian determinant meet computational requirements, excessive errors in the calculation results due to mesh quality issues are avoided. When setting boundary conditions, the constraints at the top of the riser, such as fixed constraints, hinged constraints, or elastic constraints, are accurately set based on the actual mining environment and the riser installation conditions. For the connection between the riser bottom and the seabed strata, the support forces and constraints calculated from the geotechnical mechanics model are applied. Simultaneously, boundary conditions for external environmental factors such as ocean currents and waves are considered, such as setting velocity inlets and pressure outlets on the model boundaries, to ensure the rationality and accuracy of the boundary conditions. During the calculation process, appropriate solvers and algorithm parameters, such as time step and iteration convergence accuracy, are selected. The time step is adjusted based on the dynamic characteristics of the mining process and the stability of the model to ensure that transient changes in the riser's mechanical response are captured; the iteration convergence accuracy is set to 10. -6 To ensure the convergence and accuracy of the calculation results, the solver and algorithm parameters were optimized through multiple trials and comparisons to improve computational efficiency and accuracy.

[0057] After the multi-factor coupled riser mechanical model is trained, it is validated using a validation set, and its hyperparameters are adjusted. The generalization performance of the multi-factor coupled riser mechanical model is evaluated using a test set to ensure its accuracy and reliability in practical applications. The mechanical response results obtained from numerical calculations are compared and analyzed with actual monitoring or experimental data. In this embodiment, statistical analysis methods, such as root mean square error (RMSE) and mean absolute error (MAE), are used to evaluate the degree of agreement between the model calculation results and the actual monitoring data. If discrepancies exist, the model parameters of the multi-factor coupled riser mechanical model are adjusted and optimized, and the calculations and validations are repeated until the model calculation results achieve a reasonable degree of agreement with the actual situation, thereby accurately determining the mechanical response of the natural gas hydrate extraction riser under multi-factor coupling.

[0058] Simultaneously, laboratory simulation experiments were conducted to construct an experimental model similar to actual mining conditions. In the laboratory, coupled conditions involving multiple factors such as ocean currents, waves, heat conduction, and hydrate decomposition were simulated to measure the mechanical response of the riser. The numerical calculation results were then compared and verified with the experimental data.

[0059] If there are discrepancies between the model calculation results and the actual monitoring or experimental data, analyze the causes of the discrepancies, such as unreasonable model parameter selection, inaccurate boundary condition settings, or numerical calculation errors. For each cause, take corresponding adjustment measures, such as re-optimizing model parameters, correcting boundary conditions, and improving numerical calculation methods. Then, re-perform numerical calculations and verify the results until the error between the model calculation results and the actual situation is controlled within a reasonable range. Generally, the root mean square error should be less than 5%, and the mean absolute error should be less than 3%, thereby ensuring accurate determination of the mechanical response of the natural gas hydrate extraction riser under the coupled effects of multiple factors.

[0060] Example 2 Based on the same inventive concept, this invention also discloses a system for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling, comprising: The data acquisition module is used to collect data on the natural gas hydrate extraction environment; The model training module is used to train a pre-established multi-factor coupled riser mechanics model based on data from the natural gas hydrate extraction environment. The model parameter correction module is used to divide the model parameters in the multi-factor coupled riser mechanics model into static parameters and dynamic parameters. Static parameters are parameters that remain unchanged during the mining process, while dynamic parameters are parameters that change over time. The module corrects the dynamic parameters. The model output module is used to generate the final multi-factor coupled riser mechanical model based on static parameters and corrected dynamic parameters. The test data is input into the final multi-factor coupled riser mechanical model to obtain the riser mechanical response results. The early warning module is used to provide early warnings of potential safety hazards based on the mechanical response results of the riser.

[0061] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0062] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0063] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0064] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0065] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific embodiments of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention. The above content is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims

1. A method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling, characterized in that, Includes the following steps: Collect data on the environment of natural gas hydrate extraction; Based on the data of the natural gas hydrate extraction environment, the pre-established multi-factor coupled riser mechanical model was trained; The model parameters in the multi-factor coupled riser mechanical model are divided into static parameters and dynamic parameters. The static parameters are parameters that remain unchanged during the mining process, while the dynamic parameters are parameters that change over time. The dynamic parameters are then corrected. Based on the static parameters and the corrected dynamic parameters, a final multi-factor coupled riser mechanical model is generated. The test data is then input into the final multi-factor coupled riser mechanical model to obtain the riser mechanical response results. Safety hazard warnings are issued based on the mechanical response results of the riser.

2. The method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling as described in claim 1, characterized in that, Data collected on the natural gas hydrate extraction environment includes: acquiring marine environmental data of the extraction area through sensors; obtaining seabed geological information through geophysical exploration and geological drilling; and collecting real-time operating parameters during the extraction process through pressure and flow sensors on the natural gas hydrate extraction riser.

3. The method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling as described in claim 2, characterized in that, Preprocessing of data on natural gas hydrate extraction environment: The data on natural gas hydrate extraction environment is cleaned to remove outliers and noise. The processed data is divided into training set, validation set and test set. The data is then standardized or normalized. The processed data is input into the data queue in batches.

4. The method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling as described in claim 3, characterized in that, The method for training the pre-established multi-factor coupled riser mechanics model based on the data of the natural gas hydrate extraction environment is as follows: extract a data batch from the data queue, input the data batch into the multi-factor coupled riser mechanics model, execute the forward propagation algorithm to obtain the predicted riser mechanical response, compare the predicted riser mechanical response with the true value of the data batch, calculate the loss value, and if the loss value is less than the threshold, then the training of the multi-factor coupled riser mechanics model is terminated. Otherwise, backpropagation is performed based on the loss value to calculate the gradient of the model parameters, and the model parameters are updated according to the gradient. It is then checked whether the updated model has generated new monitoring data. If so, the new data is preprocessed and added to the data queue, and training is continuously performed in a loop until the model meets the convergence condition.

5. The method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling as described in claim 1, characterized in that, In the multi-factor coupled riser mechanical model, the fluid-structure interaction performance of the riser is calculated using the added mass method, and the vibration excitation of the riser caused by waves is calculated using the Morrison equation; the thermal conductivity performance of the riser is calculated using a three-dimensional thermal conductivity model of seawater-riser-fluid inside the riser; the geotechnical mechanical properties of the riser are calculated using a constitutive model based on elastoplastic mechanics; and the multiphase flow performance inside the riser is calculated using the Euler-Lagrange equation.

6. The method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling as described in claim 5, characterized in that, The added mass method involves incorporating the drag force and inertial force of the ocean current on the riser into the riser's equation of motion through an added mass term. When calculating the riser vibration excitation caused by waves using the Morrison equation, the Morrison equation is decomposed into linear wave force and nonlinear wave force for separate calculation, and the results are superimposed. When establishing a three-dimensional heat conduction model of seawater-riser-fluid inside the riser, the heat conduction coefficient in the three-dimensional heat conduction model is corrected according to the riser material properties and temperature conditions, and the heat absorbed during the hydrate dispensing process inside the riser is incorporated into the heat conduction equation through a source term. The constitutive model uses the support force of the formation on the bottom of the riser as a boundary condition, and also incorporates the long-term effects of formation settlement and deformation on the mechanical response of the riser. The Euler-Lagrange equation treats natural gas, water, and undecomposed hydrate particles as different phases, and accurately describes the effect of the flow characteristics of multiphase flow in the pipe on the mechanical response of the riser through the interaction between the phases.

7. The method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling as described in any one of claims 1-6, characterized in that, Dynamic parameters are corrected using Kalman filtering or particle filtering algorithms.

8. The method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling as described in claim 7, characterized in that, The method for correcting the dynamic parameters is as follows: The data from the previous moment is input into the multi-factor coupled riser mechanical model to obtain the prior estimate for the current moment; The prior estimate is fused with the data at the current moment, and the optimal estimate at the current moment is calculated based on the fusion result. The corresponding data parameters are updated based on the optimal estimate, and the updated data parameters are input into the multi-factor coupled riser mechanical model to calculate the mechanical response of the riser at the next moment.

9. The method for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling as described in any one of claims 1-6, characterized in that, The multi-factor coupled riser mechanical model is discretized using the finite element method or finite difference method to solve the mechanical response of the riser under different mining stages and environmental conditions. The riser is divided into finite element meshes of appropriate size, and the mesh is refined for key parts of the riser. For the connection between the bottom of the riser and the seabed strata, the support force and constraint conditions calculated by the geotechnical mechanics model are applied.

10. A system for determining the mechanical response of a riser in natural gas hydrate extraction based on multi-factor coupling, characterized in that, include: The data acquisition module is used to collect data on the natural gas hydrate extraction environment; The model training module is used to train a pre-established multi-factor coupled riser mechanics model based on the data of the natural gas hydrate extraction environment. The model parameter correction module is used to divide the model parameters in the multi-factor coupled riser mechanics model into static parameters and dynamic parameters. The static parameters are parameters that remain unchanged during the mining process, while the dynamic parameters are parameters that change over time. The module corrects the dynamic parameters. The model output module is used to generate the final multi-factor coupled riser mechanical model based on the static parameters and the corrected dynamic parameters, input the test data into the final multi-factor coupled riser mechanical model, and obtain the riser mechanical response results. The early warning module is used to provide early warning of potential safety hazards based on the mechanical response results of the riser.

Images