Ocean structure safety forecasting method and system based on proxy model

Abstract

The invention discloses an ocean structure safety forecasting method and system based on an agent model. The method comprises the steps that structural factor parameters and environmental factor parameters needing to be considered are determined; permutation and combination are carried out on all the parameters to obtain various typical working conditions of the ocean structure; constructing a numerical model of the ocean structure, respectively calculating various typical working conditions through the numerical model, and generating a structural response database of the ocean structure under different working conditions; a high-precision ocean structure agent model meeting the precision requirement is constructed according to the structure response database; inputting structural factor and environmental factor parameters into the agent model, obtaining a safety state response result of the ocean structure under a corresponding working condition, and realizing structural safety forecasting of the ocean structure; the method can comprehensively monitor and evaluate the safety state of the ocean structure, makes up for the limitation of sensor monitoring, and is lower in use threshold compared with a finite element method.

Description

Technical Field

[0001] The present invention belongs to the technical field of marine structure safety forecasting, and more specifically, relates to a marine structure safety forecasting method and system based on an agent model. Background Art

[0002] During operation, marine structures are subjected to a variety of complex loads such as wind, waves, and currents, which pose challenges to the structures. Especially in severe weather, marine structures have to face huge waves and storms, which may cause severe impact and distortion of the structures, thus causing structural safety issues.

[0003] At present, the main means of monitoring the safety status of marine structures still relies on sensors. Sensors are placed in hot spots or key areas of marine structures, and can monitor and record the strength and stiffness of local areas in real time. However, marine structures are large in size, have many systems, and are very complex. There are mutual influences and correlations between their various parts. Only relying on local area data obtained by sensor monitoring cannot fully reflect the safety status of the entire marine structure. In addition, marine structures are expensive, and once damaged, they are very likely to cause economic losses, casualties, environmental damage and other serious consequences. However, sensors can only provide real-time data feedback, but cannot predict future trends, so they cannot predict the safety of the structure. This means that the maintenance and guarantee of marine structures can only be handled after the fact, and cannot be prevented in advance.

[0004] In order to overcome the limitations of sensor monitoring, some marine structures have begun to try to install finite element analysis software. Finite element analysis is a powerful engineering analysis tool. By dividing the entire marine structure into countless small units, when the environmental conditions that may be faced in the future are obtained, the response of the marine structure under such conditions can be simulated in advance to achieve safety forecasts. However, the threshold for using finite element software is high. On the one hand, finite element software is expensive, and it takes a lot of money to purchase and maintain a set of high-performance finite element analysis software, which limits its wide application in marine structure monitoring; on the other hand, operating finite element analysis software requires deep professional knowledge and skills. Finite element analysis involves a lot of mathematical and mechanical knowledge, and engineers need to have a solid theoretical foundation and practical experience. In addition, the software's operating interface is complex, the parameter settings are numerous, and the requirements for users are high. Therefore, a lot of time and resources need to be invested to train employees, or to hire professional analysts to operate. This undoubtedly increases operating costs and manpower burdens. What's more complicated is that for marine structures such as self-elevating platforms, when the platform moves up and down along the pile legs, its corresponding finite element model will also change. This means that after each move, engineers need to re-establish and analyze the model to ensure the accuracy of the analysis. This frequent modeling work is not only inefficient but also prone to errors, bringing additional challenges to the safety monitoring of marine structures.

[0005] In summary, most marine structures are large in size and complex in structure, and the consequences of damage are serious. However, current technology has limitations in predicting their safety status. Therefore, when designing a marine structure health monitoring system, on the one hand, it is necessary to find a method that can comprehensively monitor and evaluate the safety status of marine structures to make up for the limitations of sensor monitoring; on the other hand, it is also necessary to pay attention to cost-effectiveness and ease of operation to ensure the practicality and sustainability of the system. Summary of the invention

[0006] In view of the above defects or improvement needs of the prior art, the present invention provides a method and system for marine structure safety prediction based on an agent model, which can obtain the overall safety status of the entire marine structure and can predict the structural safety status response under various working conditions in advance; it can comprehensively monitor and evaluate the safety status of marine structures, make up for the limitations of sensor monitoring, and has a lower usage threshold than the finite element method; it is low-cost and easy to operate, and can ensure the practicality and sustainability of marine structure safety prediction.

[0007] In order to achieve the above object, one aspect of the present invention provides a method for marine structure safety prediction based on an agent model, comprising the following steps:

[0008] S1: Determine the structural factor parameters and environmental factor parameters that need to be considered according to the classification society rules and marine structure design requirements;

[0009] S2: Arrange and combine the structural factor parameters and the environmental factor parameters to obtain multiple typical working conditions that need to be calculated;

[0010] S3: constructing a numerical model of the marine structure using the finite element method, and calculating various typical working conditions determined in step S2 using the numerical model of the marine structure to obtain structural response results of the marine structure under various typical working conditions, thereby generating a structural response database of the marine structure under different working conditions;

[0011] S4: constructing a high-precision marine structure proxy model that meets the accuracy requirements according to the structural response database of the marine structure under different working conditions obtained in step S3;

[0012] S5: Inputting structural factor parameters and environmental factor parameters into the high-precision marine structure proxy model, obtaining the marine structure safety status response results under corresponding working conditions, and realizing the structural safety prediction of the marine structure.

[0013] Furthermore, the marine structure in step S1 includes a self-elevating marine platform, a semi-submersible marine platform, a floating wind turbine, etc.; the environmental factor parameters include wave parameters, current parameters and wind parameters; wave parameters include wave height, period, and wave direction; current parameters include flow velocity and flow direction; wind parameters include wind speed and wind direction;

[0014] The structural factor parameters in step S1 include draft and air gap;

[0015] Step S1 also includes setting a reasonable value range for the selected environmental factor parameters and structural factor parameters; determining the upper and lower limits of each parameter according to the design specifications and the operation area to ensure that various situations that the marine structure may encounter during service are covered;

[0016] In step S1, the number of values ​​of each parameter is determined according to the computing power and accuracy requirements of the numerical model of the marine structure.

[0017] Furthermore, the generation of the structural response database of the marine structure under different working conditions in step S3 includes: using the corresponding parameters of various typical working conditions as the input of the numerical model of the marine structure, and using the results obtained by finite element calculation as the output, so that there is an "input-output" pair under each group of working conditions; the sum of the "input-output" pairs under all working conditions constitutes the structural response database of the marine structure under different working conditions.

[0018] Furthermore, step S4 constructs a high-precision marine structure proxy model that meets the accuracy requirements based on the structural response database of the marine structure under different working conditions obtained in step S3; comprising:

[0019] S41: Based on the structural response database of the marine structure under different working conditions, an initial proxy model of the marine structure is constructed according to the principle and construction method of the proxy model;

[0020] S42: After obtaining the initial proxy model of the marine structure, verify whether its accuracy meets the requirements. If not, perform iterative optimization until the accuracy meets the requirements;

[0021] The method for constructing the marine structure proxy model in step S41 includes a linear model, a polynomial response surface model, a Kriging model, a support vector machine model, a radial basis function model, a neural network model or a combined model.

[0022] Further, in step S42, after obtaining the initial proxy model of the marine structure, verify whether its accuracy meets the requirements, and if not, perform iterative optimization until the accuracy meets the requirements; including: using finite element calculations for working conditions not included in the database, and comparing the results with the results obtained using the proxy model, and if the user requirements can be met within the error range, no iterative optimization is required; otherwise, it is considered that the proxy model does not meet the expected accuracy requirements and needs to be optimized;

[0023] The iterative optimization of the proxy model in step S42 includes adjusting the hyperparameters of the proxy model, changing the construction method, or adding sample points.

[0024] Furthermore, the proxy model in step S41 is expressed by formula (1):

[0025]

[0026] Where x = (x1, x2, ..., x n ) is the n-dimensional input variable, i.e., the n selected structural factor parameters and environmental factor parameters; is the output response obtained using the surrogate model; Represents the mapping relationship, which is obtained in the following way:

[0027] For a dataset with a training sample size of N: The corresponding observation data set is obtained by numerical method: Then the surrogate model training dataset can be expressed as [X N×n |Y N×m ];

[0028] The proxy model training data set is trained by the linear model, polynomial response surface model, Kriging model, support vector machine model, radial basis function model, neural network model or combined model to obtain an approximate fitting mapping relationship.

[0029] Furthermore, in step S42, the root mean square error RMSE and the determination coefficient R 2 As indicators for evaluating proxy models:

[0030]

[0031] Among them, y i is the observed value; is the output response obtained using the surrogate model, i.e., the predicted value; is the mean of the predicted values,

[0032] The smaller the root mean square error RMSE, the higher the accuracy of the proxy model; the determination coefficient R 2 The closer it is to 1, the higher the accuracy of the proxy model;

[0033] In the process of building a surrogate model, the root mean square error RMSE or the coefficient of determination R is used. 2 As the objective function, the process of constructing the proxy model is the process of optimizing the value of the objective function;

[0034] After obtaining the best proxy model, the finite element method is used to calculate the working conditions Z=(z1, z2, ..., z n ) T , compare the result y with the result obtained using the surrogate model Compare and evaluate the accuracy of the proxy model. If the evaluation result is good, no iterative optimization is required; otherwise, it is considered that the proxy model does not meet the expected accuracy requirements and needs to be optimized.

[0035] The second aspect of the present invention provides a marine structure safety prediction system based on an agent model, which is used to implement the marine structure safety prediction method based on an agent model, comprising:

[0036] The first main module is used to determine the structural factor parameters and environmental factor parameters that need to be considered according to the classification society rules and the design requirements of marine structures;

[0037] The second main module is used to arrange and combine the various structural factor parameters and the various environmental factor parameters to obtain a variety of typical working conditions that need to be calculated;

[0038] The third main module is used to construct a numerical model of the marine structure using the finite element method, and to calculate the various typical working conditions determined in step S2 respectively through the numerical model of the marine structure to obtain the structural response results of the marine structure under various typical working conditions, thereby generating a structural response database of the marine structure under different working conditions;

[0039] A fourth main module is used to construct a high-precision marine structure proxy model that meets the accuracy requirements according to the structural response database of the marine structure under different working conditions obtained in step S3;

[0040] The fifth main module is used to input structural factor parameters and environmental factor parameters into the high-precision marine structure proxy model, obtain the marine structure safety status response results under corresponding working conditions, and realize the structural safety prediction of the marine structure.

[0041] A third aspect of the present invention provides an electronic device, comprising a processor and a memory, wherein the processor and the memory are connected to each other;

[0042] The memory is used to store computer programs;

[0043] The processor is configured to execute the marine structure safety prediction method based on the agent model when calling the computer program.

[0044] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the marine structure safety prediction method based on the agent model.

[0045] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0046] The invention discloses a method and system for predicting marine structure safety based on a proxy model. According to the classification society's specifications and design requirements, the parameters and value ranges of structural factors and environmental factors affecting the safety of marine structures are determined, and typical working conditions to be calculated are obtained through the arrangement and combination of these parameters. The numerical model of the marine structure is constructed by using the finite element method, and the determined typical working conditions are calculated to obtain the structural response results, thereby generating a structural response database. Based on the structural response database, a suitable proxy model construction method is selected and iterative optimization is performed until the accuracy of the proxy model meets the requirements. The rapid response characteristics of the proxy model are used to combine the real-time sensor data of the marine structure to predict its safety status in real time. The invention comprehensively considers the dual effects of structural factors and environmental factors, and can comprehensively evaluate the safety status of the marine structure. The use of the proxy model simplifies the calculation process and improves the prediction efficiency. The safety status can be predicted in combination with real-time data, which is helpful to take preventive measures in time. The invention can break through the limitations of traditional sensor monitoring, such as the inability of local data to fully reflect the overall safety status and the inability to predict future change trends. Through the proxy model technology, efficient prediction of the overall structural safety of marine structures can be achieved, which is of great significance for improving the safety of marine structures and reducing potential losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A schematic diagram of a flow chart of a method for predicting safety of marine structures based on an agent model according to an embodiment of the present invention;

[0048] Figure 2 A schematic diagram of a process of predicting stress and deformation response of a self-elevating offshore platform during service by a marine structure safety prediction method based on an agent model according to an embodiment of the present invention;

[0049] Figure 3 A schematic diagram of prediction results of a method for predicting safety of marine structures based on a proxy model in an embodiment of the present invention for predicting stress and deformation responses of a jack-up marine platform during service;

[0050] Figure 4 A schematic diagram of the structure of a marine structure safety forecasting system based on an agent model according to an embodiment of the present invention;

[0051] Figure 5 The figure is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0053] Based on the above reasons, if Figure 1 As shown, one aspect of the present invention provides a method for marine structure safety prediction based on an agent model, comprising the following steps:

[0054] S1: Determine the structural factor parameters and environmental factor parameters that need to be considered according to the classification society rules and marine structure design requirements;

[0055] S2: Arrange and combine the structural factor parameters and the environmental factor parameters to obtain multiple typical working conditions that need to be calculated;

[0056] S3: constructing a numerical model of the marine structure using the finite element method, and calculating various typical working conditions determined in step S2 using the numerical model of the marine structure to obtain structural response results of the marine structure under various typical working conditions, thereby generating a structural response database of the marine structure under different working conditions;

[0057] S4: constructing a high-precision marine structure proxy model that meets the accuracy requirements according to the structural response database of the marine structure under different working conditions obtained in step S3;

[0058] S5: Inputting structural factor parameters and environmental factor parameters into the high-precision marine structure proxy model that meets the accuracy requirements, obtaining the marine structure safety status response results under corresponding working conditions, and realizing the structural safety prediction of the marine structure.

[0059] Further, in step S1, when selecting the environmental factor parameters to be considered, the factors affecting the safety of the marine structure are considered; in general, the marine structure will be subjected to wave loads and current loads, so wave parameters, current parameters and draft need to be considered; wherein, wave parameters include wave height, period, wave direction, etc.; current parameters include flow velocity and flow direction; if the marine structure has a large windward area, it will be subjected to a large wind load, and at this time, wind parameters, such as wind speed, wind direction and factors affecting the windward area need to be considered; factors affecting the windward area include air gap, etc.; in a specific embodiment of the present invention, taking the prediction of stress and deformation response of a self-elevating marine platform during service as an example, Figure 2Schematic diagram of the prediction method of the stress and deformation response of the self-elevating offshore platform during its service; in the process of predicting the stress and deformation response of the self-elevating offshore platform during its service, when selecting the environmental factor parameters that need to be considered, starting from the factors affecting the safety of the offshore platform, since the self-elevating offshore platform is subject to wind loads, wind speed and wind direction need to be considered; among them, the wind-exposed area is generally related to the draft and air gap; since the self-elevating offshore platform is subject to wave loads, parameters such as wave height, period, and draft need to be considered; since the self-elevating offshore platform is subject to flow loads, parameters such as flow velocity and flow direction need to be considered.

[0060] Further, in step S1, when selecting the structural factor parameters to be considered, the factors affecting the response of the marine structure to the load are considered; in a specific embodiment of the present invention, taking the prediction of the stress and deformation response of a self-elevating marine platform during service as an example, when selecting the structural factor parameters to be considered, the factors affecting the response of the marine platform to the load are considered. For a self-elevating platform, its platform structure can move up and down along the pile legs. When it moves to different positions, even if the load it bears is the same, the response of the marine platform is different. Therefore, the draft and air gap at the pile leg height below the main deck need to be taken into consideration.

[0061] Furthermore, step S1 also includes setting a reasonable range of values ​​for the selected environmental factor parameters and structural factor parameters; specifically, the upper and lower limits of each parameter are determined according to the design specifications and the operating area to ensure that various situations that the marine structure may encounter during its service are covered; in a specific embodiment of the present invention, taking the prediction of the stress and deformation response of a self-elevating marine platform during its service as an example, the upper and lower limits of each parameter are determined according to the design specifications and the operating area to ensure that various situations that the marine platform may encounter during its service are covered, and for environmental factor parameters, values ​​are taken according to empirical distribution; for example, the specific values ​​of wave height, period, flow velocity, and wind speed are determined according to the wind, wave, and flow environment conditions in the area where the marine platform is to be in service in recent years, and the wave direction, wind direction, and flow direction can be uniformly taken at certain angles; for structural factor parameters, values ​​are uniformly taken at fixed length intervals; for example, for draft and air gap, values ​​can be taken at fixed length intervals.

[0062] Furthermore, in step S1, the value of each parameter is determined according to the computing power that the operating mechanism or unit can mobilize and the requirements for accuracy; when performing numerical simulation or calculation, each parameter (such as wind speed, wave height, flow rate, etc.) needs to take multiple values ​​within a certain range to cover all possible situations; the more values ​​the parameter takes, the more accurate the simulation result is usually, but it also means that more calculations are required; if the computing power is limited, it may not be able to process too many data points, so it is necessary to reasonably select the number of parameter values ​​while ensuring a certain accuracy. On the contrary, if the requirements for forecast accuracy are very high, it may be necessary to increase the number of parameter values, but this will put higher requirements on computing power;

[0063] In a specific embodiment of the present invention, taking the prediction of stress and deformation response of a self-elevating offshore platform during service as an example, 4 values ​​are taken for wave height x1, period x2, and flow velocity x3. Since the self-elevating platform in this embodiment is symmetrical, the wind direction x4, wave direction x5, and flow direction x6 have a value range of 0-180°, with an average interval of 15°. The draft x7 takes 10 values ​​on average, and the air gap x8 takes 4 values ​​on average. The parameters are combined to obtain the working conditions to be calculated. The total number of training samples N = 4×4×4×13×13×10×4 = 432640. 432640 working conditions are obtained.

[0064] Further, step S3 includes establishing a finite element numerical model of the marine structure under each set of structural factor parameters using finite element software (numerical simulation software) based on the marine structure design drawings and specifications and according to the structural factor parameters determined in step S1;

[0065] In step S3, various typical working conditions determined in step S2 are calculated respectively by the numerical model of the marine structure to obtain the structural response results of the marine structure under various typical working conditions, thereby generating a structural response database of the marine structure under different working conditions; including:

[0066] Various typical working conditions determined by the numerical model calculation step S2 of the marine structure are used to obtain finite element calculation results; corresponding parameters of various typical working conditions are used as inputs of the numerical model of the marine structure; the results obtained by finite element calculation are used as outputs, and there are "input-output" pairs under each group of working conditions; the sum of the "input-output" pairs under all working conditions constitutes a structural response database of the marine structure under different working conditions;

[0067] In a specific embodiment of the present invention, finite element software is used to calculate each working condition determined in the second step, and the parameters x = (x1, x2, ..., x8) of each working condition are used as input, and the stress response result y1 and the displacement response result y2 obtained by finite element calculation are used as output. Then, there is an "input-output" pair under each group of working conditions, and the sum of the "input-output" pairs under all working conditions constitutes the offshore platform stress and deformation response database [X 432640×8 |Y 432640×2 ].

[0068] Furthermore, in step S4, a high-precision marine structure proxy model that meets the accuracy requirements is constructed based on the structural response database of the marine structure under different working conditions obtained in step S3; comprising:

[0069] S41: Based on the structural response database of the marine structure under different working conditions, an initial proxy model of the marine structure is constructed according to the principle and construction method of the proxy model;

[0070] S42: After obtaining the initial proxy model of the marine structure, verify whether its accuracy meets the requirements. If not, perform iterative optimization until the accuracy meets the requirements;

[0071] Furthermore, the method for constructing the marine structure proxy model in step S41 includes a linear model, a polynomial response surface model, a Kriging model, a support vector machine model, a radial basis function model, a neural network model or a combined model.

[0072] Furthermore, the proxy model in step S41 is expressed by formula (1):

[0073]

[0074] Where x = (x1, x2, ..., x n ) is the n-dimensional input variable, i.e., the n selected structural factor parameters and environmental factor parameters; is the output response obtained using the surrogate model; Represents the mapping relationship, which is obtained in the following way:

[0075] For a dataset with a training sample size of N: The corresponding observation data set is obtained by numerical method: Then the surrogate model training dataset can be expressed as [X N×n |Y N×m ];

[0076] The proxy model training data set is trained through the linear model, polynomial response surface model, Kriging model, support vector machine model, radial basis function model, neural network model or combined model mentioned above to obtain an approximate fitting mapping relationship.

[0077] Further, in step S42, after obtaining the initial proxy model of the marine structure, verify whether its accuracy meets the requirements, and if not, perform iterative optimization until the accuracy meets the requirements; including: using finite element calculations for working conditions not included in the database, and comparing the results with the results obtained using the proxy model, and if the user requirements can be met within the error range, no iterative optimization is required; otherwise, it is considered that the proxy model does not meet the expected accuracy requirements and needs to be optimized;

[0078] During optimization, the accuracy of the proxy model can be improved by adjusting the hyperparameters of the proxy model, changing the construction method, adding sample points, etc., and continuous optimization and adjustment are performed until the accuracy of the proxy model meets the usage requirements.

[0079] Specifically, in step S42, the root mean square error RMSE and the determination coefficient R 2 As indicators for evaluating proxy models:

[0080]

[0081] Among them, y i is the observed value; is the output response obtained using the surrogate model, i.e., the predicted value; is the mean of the predicted values,

[0082] The smaller the root mean square error RMSE, the higher the accuracy of the proxy model; the determination coefficient R 2 The closer it is to 1, the higher the accuracy of the proxy model. In the process of building a proxy model, the root mean square error RMSE or the coefficient of determination R can be used. 2 As the objective function, the process of constructing the proxy model is the process of optimizing the value of the objective function;

[0083] After obtaining the best proxy model, the finite element method is used to calculate the working conditions Z=(z1, z2, ..., z n ) T , compare the result y with the result obtained using the surrogate model Compare and evaluate the accuracy of the proxy model. If the evaluation result is good, no iterative optimization is required; otherwise, it is considered that the proxy model does not meet the expected accuracy requirements and needs to be optimized; adjust the hyperparameters of the proxy model, change the construction method, expand the training sample size N, etc. to rebuild the proxy model to improve its accuracy. Through continuous optimization and adjustment, the accuracy of the proxy model meets the usage requirements;

[0084] After obtaining a proxy model that meets the requirements, the structural factor parameters and environmental factor parameters are input to predict the structural safety of the marine structure.

[0085] Furthermore, in a specific embodiment of the present invention, since the finite element calculations are all within the linear elastic range, a linear model is selected to be constructed; the offshore platform stress and deformation response database obtained in step S3 is used to establish a proxy model, and the construction process and the final proxy model are implemented by writing Qt C++ code to obtain a program with a visual interface; after obtaining the proxy model, the accuracy is verified until the accuracy meets the requirements.

[0086] The specific process of constructing the proxy model is as follows:

[0087] The proxy model established using the offshore platform stress and deformation response database obtained in step S3 is expressed by equation (4):

[0088]

[0089] Where j is the number of selected structural factor parameters and environmental factor parameters; k j (j=1, 2, ...8) is the feature weight value, b is the intercept of the model;

[0090] In order to obtain the feature weight value k and the model intercept b, let the proxy model parameter matrix

[0091] but:

[0092]

[0093] Using the root mean square error RMSE as the evaluation indicator, the goal is to obtain a set of values ​​that minimizes the RMSE. After obtaining the proxy model, take 10 sets of values ​​that are not in X 432640×8 The working condition set Z 10×8 , Z 10×8 Substituting into the constructed proxy model, we can get the predicted value set Z 10×8 Using finite element calculation, we can get the observation set Y 10×2 ;

[0094] calculate and Y 10×2 Root mean square error RMSE and determination coefficient R of two data sets 2 , if the result is good, it means that the constructed multidimensional linear regression proxy model has high accuracy and can be used to predict the stress response and deformation response of the self-elevating offshore platform during operation.

[0095] Furthermore, step S5 includes calling the high-precision marine structure proxy model that meets the accuracy requirements obtained in step S4 when the marine structure encounters specific sea conditions, and inputting environmental factor parameters and structural factor parameters corresponding to the sea conditions into the proxy model to obtain a response result of the safety status of the marine structure under the corresponding sea conditions; step S5 also includes inputting environmental factor parameters and structural factor parameters corresponding to the sea conditions into the proxy model when the sea conditions that the marine structure may face are known in advance, so as to obtain a response result of the safety status of the marine structure under the corresponding sea conditions, thereby realizing a safety forecast of the marine structure.

[0096] In a specific embodiment of the present invention, when the offshore platform encounters a specific sea condition, the offshore platform proxy model that meets the accuracy requirements obtained in step S4 is called, and the environmental factor parameters and structural factor parameters corresponding to the current sea condition are input to obtain the stress and deformation response of the offshore platform structure under this situation; in addition, if the sea condition that the offshore platform may face is known through weather forecasting or other methods, parameters can also be input to predict the stress and deformation response of the offshore platform; Figure 3 As shown in the figure, it is a schematic diagram of using the proxy model for safety prediction, wherein the left box is the input part, input the structural parameters and environmental parameters considered, click the "Calculate" button, and the stress response and deformation response prediction results of the self-elevating offshore platform under the input sea conditions will be displayed in the right box.

[0097] like Figure 4 As shown, the second aspect of the present invention provides a marine structure safety prediction system based on an agent model, which is used to implement the above prediction method, including:

[0098] The first main module is used to determine the structural factor parameters and environmental factor parameters that need to be considered according to the classification society rules and the design requirements of marine structures, and to set a reasonable value range for each factor parameter;

[0099] The second main module is used to arrange and combine the various structural factor parameters and the various environmental factor parameters to obtain a variety of typical working conditions that need to be calculated;

[0100] The third main module is used to construct a numerical model of the marine structure using the finite element method, and to calculate the various typical working conditions determined in step S2 respectively through the numerical model of the marine structure to obtain the structural response results of the marine structure under various typical working conditions, thereby generating a structural response database of the marine structure under different working conditions;

[0101] A fourth main module is used to construct a high-precision marine structure proxy model that meets the accuracy requirements according to the structural response database of the marine structure under different working conditions obtained in step S3;

[0102] The fifth main module is used to input structural factor parameters and environmental factor parameters into the high-precision marine structure proxy model, obtain the marine structure safety status response results under corresponding working conditions, and realize the structural safety prediction of the marine structure.

[0103] It should be noted that the marine structure safety forecasting system based on the agent model provided in this embodiment can be a computer program (including program code) running in a computer device, for example, the marine structure safety forecasting system based on the agent model is an application software; the marine structure safety forecasting system based on the agent model can be used to execute the corresponding steps in the above method provided in the embodiment of the present application.

[0104] In some feasible implementations, the marine structure safety forecasting system based on the agent model provided in this embodiment can be implemented in a combination of software and hardware. As an example, the marine structure safety forecasting system based on the agent model provided in the embodiment of the present application can be a processor in the form of a hardware decoding processor, which is programmed to execute the marine structure safety forecasting method based on the agent model provided in the embodiment of the present application. For example, the processor in the form of a hardware decoding processor can adopt one or more application specific integrated circuits (ASIC, Application Specific Integrated Circuit), digital signal processors (digital signal processor, DSP), programmable logic devices (PLD, Programmable Logic Device), complex programmable logic devices (CPLD, Complex Programmable Logic Device), field programmable gate arrays (FPGA, Field-Programmable Gate Array) or other electronic components.

[0105] In some feasible implementations, the marine structure safety prediction system based on the agent model provided in this embodiment can be implemented in software form, which can be software in the form of programs and plug-ins, and includes a series of modules to implement the marine structure safety prediction method based on the agent model provided in the embodiment of the present invention.

[0106] The marine structure safety prediction system based on the proxy model provided in this embodiment determines the parameters and value ranges of the structural factors and environmental factors that affect the safety of the marine structure according to the classification society specifications and design requirements, and obtains the typical working conditions that need to be calculated through the arrangement and combination of these parameters; uses the finite element method to construct a numerical model of the marine structure, and calculates the determined typical working conditions to obtain the structural response results, thereby generating a structural response database; based on the structural response database, selects a suitable proxy model construction method and performs iterative optimization until the accuracy of the proxy model meets the requirements; utilizes the rapid response characteristics of the proxy model, combined with the real-time sensor data of the marine structure, and predicts its safety status in real time; the present invention comprehensively considers the dual effects of structural factors and environmental factors, and can comprehensively evaluate the safety status of the marine structure; simplifies the calculation process and improves the prediction efficiency through the use of the proxy model; can predict the safety status in combination with real-time data, which is helpful to take preventive measures in time; the present invention can break through the limitations of traditional sensor monitoring, such as the inability of local data to fully reflect the overall safety status and the inability to predict future change trends. Through the proxy model technology, efficient prediction of the overall structural safety of marine structures can be achieved, which is of great significance for improving the safety of marine structures and reducing potential losses.

[0107] A third aspect of the present invention further provides an electronic device, Figure 5 Schematic diagram of the structure of the electronic device of this embodiment. Figure 5 As shown, the electronic device 1000 in this embodiment may include: a processor 1001, a network interface 1004 and a memory 1005. In addition, the above-mentioned electronic device 1000 may also include: a user interface 1003, and at least one communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and communication between these components. Among them, the user interface 1003 may include a display screen (Display), a keyboard (Keyboard), and the user interface 1003 may optionally include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1004 may be a high-speed RAM memory, or it may be a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 1005 may optionally also be at least one storage device located away from the aforementioned processor 1001. As Figure 5 As shown, the memory 1005 as a computer-readable storage medium may include an operating system, a network communication module, a user interface module, and a device control application program.

[0108] like Figure 5In the electronic device 1000 shown, the network interface 1004 can provide a network communication function; the user interface 1003 is mainly used to provide an input interface for the user; and the processor 1001 can be used to call the device control application stored in the memory 1005 to achieve:

[0109] According to the classification society rules and the design requirements of marine structures, determine the structural factor parameters and environmental factor parameters that need to be considered, and set a reasonable value range for each factor parameter;

[0110] Arrange and combine the various structural factor parameters and the various environmental factor parameters to obtain a variety of typical working conditions that need to be calculated;

[0111] A numerical model of the marine structure is constructed by using the finite element method, and various typical working conditions determined in the previous step are calculated respectively by using the numerical model of the marine structure to obtain the structural response results of the marine structure under various typical working conditions, thereby generating a structural response database of the marine structure under different working conditions;

[0112] Constructing a high-precision marine structure proxy model that meets the accuracy requirements based on the structural response database of the marine structure under different working conditions;

[0113] Structural factor parameters and environmental factor parameters are input into the high-precision marine structure proxy model to obtain the marine structure safety status response results under corresponding working conditions, thereby realizing the structural safety prediction of the marine structure.

[0114] It should be understood that in some feasible implementations, the processor 1001 may be a central processing unit (CPU), or other general-purpose processors, DSPs, ASICs, FPGAs or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0115] In a specific implementation, the electronic device 1000 can execute the above-mentioned functions through its built-in functional modules. Figure 1 For the implementation methods provided in each step, please refer to the implementation methods provided in the above steps for details, which will not be repeated here.

[0116] The electronic device provided in this embodiment determines the parameters and value ranges of structural factors and environmental factors that affect the safety of marine structures according to the classification society specifications and design requirements, and obtains the typical working conditions that need to be calculated through the arrangement and combination of these parameters; uses the finite element method to construct a numerical model of the marine structure, and calculates the determined typical working conditions to obtain the structural response results, thereby generating a structural response database; based on the structural response database, selects a suitable proxy model construction method and performs iterative optimization until the accuracy of the proxy model meets the requirements; utilizes the rapid response characteristics of the proxy model, combined with the real-time sensor data of the marine structure, and predicts its safety status in real time; the present invention comprehensively considers the dual effects of structural factors and environmental factors, and can comprehensively evaluate the safety status of marine structures; simplifies the calculation process through the use of proxy models, and improves the efficiency of prediction; can combine real-time data to predict the safety status, which helps to take preventive measures in time; the present invention can break through the limitations of traditional sensor monitoring, such as local data cannot fully reflect the overall safety status, cannot predict future change trends, etc. Through the proxy model technology, efficient prediction of the overall structural safety of marine structures can be achieved, which is of great significance for improving the safety of marine structures and reducing potential losses.

[0117] The present application also provides a computer-readable storage medium, which stores a computer program and is executed by a processor to implement Figure 1 For the methods provided in each step, please refer to the implementation methods provided in the above steps for details, which will not be repeated here.

[0118] The computer-readable storage medium provided in this embodiment determines the parameters and value ranges of structural factors and environmental factors that affect the safety of marine structures according to the classification society specifications and design requirements, and obtains the typical working conditions that need to be calculated through the arrangement and combination of these parameters; uses the finite element method to construct a numerical model of the marine structure, and calculates the determined typical working conditions to obtain the structural response results, thereby generating a structural response database; based on the structural response database, selects a suitable proxy model construction method and performs iterative optimization until the accuracy of the proxy model meets the requirements; utilizes the rapid response characteristics of the proxy model, combines the real-time sensor data of the marine structure, and predicts its safety status in real time; the present invention comprehensively considers the dual effects of structural factors and environmental factors, and can comprehensively evaluate the safety status of marine structures; simplifies the calculation process and improves the efficiency of prediction through the use of proxy models; can combine real-time data to predict the safety status, which helps to take preventive measures in time; the present invention can break through the limitations of traditional sensor monitoring, such as local data cannot fully reflect the overall safety status, cannot predict future change trends, etc. Through the proxy model technology, efficient prediction of the overall structural safety of marine structures can be achieved, which is of great significance for improving the safety of marine structures and reducing potential losses.

[0119] Any reference to memory, storage, database or other medium used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0120] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting safety of marine structures based on an agent model, characterized in that: The steps include: S1: Determine the structural factor parameters and environmental factor parameters that need to be considered according to the classification society rules and marine structure design requirements; S2: Arrange and combine the structural factor parameters and the environmental factor parameters to obtain a variety of typical working conditions that need to be calculated; S3: constructing a numerical model of the marine structure using the finite element method, and calculating various typical working conditions determined in step S2 using the numerical model of the marine structure to obtain structural response results of the marine structure under various typical working conditions, thereby generating a structural response database of the marine structure under different working conditions; S4: constructing a high-precision marine structure proxy model that meets the accuracy requirements according to the structural response database of the marine structure under different working conditions obtained in step S3; S5: Inputting structural factor parameters and environmental factor parameters into the high-precision marine structure proxy model, obtaining the marine structure safety status response results under corresponding working conditions, and realizing the structural safety prediction of the marine structure.

2. The method for marine structure safety prediction based on agent model according to claim 1, characterized in that: The environmental factor parameters in step S1 include wave parameters, ocean current parameters and wind parameters; wave parameters include wave height, period and wave direction; ocean current parameters include flow velocity and flow direction; wind parameters include wind speed and wind direction; The structural factor parameters in step S1 include draft and air gap. Step S1 also includes setting a reasonable value range for the selected environmental factor parameters and structural factor parameters; determining the upper and lower limits of each parameter according to the design specifications and the operating area to ensure that various situations that the marine structure may encounter during its service are covered.

3. The method for marine structure safety prediction based on agent model according to claim 1, characterized in that: The generation of the structural response database of the marine structure under different working conditions in step S3 includes: using the corresponding parameters of various typical working conditions as the input of the numerical model of the marine structure, and using the results obtained by finite element calculation as the output, so that there is an "input-output" pair under each group of working conditions; the sum of the "input-output" pairs under all working conditions constitutes the structural response database of the marine structure under different working conditions.

4. The method for marine structure safety prediction based on agent model according to claim 1, characterized in that: Step S4: constructing a high-precision marine structure proxy model that meets the accuracy requirements based on the structural response database of the marine structure under different working conditions obtained in step S3; include: S41: Based on the structural response database of the marine structure under different working conditions, an initial proxy model of the marine structure is constructed according to the principle and construction method of the proxy model; S42: After obtaining the initial proxy model of the marine structure, verify whether its accuracy meets the requirements. If not, perform iterative optimization until the accuracy meets the requirements; The method for constructing the marine structure proxy model in step S41 includes a linear model, a polynomial response surface model, a Kriging model, a support vector machine model, a radial basis function model, a neural network model or a combined model.

5. The method for marine structure safety prediction based on agent model according to claim 4, characterized in that: In step S42, after obtaining the initial proxy model of the marine structure, verify whether its accuracy meets the requirements. If not, perform iterative optimization until the accuracy meets the requirements; including: using finite element calculations for working conditions not included in the database, comparing the results with the results obtained using the proxy model, and if the error range can meet the user's requirements, no iterative optimization is required; otherwise, it is considered that the proxy model does not meet the expected accuracy requirements and needs to be optimized; The iterative optimization of the proxy model in step S42 includes adjusting the hyperparameters of the proxy model, changing the construction method, or adding sample points.

6. The method for marine structure safety prediction based on agent model according to claim 4, characterized in that: The proxy model in step S41 is expressed by formula (1): Where x = (x1, x2, ..., x n ) is the n-dimensional input variable, i.e., the n selected structural factor parameters and environmental factor parameters; is the output response obtained using the surrogate model; Represents the mapping relationship, which is obtained in the following way: For a dataset with a training sample size of N: The corresponding observation data set is obtained by numerical method: Then the surrogate model training dataset can be expressed as [X N×n |Y N×m ]; The proxy model training data set is trained by the linear model, polynomial response surface model, Kriging model, support vector machine model, radial basis function model, neural network model or combined model to obtain an approximate fitting mapping relationship.

7. The method for marine structure safety prediction based on agent model according to claim 4, characterized in that: In step S42, the root mean square error RMSE and the determination coefficient R 2 As indicators for evaluating proxy models: Among them, y i is the observed value; is the output response obtained using the surrogate model, i.e., the predicted value; is the mean of the predicted values, The smaller the root mean square error RMSE, the higher the accuracy of the proxy model; the determination coefficient R 2 The closer it is to 1, the higher the accuracy of the proxy model; In the process of building a surrogate model, the root mean square error RMSE or the coefficient of determination R is used. 2 As the objective function, the process of constructing the proxy model is the process of optimizing the value of the objective function; After obtaining the best proxy model, the finite element method is used to calculate the working conditions Z=(z1, z2, ..., z n ) T , compare the result y with the result obtained using the surrogate model Compare and evaluate the accuracy of the proxy model. If the evaluation result is good, no iterative optimization is required; otherwise, it is considered that the proxy model does not meet the expected accuracy requirements and needs to be optimized.

8. A marine structure safety prediction system based on an agent model, characterized in that: A method for predicting marine structure safety based on an agent model as claimed in any one of claims 1 to 7, comprising: The first main module is used to determine the structural factor parameters and environmental factor parameters that need to be considered according to the classification society rules and the design requirements of marine structures, and to set a reasonable value range for each factor parameter; The second main module is used to arrange and combine the various structural factor parameters and the various environmental factor parameters to obtain a variety of typical working conditions that need to be calculated; The third main module is used to construct a numerical model of the marine structure using the finite element method, and to calculate the various typical working conditions determined in step S2 respectively through the numerical model of the marine structure to obtain the structural response results of the marine structure under various typical working conditions, thereby generating a structural response database of the marine structure under different working conditions; A fourth main module is used to construct a high-precision marine structure proxy model that meets the accuracy requirements according to the structural response database of the marine structure under different working conditions obtained in step S3; The fifth main module is used to input structural factor parameters and environmental factor parameters into the high-precision marine structure proxy model, obtain the marine structure safety status response results under corresponding working conditions, and realize the structural safety prediction of the marine structure.

9. An electronic device, characterized in that: comprising a processor and a memory, wherein the processor and the memory are connected to each other; The memory is used to store computer programs; The processor is configured to execute the marine structure safety prediction method based on the agent model as described in any one of claims 1 to 7 when calling the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the marine structure safety prediction method based on an agent model as described in any one of claims 1 to 7.

Images