A method and system for optimizing the shape of a dry tree cylindrical platform pendulum

Abstract

The present application relates to a kind of for dry tree cylinder type platform heave body shape optimization design method and system, it includes: based on the design variable of determined heave body, constraint condition and objective function, the initial design variable of heave body is generated, wherein, with heave body geometric feature as design variable;Adaptation function is established using trained neural network model, and the fitness value of geometric feature variable is calculated, to evaluate the performance of geometric feature variable on objective function;Judge whether the calculation of geometric feature variable fitness value meets termination condition, meet then output best heave body shape design scheme, otherwise the fitness value of geometric feature variable is selected, crossed and mutated, and new optimization variable is generated, and the fitness value is recalculated.The present application can obtain high-performance heave body type, improve the heave motion of platform, solve the heave body shape optimization design problem in the field of ocean engineering;It can be applied in the field of ocean structure optimization design.

Description

Technical Field

[0001] This invention relates to the field of marine structure optimization design technology, and in particular to a method and system for optimizing the shape of a helical body of a dry tree cylindrical platform. Background Technology

[0002] With the rapid growth in global demand for oil and gas resources and the decreasing availability of onshore oil and gas, offshore oil and gas production has become a significant contributor to the increase in oil and gas resources. Dry production systems on offshore oil production platforms can autonomously perform drilling and completion functions, avoiding production interruptions. To apply dry production trees to cylindrical platforms, which are characterized by simple structure, ease of construction and maintenance, and superior motion performance, further improvements to the heave motion performance of cylindrical platforms are needed. Designing and optimizing high-performance heave suppression structures has always been a key focus and challenge in the field of marine engineering. Currently, the damping performance of dry production tree platforms can be improved by designing and optimizing the shape of the heave body, thereby mitigating the platform's heave motion.

[0003] Existing fast methods for hull form optimization based on the expectation-maximization algorithm (EMA) in the literature suffer from high computational complexity, requiring the introduction of penalty functions or other methods to handle constraint problems. Existing hull form optimization design methods and systems based on geometric deformation space features primarily focus on the geometric characteristics of the shape when dealing with the shape optimization problem of marine structures, neglecting the physical properties and constraints of the structure, and failing to incorporate important considerations into the geometric deformation space feature method. Existing high-energy-efficiency hull form optimization design methods based on two stages, when using multi-objective optimization algorithms to optimize structures, struggle to determine the objective function for complex nonlinear problems. Furthermore, existing methods for high-speed planing boat resistance optimization based on machine learning and CFD methods, when using genetic algorithms for structural optimization, rely on known mathematical models as the objective function, failing to fully consider the nonlinearity of the structure's hydrodynamic characteristics. Summary of the Invention

[0004] To address the aforementioned problems, the purpose of this invention is to provide a method and system for optimizing the shape of a heave body on a dry tree cylindrical platform. This method can achieve a high-performance heave body shape, improve the platform's heave motion, and solve the problem of optimizing the shape of a heave body in the field of marine engineering.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for optimizing the shape design of a helical body on a cylindrical platform, comprising: generating initial design variables of the helical body based on determined design variables, constraints, and objective function of the helical body, wherein the geometric features of the helical body are used as design variables; establishing a fitness function using a trained neural network model, calculating the fitness values ​​of the geometric feature variables to evaluate the performance of the geometric feature variables on the objective function; determining whether the calculation of the fitness values ​​of the geometric feature variables meets the termination condition, if so, outputting the optimal helical body shape design scheme, otherwise, performing selection, crossover, and mutation operations on the fitness values ​​of the geometric feature variables to generate new optimization variables, and recalculating the fitness values.

[0006] Furthermore, the geometric characteristics of the helical body are used as design variables, including: the diameter, width, thickness, and the location and number of openings of the helical body.

[0007] Furthermore, the constraints include: platform size limitations and changes in displacement.

[0008] Furthermore, the objective function is: the heave motion, added mass, and damping of the dry tree cylindrical platform.

[0009] Furthermore, the training of the neural network model includes:

[0010] Determine the design variables and objective function of the neural network model, and construct the database;

[0011] Visualize the design variables and objective function to intuitively understand the relationship between them;

[0012] The design variables and objective function are normalized, and geometric feature processing and structural material field characteristics are calculated. Feature extraction and transformation are performed on the normalized data, and the transformed dataset is divided into training set and test set.

[0013] The convolutional neural network model is trained and tuned using a training set, and the trained neural network model is tested using a test set.

[0014] Furthermore, the design variables and objective function of the neural network model are determined, and a database is constructed, including:

[0015] The heave motion, added mass, damping, and flow field of the dry tree cylindrical platform are selected as objective functions, while the geometric characteristics of the heave body and marine environmental conditions are selected as design variables.

[0016] The heave motion, added mass, damping, and hydrodynamic characteristics of the initial scheme group were obtained by using potential flow theory calculations, CFD numerical calculations, and experiments. A database was constructed based on the existing calculation and experimental data.

[0017] Furthermore, the design variables and objective function are visualized, including:

[0018] Multicollinearity analysis was performed on the design variables, and the variance inflation coefficient was used to assess the multicollinearity among the design variables.

[0019] The correlation coefficient between the design variables and the objective function was evaluated using the Pearson correlation coefficient, and the correlation coefficient was plotted as a correlation matrix heatmap.

[0020] A system for optimizing the shape of a helical body on a cylindrical platform shaped like a dry tree includes: a first processing module that generates initial design variables for the helical body based on determined design variables, constraints, and an objective function, wherein geometric features of the helical body are used as design variables; a second processing module that establishes a fitness function using a trained neural network model and calculates fitness values ​​for the geometric feature variables to evaluate their performance on the objective function; and an optimization module that determines whether the calculation of the fitness values ​​of the geometric feature variables meets the termination condition. If it does, the optimal helical body shape design scheme is output; otherwise, the fitness values ​​of the geometric feature variables are subjected to selection, crossover, and mutation operations to generate new optimization variables, and the fitness values ​​are recalculated.

[0021] A computer-readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by a computing device, cause the computing device to perform any of the methods described above.

[0022] A computing device includes: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for performing any of the methods described above.

[0023] The present invention has the following advantages due to the adoption of the above technical solutions:

[0024] 1. Traditional methods are based on mathematical models or empirical formulas for optimization design, which have limited ability to model nonlinear problems. However, the neural network model of this invention can construct the complex nonlinear relationship between the dry tree cylindrical platform and marine environmental conditions.

[0025] 2. Neural network algorithms possess adaptive learning capabilities, allowing for parameter adjustments and model updates based on data, thus progressively optimizing the model. This enables neural networks to adapt to different marine structural problems and adaptively model the complexity of the problem and the characteristics of the data.

[0026] 3. This invention employs a genetic algorithm as a multi-objective optimization algorithm, capable of handling multiple conflicting optimization objectives. Combined with a neural network algorithm, it can achieve trade-offs and optimizations among multiple objectives, generating a set of non-dominated solutions on the Pareto front.

[0027] 4. This invention employs a neural network algorithm to automatically extract important features from the input data, reducing the need for manual feature engineering. This has certain advantages for complex data and a large number of variables in marine structure optimization problems. Attached Figure Description

[0028] Figure 1 This is a flowchart of the optimization design method in an embodiment of the present invention;

[0029] Figure 2 This is a flowchart of the neural network model training process in an embodiment of the present invention. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0031] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0032] To address the challenge of optimizing the shape of heave bodies in marine engineering, this invention provides a method and system for optimizing the shape of heave bodies on a tree-shaped cylindrical platform. This method combines machine learning algorithms and multi-objective optimization algorithms to optimize the shape of the heave body on the tree-shaped cylindrical platform, taking into account the nonlinear problems in the complex motion process of the platform, thereby obtaining a high-performance heave body shape and improving the heave motion of the platform.

[0033] In one embodiment of the present invention, a method for optimizing the shape of a cylindrical platform heave is provided. In this embodiment, a genetic optimization algorithm is used for optimization design; specifically, as shown... Figure 1 As shown, the method includes the following steps:

[0034] 1) Based on the determined design variables, constraints and objective function of the helical body, generate the initial design variables of the helical body, wherein the geometric characteristics of the helical body are used as design variables;

[0035] 2) Establish a fitness function using a trained neural network model and calculate the fitness values ​​of the geometric feature variables to evaluate the performance of the geometric feature variables on the objective function;

[0036] 3) Determine whether the calculation of the fitness value of the geometric feature variable meets the termination condition. If it does, output the optimal heave shape design scheme; otherwise, perform selection, crossover and mutation operations on the fitness value of the geometric feature variable to generate new optimization variables and recalculate the fitness value.

[0037] In step 1) above, the geometric characteristics of the helical body are used as design variables, including: the diameter, width, thickness, and the location and number of openings of the helical body. The initial population is constructed based on the geometric characteristics of the helical body.

[0038] In step 1) above, the constraints include: platform size limitations and changes in displacement.

[0039] In step 1) above, the objective function is: the heave motion, added mass, and damping of the dry tree cylindrical platform.

[0040] In step 2) above, if Figure 2 As shown, training a neural network model includes the following steps:

[0041] 2.1) Determine the design variables and objective function of the neural network model, and construct the database;

[0042] 2.2) Visualize the design variables and objective function to intuitively obtain the relationship between them;

[0043] 2.3) Normalize the design variables and objective function, and use geometric feature processing to calculate the characteristics of the structural material field. Extract and transform the normalized data, and divide the transformed dataset into training set and test set.

[0044] The min-max normalization method is used to normalize the data, ensuring data quality and consistency. The normalization formula is:

[0045]

[0046] 2.4) The training set is used to train and optimize the convolutional neural network model (CNN), and the test set is used to test the trained neural network model.

[0047] In step 2.1) above, the design variables and objective function of the neural network model are determined, and the database is constructed, specifically as follows:

[0048] 2.1.1) The heave motion, added mass, damping and flow field of the dry tree cylindrical platform are selected as the objective function, and the geometric characteristics of the heave body and marine environmental conditions are selected as the design variables.

[0049] The design variables include: main cylinder diameter, sag diameter, width, thickness, and the location and number of openings.

[0050] 2.1.2) The heave motion, added mass, damping and hydrodynamic characteristics of the initial scheme group were obtained by using potential flow theory calculation, CFD numerical calculation and experiment. A database was constructed based on the existing calculation data and experimental data.

[0051] In step 2.2) above, the design variables and objective function are visualized, including the following steps:

[0052] 2.2.1) Perform multicollinearity analysis on the design variables such as diameter, width, thickness, and location and number of openings of the helical body, and use the variance expansion coefficient (VIF) to evaluate the multicollinearity among the design variables;

[0053] VIF is:

[0054]

[0055] Among them, R i 2 To test the statistical indicators.

[0056] 2.2.2) The Pearson correlation coefficient is used to evaluate the correlation coefficient between the design variables such as the diameter, width, thickness, and number of openings of the heave body and the objective function (heave motion, added mass, damping, flow field, etc. of the dry tree cylindrical platform), and the correlation coefficient is plotted as a correlation matrix heat map.

[0057] Among them, the correlation coefficient R X,Y for:

[0058]

[0059] In the formula, cov(X,Y) is the covariance between the design variables and the objective function, and σ X σ Y The standard deviation between the design variables and the objective function.

[0060] Following step 3) above, a verification step is also included. A computational model is established using optimized design data, and the optimal design scheme for the generated dry tree cylindrical platform heave body shape is analyzed and verified using trained neural network models, potential flow theory, CFD methods, experiments, and other methods.

[0061] In summary, this invention employs a convolutional neural network (CNN) model to construct a nonlinear hydrodynamic model of a dry-tree cylindrical platform with a heave body, overcoming the limitations of traditional mathematical models or empirical formulas. A neural network algorithm is introduced into the genetic algorithm, using the trained model as the objective function to replace the simplified mathematical model, thus addressing the limitations of the objective function in multi-objective optimization algorithms. A computational model is established using optimal design data, and the optimal design scheme for the heave body shape of the generated dry-tree cylindrical platform is analyzed and verified using trained machine learning models, potential flow theory, CFD algorithms, and experiments. Multiple methods are employed to verify the performance of the optimal design model.

[0062] In one embodiment of the present invention, a system for optimizing the shape of a swaying body of a cylindrical platform for trunk trees is provided, comprising:

[0063] The first processing module generates the initial design variables of the helical body based on the determined design variables, constraints and objective function of the helical body, wherein the geometric characteristics of the helical body are used as design variables.

[0064] The second processing module uses a trained neural network model to establish a fitness function and calculates the fitness values ​​of the geometric feature variables to evaluate the performance of the geometric feature variables on the objective function.

[0065] The optimization module determines whether the calculation of the fitness value of the geometric feature variable meets the termination condition. If it does, it outputs the optimal heave shape design scheme; otherwise, it performs selection, crossover, and mutation operations on the fitness value of the geometric feature variable to generate new optimization variables and recalculates the fitness value.

[0066] Furthermore, the geometric characteristics of the helical body are used as design variables, including: the diameter, width, thickness, and the location and number of openings of the helical body.

[0067] Furthermore, the constraints include: platform size limitations and changes in displacement.

[0068] Furthermore, the objective function is: the heave motion, added mass, and damping of the dry tree cylindrical platform.

[0069] Furthermore, the training of the neural network model includes:

[0070] Determine the design variables and objective function of the neural network model, and construct the database;

[0071] Visualize the design variables and objective function to intuitively understand the relationship between them;

[0072] The design variables and objective function are normalized, and geometric feature processing and structural material field characteristics are calculated. Feature extraction and transformation are performed on the normalized data, and the transformed dataset is divided into training set and test set.

[0073] The convolutional neural network model is trained and tuned using a training set, and the trained neural network model is tested using a test set.

[0074] Furthermore, the design variables and objective function of the neural network model are determined, and a database is constructed, including:

[0075] The heave motion, added mass, damping, and flow field of the dry tree cylindrical platform are selected as objective functions, while the geometric characteristics of the heave body and marine environmental conditions are selected as design variables.

[0076] The heave motion, added mass, damping, and hydrodynamic characteristics of the initial scheme group were obtained by using potential flow theory calculations, CFD numerical calculations, and experiments. A database was constructed based on the existing calculation and experimental data.

[0077] Furthermore, the design variables and objective function are visualized, including:

[0078] Multicollinearity analysis was performed on the design variables, and the variance inflation coefficient was used to assess the multicollinearity among the design variables.

[0079] The correlation coefficient between the design variables and the objective function was evaluated using the Pearson correlation coefficient, and the correlation coefficient was plotted as a correlation matrix heatmap.

[0080] The system provided in this embodiment is used to execute the above-described method embodiments. For specific processes and details, please refer to the above embodiments, which will not be repeated here.

[0081] A schematic diagram of a computing device structure is provided in one embodiment of the present invention. The computing device can be a terminal, which may include: a processor, a communication interface, memory, a display screen, and an input device. The processor, communication interface, and memory communicate with each other via a communication bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and a computer program. When the computer program is executed by the processor, it implements a method for optimizing the shape of a cylindrical platform helical body. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, a management network, NFC (Near Field Communication), or other technologies. The display screen can be a liquid crystal display or an e-ink display. The input device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad mounted on the casing of the computing device, or an external keyboard, touchpad, or mouse. The processor can call logical instructions stored in the memory.

[0082] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0083] In one embodiment of the present invention, a computer program product is provided, the computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, and when the program instructions are executed by a computer, the computer is able to perform the methods provided in the above-described method embodiments.

[0084] In one embodiment of the present invention, a non-transitory computer-readable storage medium is provided, which stores server instructions that cause a computer to perform the methods provided in the above embodiments.

[0085] The computer-readable storage medium provided in the above embodiments has a similar implementation principle and technical effect to the above method embodiments, and will not be described again here.

[0086] 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.

[0087] 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.

[0088] 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.

[0089] 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 them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing the shape of a cylindrical platform swaying over a dry tree, characterized in that, include: Based on the determined design variables, constraints, and objective function of the helical body, the initial design variables of the helical body are generated, wherein the geometric characteristics of the helical body are used as design variables; the geometric characteristics of the helical body as design variables include: the diameter, width, thickness, and location and number of openings of the helical body; the constraints include: platform size restrictions and changes in drainage volume; the objective function is: the helical motion, added mass, and damping of the dry tree cylindrical platform; A fitness function is established using a trained neural network model, and the fitness values ​​of the geometric feature variables are calculated to evaluate the performance of the geometric feature variables on the objective function. Determine whether the calculation of the fitness value of the geometric feature variable meets the termination condition. If it does, output the optimal heave shape design scheme. Otherwise, perform selection, crossover and mutation operations on the fitness value of the geometric feature variable to generate new optimization variables and recalculate the fitness value. The training of the neural network model includes: determining the design variables and objective function of the neural network model, and constructing a database, including: selecting the heave motion, added mass, damping, and flow field of the dry tree cylindrical platform as the objective function, and the geometric characteristics of the heave body and marine environmental conditions as design variables; using potential flow theory calculation, CFD numerical calculation, and experiments to obtain the heave motion, added mass, damping, and flow field hydrodynamic characteristics of the initial scheme group, and constructing a database based on existing calculation data and experimental data; The design variables and objective function are visualized to intuitively obtain the relationship between them. This includes: performing multicollinearity analysis on the design variables and using the variance inflation coefficient to assess the multicollinearity among the design variables; using the Pearson correlation coefficient to assess the correlation coefficient between the design variables and the objective function, and plotting the correlation coefficient into a correlation matrix heatmap. The design variables and objective function are normalized, and geometric feature processing and structural material field characteristics are calculated. Feature extraction and transformation are performed on the normalized data, and the transformed dataset is divided into training set and test set. The training set is used to train and optimize the convolutional neural network model, and the test set is used to test the trained neural network model. It also includes a verification step: using optimized design data to establish a computational model, and using the trained neural network model, potential flow theory, CFD method, and experimental method to analyze and verify the optimal design scheme of the generated dry tree cylindrical platform heave body shape.

2. A system for optimizing the shape of a cylindrical platform swaying over a dry tree, used to implement the method for optimizing the shape of a cylindrical platform swaying over a dry tree as described in claim 1, characterized in that, include: The first processing module generates the initial design variables of the helical body based on the determined design variables, constraints and objective function of the helical body, wherein the geometric characteristics of the helical body are used as design variables. The second processing module uses a trained neural network model to establish a fitness function and calculates the fitness values ​​of the geometric feature variables to evaluate the performance of the geometric feature variables on the objective function. The optimization module determines whether the calculation of the fitness value of the geometric feature variable meets the termination condition. If it does, it outputs the optimal heave shape design scheme; otherwise, it performs selection, crossover, and mutation operations on the fitness value of the geometric feature variable to generate new optimization variables and recalculates the fitness value.

3. A computer-readable storage medium for storing one or more programs, characterized in that, The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform the method as described in claim 1.

4. A computing device, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing the method of claim 1.

Images