A method and apparatus for simulating and predicting underwater impact on a cabin
Through machine learning and finite element simulation methods, underwater explosion finite element simulation model is constructed and dynamic simulation is performed to generate the underwater shock simulation prediction model of the tank, which solves the long calculation time and prediction of underwater shock wave simulation in the existing technology. The analysis of difficult problems has achieved accurate prediction of the impact of underwater shock waves on the cabin.
Patent Information
- Application Number
- CN202410709968.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-03
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-06-03
AI Technical Summary
When the prior art simulates the impact of underwater shock waves on surface ships and engineering structures, the calculation time is long and the analysis is difficult to provide accurate simulation predictions.
Using the underwater impact simulation prediction method of the tank based on machine learning and finite element simulation, the underwater explosion finite element simulation model of the water surface simulation chamber is constructed, and the finite element dynamic simulation is performed, the response data is obtained, and the machine learning model is used for training. A tank underwater impact simulation prediction model is generated to identify the maximum impact velocity and acceleration of the target tank subject to underwater impact.
It reduces the difficulty of calculation, shortens the calculation time, provides accurate simulation experiments and prediction results, and improves analysis efficiency.
Smart Images

Figure CN118485022B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of machine simulation analysis, and in particular to a method and device for simulating and predicting underwater impact of a cabin based on machine learning and finite element simulation, and a prediction device. Background Art
[0002] Underwater shock waves are a physical phenomenon caused by explosives (such as TNT) exploding in water. When explosives explode in water, the detonation wave first propagates from the center of the explosion to the explosives. When the detonation wave propagates to the interface between the surface of the charge and the water, the extremely high-speed explosion products spread to the surroundings, forming an initial shock wave. This shock wave will produce strong compression and movement on the surrounding water, the density of the water will increase instantly, and the particle movement speed can reach several thousand meters per second. This shock wave will propagate in the water and exert pressure on the surrounding medium, causing the medium to break.
[0003] The effects of underwater shock waves are mainly manifested in the following aspects:
[0004] 1. Destructive effect: Underwater shock waves have high pressure and speed, which can cause damage to the surrounding medium. This destructive effect can break the medium, thereby achieving engineering purposes, such as underwater blasting excavation, rock breaking, etc.
[0005] 2. Transfer of pressure: Underwater shock waves can transfer pressure, making it act more gently and evenly on the surrounding medium. This pressure transfer effect can make the medium be subjected to uniform pressure, thus avoiding excessive stress concentration in the medium during the explosion process.
[0006] 3. Impact on organisms: Underwater shock waves will also have an impact on organisms. When the density of the animal and the water is close, the shock wave will directly pass through the animal and propagate forward; if there is an air cavity in the animal's body, the shock wave compresses the air and causes damage to the cavity wall, which in turn causes injury or even death to the animal. In addition, underwater shock waves will also interfere with the predation, mating, communication and other behaviors of organisms, and have a certain impact on the marine ecological environment.
[0007] Therefore, when carrying out underwater blasting and other projects, it is necessary to fully consider the role of underwater shock waves and take corresponding safety measures to ensure the safety of the project and the sustainability of the environment. At the same time, it is also necessary to conduct in-depth research on the propagation mechanism and time-frequency characteristics of underwater shock waves to better guide engineering practice.
[0008] In existing application fields, underwater shock waves mainly act in the following aspects:
[0009] Military and Civilian Submarine Design: When designing submarines or other underwater vehicles, understanding the maximum impact velocity under the action of underwater explosions or shock waves is crucial to improving the safety of the vehicle design.
[0010] Protection of marine engineering structures: Offshore platforms, bridge piers and other structures may be threatened by underwater explosions. Predicting the response of these structures under the action of shock waves is of great significance for their design and reinforcement.
[0011] Underwater pipeline and equipment safety: During the design and operation of underwater facilities such as oil and gas pipelines and submarine cables, it is necessary to evaluate the possible damage caused by shock waves to ensure their safe operation.
[0012] Underwater archaeology and salvage operations: During underwater archaeology or salvage operations, explosives may be used, so understanding the impact of shock waves on the surrounding cabin will help protect artifacts and equipment.
[0013] The shock wave generated by underwater explosions propagates underwater and has an impact on surface ships, engineering structures, pipelines and other objects, causing damage to them and internal life forms. Therefore, it is of practical significance to study the impact effect of underwater explosions on surface objects. The parameter usually related to damage is the maximum impact velocity (or maximum impact acceleration). In addition to actual explosion tests, it is common to use finite element simulation technology to conduct simulation experiments. However, although this type of experiment is relatively convenient, there are still problems such as long calculation time and difficulty in analysis. Summary of the invention
[0014] In order to solve the technical problems existing in the prior art, the embodiment of the present invention provides a method and device for simulating and predicting underwater impact of a cabin based on machine learning and finite element simulation, and a prediction device. The technical solution is as follows:
[0015] On the one hand, a method for hull underwater impact simulation prediction based on machine learning and finite element simulation is provided, the method comprising:
[0016] S1. Construct an underwater explosion finite element simulation model Q of a water surface simulation cabin;
[0017] S2. Perform finite element dynamic simulation on the underwater explosion finite element simulation model Q to simulate the impact of the underwater shock wave on the surface simulation cabin, and obtain the response data set T of the surface simulation cabin under the impact load, where:
[0018] T={T(Q),C(x)},
[0019] Where: T(Q) = ∏T{Q(1), Q(2)...Q(x)}; C(x) represents the underwater explosion finite element simulation parameters based on the underwater impact scenario x, T(Q) represents the response data set made by the underwater explosion finite element simulation model Q under the underwater shock wave simulation, and the response data set is composed of the response data accumulated by each sub-model Q(x) in the corresponding underwater impact scenario x, where:
[0020] ∏T{Q(1),Q(2)...Q(x)}=∏{T(Q(1)),T(Q(2))...T(Q(x))};
[0021] T(Q(x)) represents the response data of the sub-model Q(x) under the configured finite element simulation calculation entity unit formula LS-DYNAID(n), where n is the formula address;
[0022] S3, taking the response data set T as input, training and learning the machine learning model, and generating a cabin underwater impact simulation prediction model P;
[0023] S4, labeling training data, performing model optimization iterative training on the cabin underwater impact simulation prediction model P through the labeled training data, and deploying the qualified cabin underwater impact simulation prediction model P on the background server;
[0024] S5. Obtain the structure and performance data of the target cabin and input them into the background server; the background server calls the cabin underwater impact simulation prediction model P to identify and predict the structure and performance data of the target cabin, identify and output the prediction results of the maximum impact velocity and / or acceleration of the corresponding target cabin subjected to underwater impact.
[0025] Optionally, in S1, constructing an underwater explosion finite element simulation model Q of a water surface simulation cabin includes:
[0026] Construct a water surface simulation cabin model;
[0027] Based on the water surface simulation cabin model, an underwater explosion finite element simulation model Q of the water surface simulation cabin model is simulated and constructed;
[0028] The underwater explosion finite element simulation model Q is configured with corresponding underwater explosion finite element simulation parameters C(x), where C(x) includes the following parameters for different underwater impact scenarios x:
[0029] C(x)=(a(x), b(x), c(x))
[0030] Wherein, a represents the explosive equivalent; b represents the explosive position; c represents the structure and performance parameters of the surface simulation cabin; a(x) represents the explosive equivalent under underwater impact scenario x, b(x) represents the explosive position under underwater impact scenario x, and c(x) represents the structure and performance parameters of the surface simulation cabin under underwater impact scenario x;
[0031] By using the finite element simulation application, based on the corresponding underwater explosion finite element simulation parameters C(x) configured by the underwater explosion finite element simulation model Q, the corresponding underwater explosion finite element simulation model Q is generated.
[0032] Optionally, generating a corresponding underwater explosion finite element simulation model Q also includes:
[0033] Randomly generate N underwater explosion finite element simulation models Q according to the a, b and c parameters in C(x):
[0034] Q=polyfit{Q(1), Q(2)...Q(x)},
[0035] Where polyfit{Q(x)} represents the model constructed based on the parameters defined by the model simulation under the underwater impact scenario x.
[0036] Optionally, in S2, finite element dynamic simulation is performed on the underwater explosion finite element simulation model Q, including:
[0037] Mesh each sub-model Q(x) constructed under different underwater impact scenarios x to obtain a discrete unit model of each sub-model Q(x);
[0038] According to the model attributes of each discrete unit model, select the finite element simulation calculation solid unit formula LS-DYNA ID (n) that matches the model attributes from ANSYS LS-DYNA;
[0039] Bind the finite element simulation calculation entity unit formula LS-DYNA ID(n) to the corresponding discrete unit model.
[0040] Optionally, the finite element simulation calculation entity unit formula LS-DYNA ID(n) is used to calculate the response data of the corresponding discrete unit model under the underwater impact scenario x.
[0041] Optionally, in S4, the training data is labeled, and the cabin underwater impact simulation prediction model P is subjected to model optimization iterative training through the labeled training data, and the qualified cabin underwater impact simulation prediction model P is deployed on the background server, including:
[0042] Preset model iteration stop conditions;
[0043] Construct labeled training data for several underwater impact scenarios x;
[0044] The labeled training data is used to conduct supervised training learning on the cabin underwater impact simulation prediction model P until the preset model iteration stop condition is reached, and a qualified cabin underwater impact simulation prediction model P is obtained;
[0045] The qualified cabin underwater impact simulation prediction model P is deployed on the background server for cabin underwater impact simulation prediction.
[0046] On the other hand, a device for simulating and predicting underwater impact of a cabin based on machine learning and finite element simulation is provided, and the device is used to implement a method for simulating and predicting underwater impact of a cabin based on machine learning and finite element simulation, and the device comprises:
[0047] A data acquisition unit is used to obtain the structure and performance data of the target cabin and input it into the backend server;
[0048] The backend server is used to call the underwater impact simulation prediction model P of the cabin to identify and predict the structure and performance data of the target cabin, match and output the prediction results of the maximum impact velocity and / or acceleration of the target cabin subjected to underwater impact and output them;
[0049] The data acquisition unit is connected to the background server for communication.
[0050] On the other hand, a prediction device is provided, comprising: a processor; a memory, wherein computer-readable instructions are stored in the memory, and when the computer-readable instructions are executed by the processor, any one of the above-mentioned methods for predicting underwater impact simulation of a cabin based on machine learning and finite element simulation is implemented.
[0051] On the other hand, a computer-readable storage medium is provided, in which at least one instruction is stored, and the at least one instruction is loaded and executed by a processor to implement any one of the above-mentioned methods for hull underwater impact simulation prediction based on machine learning and finite element simulation.
[0052] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:
[0053] The present invention constructs an underwater explosion finite element simulation model Q of a surface simulation cabin and performs finite element dynamic simulation to simulate the impact of underwater shock waves on the surface simulation cabin, and obtains response data of the surface simulation cabin under the impact load. The response data is then used as input and a machine learning model is used to train, learn and identify the response data to generate a cabin underwater impact simulation prediction model P, which is used to identify and predict the structure and performance data of the target cabin, and identify and output the predicted results of the maximum impact velocity and / or acceleration of the target cabin subjected to underwater impact.
[0054] Therefore, it can provide accurate simulation experiments and predictions for surface ships, engineering structures, pipelines, etc., reduce the difficulty of calculation, greatly shorten the calculation time by combining with AI models, and use finite element simulation to reduce the difficulty of analysis and improve efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0056] Figure 1 It is a flow chart of a method for simulating and predicting underwater impact of a cabin based on machine learning and finite element simulation provided by an embodiment of the present invention;
[0057] Figure 2 It is a three-dimensional structural schematic diagram of an underwater explosion finite element simulation model provided by an embodiment of the present invention;
[0058] Figure 3 It is a structural block diagram of a cabin underwater impact simulation prediction device based on machine learning and finite element simulation provided by an embodiment of the present invention;
[0059] Figure 4 It is a structural schematic diagram of a prediction device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0060] The technical solution of the present invention is described below in conjunction with the accompanying drawings.
[0061] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.
[0062] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.
[0063] In the embodiments of the present invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are consistent.
[0064] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.
[0065] An embodiment of the present invention provides a method for hull underwater impact simulation prediction based on machine learning and finite element simulation. The method can be implemented by a prediction device, which can be a terminal or a server.
[0066] like Figure 1 The flowchart of the method for hull underwater impact simulation prediction based on machine learning and finite element simulation is shown. The processing flow of the method may include the following steps:
[0067] S1. Construct an underwater explosion finite element simulation model Q of a water surface simulation cabin;
[0068] S2. Perform finite element dynamic simulation on the underwater explosion finite element simulation model Q to simulate the impact of the underwater shock wave on the surface simulation cabin, and obtain the response data set T of the surface simulation cabin under the impact load, where:
[0069] T={T(Q),C(x)},
[0070] Where: T(Q) = ∏T{Q(1), Q(2)...Q(x)}; C(x) represents the underwater explosion finite element simulation parameters based on the underwater impact scenario x, T(Q) represents the response data set made by the underwater explosion finite element simulation model Q under the underwater shock wave simulation, and the response data set is composed of the response data accumulated by each sub-model Q(x) in the corresponding underwater impact scenario x, where:
[0071] ∏T{Q(1),Q(2)...Q(x)}=∏{T(Q(1)),T(Q(2))...T(Q(x))};
[0072] T(Q(x)) represents the response data of the sub-model Q(x) under the configured finite element simulation calculation entity unit formula LS-DYNAID(n), where n is the formula address;
[0073] S3, taking the response data set T as input, training and learning the machine learning model, and generating a cabin underwater impact simulation prediction model P;
[0074] S4, labeling training data, performing model optimization iterative training on the cabin underwater impact simulation prediction model P through the labeled training data, and deploying the qualified cabin underwater impact simulation prediction model P on the background server;
[0075] S5. Obtain the structure and performance data of the target cabin and input them into the background server; the background server calls the cabin underwater impact simulation prediction model P to identify and predict the structure and performance data of the target cabin, identify and output the prediction results of the maximum impact velocity and / or acceleration of the corresponding target cabin subjected to underwater impact.
[0076] This scheme mainly constructs an underwater explosion finite element simulation model Q of the surface simulation cabin and performs finite element dynamic simulation to simulate the impact of underwater shock waves on the surface simulation cabin, and obtains the response data of the surface simulation cabin under the impact load. Then, the response data is used as input and the machine learning model is used to train, learn and identify the response data to generate an underwater impact simulation prediction model P for the cabin, which is used to identify and predict the structure and performance data of the target cabin, and identify and output the prediction results of the maximum impact velocity and / or acceleration corresponding to the target cabin subjected to underwater impact.
[0077] The implementation process and principle of this solution will be described in detail from steps S1 to S5.
[0078] Optionally, S1, constructing an underwater explosion finite element simulation model Q of a water surface simulation cabin, comprising:
[0079] Construct a water surface simulation cabin model;
[0080] Based on the water surface simulation cabin model, an underwater explosion finite element simulation model Q of the water surface simulation cabin model is simulated and constructed;
[0081] The corresponding underwater explosion finite element simulation parameters C(x) are configured for the underwater explosion finite element simulation model Q. C(x) includes the following parameters under different x:
[0082] C(x)=(a(x), b(x), c(x)),
[0083] a, explosive equivalent;
[0084] b, location of the explosive;
[0085] c. Structure and performance parameters of the surface simulation cabin;
[0086] a(x), b(x), c(x) are the explosive equivalent, explosive position, and structural and performance parameters of the surface simulation cabin at x, respectively;
[0087] By using the finite element simulation application, based on the corresponding underwater explosion finite element simulation parameters C(x) configured for the underwater explosion finite element simulation model Q, a corresponding underwater explosion finite element simulation model Q is generated.
[0088] like Figure 2 A finite element simulation model of underwater explosion is shown (actually a simulation model under the xyz three-dimensional coordinate axis, and the boundary of the simulation model is not shown). In this embodiment, a stiffening plate model of a water surface simulation cabin is used to simulate the impact force of underwater explosion. A finite element simulation analysis three-dimensional coordinate system with the center of the stiffening plate as the origin can be established, so that the position of each point (object) in the model can be obtained, including the position of the explosion point.
[0089] At a certain distance below the origin in the model, that is, on the -z axis, explosives (TNT, trinitrotoluene) are simulated. Subsequently, through finite element simulation of each model, the shock wave effect of the explosives on the bottom of the water simulation cabin is simulated, and the maximum impact velocity and impact acceleration of the water simulation cabin are simulated and calculated.
[0090] The underwater explosion finite element simulation model Q can be constructed by finite element simulation. The construction steps can be as follows:
[0091] The construction of a finite element simulation model is a complex but structured process. It is mainly based on the principle of the finite element method, which divides a continuous physical system into a finite number of small elements (i.e., finite elements), and then converts the continuous problem into a discrete problem for calculation and solution. The following are the main steps to build a finite element simulation model:
[0092] 1. Problem definition:
[0093] Clarify the purpose of the simulation, the physical phenomena that need to be simulated, and the relevant boundary conditions (for underwater explosion shock wave simulation, the finite element simulation data is determined according to this scheme C(x)).
[0094] Determine the required accuracy and scope of the model.
[0095] 2. Model building:
[0096] Create a geometric model: Based on the problem description, build the geometric shape that represents the physical system.
[0097] Meshing: Discretize the geometric model into a series of finite element units. The unit type can be selected as needed, such as line units, surface units, shell units, beam units and solid units.
[0098] Define material properties: Assign corresponding material properties to each finite element, such as elastic modulus, density, Poisson's ratio, etc.
[0099] 3. Solver selection:
[0100] According to the nature of the problem, select a suitable finite element solver, such as finite element method, finite difference method, etc.
[0101] 4. Boundary conditions and load application:
[0102] According to the actual problem, appropriate boundary conditions are imposed on the model, such as fixation, constraints, etc.
[0103] Apply external loads such as force, pressure, temperature, etc. to the model.
[0104] 5. Solution:
[0105] Run the solver to calculate the response values of each finite element, such as displacement, stress, strain, etc.
[0106] 6. Post-processing:
[0107] Visualize the calculation results, such as drawing displacement cloud diagrams, stress distribution diagrams, etc.
[0108] Conduct statistical analysis and optimization analysis to extract key information and evaluate the effectiveness and accuracy of the model.
[0109] Combined with Figure 2 As shown, the main finite element model of this scheme is established: a structural model of the water surface simulation cabin is established, including material properties, geometric shapes and boundary conditions, etc., to ensure that the model can accurately reflect the actual situation.
[0110] Based on machine learning, the experimental parameters are used for training and simulation prediction to determine the impact force, maximum impact acceleration, and maximum impact velocity on surface compartment A and underwater pipeline B under different TNT equivalents, explosion source distances, and locations.
[0111] Here are the steps:
[0112] Establish a finite element simulation model of underwater explosion.
[0113] Set the explosion parameters a, explosive (TNT) equivalent (e.g. 1kg to 10kg); b, TNT position (1 to 10 meters underwater); c, surface simulation cabin structure and performance parameters (such as the thickness, area, material properties (involving different strain limit simulations), density, etc. of the surface simulation cabin);
[0114] Finite element simulation analysis is used to calculate the maximum impact velocity and / or acceleration suffered by the simulated water surface cabin (scaled to 1m and 2m in length and width).
[0115] Dynamic simulation analysis: Use finite element software to perform dynamic simulation, simulate the process of underwater shock waves on the cabin, and obtain the response data of the cabin under the impact load. LS-DYNA ID(n), the solid unit formula selected in ANSYS LS-DYNA: ID(n), is used to calculate and analyze the large deformation static and dynamic response of the structure.
[0116] a(x), b(x), c(x) are the explosive equivalent, explosive position, and structure and performance parameters of the surface simulation cabin at x. The specific parameters can be set by the administrator.
[0117] In this scheme, in order to obtain more training data, the numerical combinations of a, b and c parameters (for example, parameter a can be divided into several equivalent levels according to the arithmetic interval of 0.05, so that several groups of C(x) can be obtained) can be used to construct several underwater explosion finite element simulation models Q. The x variable can be set according to the arithmetic interval.
[0118] Optionally, when generating the corresponding underwater explosion finite element simulation model Q, it also includes:
[0119] Randomly generate N underwater explosion finite element simulation models Q according to the a, b and c parameters in C(x), namely:
[0120] Q=polyfit{Q(1), Q(2)...Q(x)},
[0121] x represents the underwater impact scenario,
[0122] polyfit{Q(x)} represents the model constructed based on the parameters defined by the model simulation under the underwater impact scenario x.
[0123] In order to obtain as much finite element simulation data of the model as possible and improve the accuracy of the training sample set (the more data, the better the model training effect and the more accurate it is), this scheme randomly generates 500 underwater explosion finite element simulation models according to the a, b and c parameters.
[0124] For each set of underwater explosion finite element simulation parameters C(x), an underwater explosion finite element simulation model Q can be constructed. For example, C(1) corresponds to Q(1), and 500 underwater explosion finite element simulation models Q can be randomly generated.
[0125] Here, polyfit{Q(1), Q(2)...Q(x)} means that the model Q of this scheme is an aggregate of 500 models. Model Q contains 500 variable models generated by different C(x). It also means that this scheme can obtain 500 sets of data generated by explosion impact (response data set T) through finite element simulation.
[0126] S2. Perform finite element dynamic simulation on the underwater explosion finite element simulation model Q to simulate the impact of the underwater shock wave on the surface simulation cabin, and obtain a response data set T of the surface simulation cabin under the impact load, wherein:
[0127] T={T(Q),C(x)},
[0128] Where: T(Q) = ∏T{Q(1), Q(2)...Q(x)}; C(x) represents the underwater explosion finite element simulation parameters based on the underwater impact scenario x, T(Q) represents the response data set made by the underwater explosion finite element simulation model Q under the underwater shock wave simulation, and the response data set is composed of the response data accumulated by each sub-model Q(x) in the corresponding underwater impact scenario x, wherein:
[0129] ∏T{Q(1),Q(2)...Q(x)}=∏{T(Q(1)),T(Q(2))...T(Q(x))};
[0130] T(Q(x)) represents the response data of the sub-model Q(x) under the configured finite element simulation calculation entity unit formula LS-DYNA ID(n), where n is the formula address;
[0131] Dynamic simulation analysis: Use finite element software such as ANSYS to perform dynamic simulation, simulate the effect of underwater shock waves on the cabin, and obtain the response data of the cabin under impact load.
[0132] LS-DYNA ID(n), the solid element formulation selected in ANSYS LS-DYNA: ID(n), is used to calculate and analyze the large deformation static and dynamic responses of the structure.
[0133] In LS-DYNA, the relevant formulas of solid elements involve multiple aspects, including but not limited to the calculation of volume, mass, kinetic energy and internal energy.
[0134] 1. The volume calculation formula is: Volume = Area × Thickness. In LS-DYNA, the volume of a solid element is usually obtained by calculating its surface area and multiplying it by its thickness.
[0135] 2. The mass calculation formula is: mass = density × volume. The mass calculation formula of a solid unit is related to its density and volume.
[0136] 3. The calculation formula of kinetic energy is: kinetic energy = 0.5 × mass × velocity squared. The kinetic energy of a solid unit can be calculated by its mass and velocity squared.
[0137] 4. The calculation formula of internal energy is: internal energy = specific heat capacity × mass × temperature change. The calculation formula of the internal energy of a solid unit involves its specific heat capacity, mass and temperature change.
[0138] In addition, for hexahedral solid elements and tetrahedral solid elements, LS-DYNA also provides different unit formulas, such as single-point synchronous rotation elements, constant stress elements, selective reduced integration solid elements, full integration elements, etc. These unit formulas have different characteristics and application scenarios.
[0139] Therefore, when actually performing finite element simulation calculations on each sub-model Q(x) in this solution, the corresponding entity unit formula required for the finite element calculation can be selected according to the attribute parameters of each sub-model Q(x).
[0140] Optionally, in step S2, when performing finite element dynamic simulation on the underwater explosion finite element simulation model Q, it also includes:
[0141] Meshing is performed on each sub-model Q(x) constructed under different underwater impact scenarios in the underwater explosion finite element simulation model Q to obtain a discrete unit model of each sub-model Q(x);
[0142] According to the model attributes of each discrete unit model, select the finite element simulation calculation solid unit formula LS-DYNA ID (n) that matches the model attributes from ANSYS LS-DYNA;
[0143] Bind the finite element simulation calculation entity unit formula LS-DYNA ID(n) to the corresponding discrete unit model.
[0144] In this embodiment, the selection of specific formulas and parameters matched by each sub-model Q(x) is determined by the administrator based on factors such as the characteristics of the model, the purpose of the analysis, and the limitation of computing resources.
[0145] Under the underwater shock wave simulation, the underwater explosion finite element simulation model Q can record the response data of the simulation made by the finite element system.
[0146] Therefore, each sub-model Q(x) will produce corresponding response data, namely T(Q(x)) (including maximum impact velocity, acceleration, etc.) in the corresponding underwater impact scene x under underwater shock wave simulation. The final response data can be determined based on the calculation results.
[0147] T(Q(x)) contains the response data of the sub-model Q(x) under the configured finite element simulation calculation entity unit formula LS-DYNA ID(n), where n is the formula address. During the finite element simulation calculation, the corresponding configured entity unit formula will be selected from ANSYS LS-DYNA according to the formula address n configured for the sub-model Q(x) to calculate the explosion parameters of the current sub-model Q(x) and obtain the corresponding response data T(Q(x)). The finite element simulation calculation entity unit formula LS-DYNA ID(n) is used to calculate the response data of the corresponding discrete unit model under the underwater impact scenario x.
[0148] Because there are several sub-models, it is necessary to calculate the set of response data made by all sub-models Q(x) under underwater shock wave simulation. This set is composed of the cumulative response data made by each sub-model Q(x) in the corresponding underwater shock scene x, where:
[0149] T(Q)=∏T{Q(1),Q(2)......Q(x)}=∏{T(Q(1)),T(Q(2))......T(Q(x))}, where ∏ represents the cumulative data of the response data of each sub-model, indicating that the finite element simulation data collected this time is the cumulative set of response data of several sub-models.
[0150] In this scheme, the response data of the above-mentioned underwater explosion finite element simulation model will be used as training data, and the corresponding prediction model will be constructed using the machine learning model.
[0151] S3, taking the response data set T as input, using the machine learning model to train, learn and identify the response data set T, and generating a cabin underwater impact simulation prediction model P;
[0152] Specifically, machine learning model training: collect simulation data as a training set, select a suitable machine learning algorithm (such as neural network, support vector machine, random forest, etc.) to train the model so that it can learn the maximum impact velocity of the cabin under different conditions.
[0153] Use the trained machine learning model to predict the maximum impact velocity for new designs or conditions and verify the accuracy of the model by comparing it with additional simulation or experimental data.
[0154] Based on the prediction results and verification feedback, the simulation model and machine learning algorithm are continuously adjusted to improve the accuracy and reliability of the prediction.
[0155] The response data set T includes T(Q) and C(x), where C(x) represents the underwater explosion finite element simulation parameters based on the underwater impact scenario x, and T(Q) represents the response data set of the underwater explosion finite element simulation model Q under underwater shock wave simulation.
[0156] Therefore, the underwater explosion finite element simulation parameters corresponding to the underwater impact scenario x have corresponding underwater shock wave simulation response data, that is, under the underwater explosion finite element simulation parameters of the underwater impact scenario x, there are corresponding underwater impact maximum velocity and impact acceleration data. Through several groups of response data sets T, a training data set is formed and model learning training is performed, so that the model learns the data characteristic relationship between the underwater explosion finite element simulation parameters of different underwater impact scenarios x and the corresponding response data (underwater impact maximum velocity and impact acceleration data), so that the model can predict and output the corresponding underwater shock wave response data to the cabin according to the underwater explosion finite element simulation parameters, and quickly obtain the maximum impact velocity of the simulated cabin under the action of the underwater shock wave.
[0157] The steps of using a machine learning model to train the finite element data (the response data set T in this embodiment) generally involve the following key steps:
[0158] 1. Data preparation:
[0159] Collect finite element data: This includes various types of data obtained from finite element analysis, such as stress, strain, displacement, etc.
[0160] Data preprocessing: Cleaning and organizing data, including missing value processing, outlier processing, data standardization or normalization, etc., so that machine learning algorithms can learn and generalize better.
[0161] 2. Feature Engineering:
[0162] Feature extraction: Extract meaningful features from raw finite element data. These features should be able to reflect the inherent laws and patterns of the data.
[0163] Feature selection: Select the most representative feature subset based on the requirements of the machine learning task to improve the performance of the model.
[0164] 3. Model selection and construction:
[0165] Choose the appropriate machine learning algorithm: According to the nature of the problem and the characteristics of the data, choose the appropriate machine learning algorithm, such as linear regression, decision tree, neural network, etc.
[0166] Build model: Build a machine learning model using the selected algorithm and extracted features.
[0167] 4. Model training:
[0168] Set training parameters: including learning rate, number of iterations, regularization parameters, etc. These parameters will affect the training process and performance of the model.
[0169] Training model: Use the preprocessed finite element data and extracted features to train the model, and continuously adjust the model parameters through the optimization algorithm to minimize the prediction error.
[0170] 5. Model evaluation and optimization:
[0171] Evaluate model performance: Use test sets or cross-validation methods to evaluate model performance, such as accuracy, recall, F1 value, etc.
[0172] Model optimization: Adjust and optimize the model based on the evaluation results, such as adjusting model parameters, changing algorithms, adding features, etc., to improve the performance of the model.
[0173] 6. Model deployment and application:
[0174] Model deployment: Deploy the trained model to actual application scenarios to predict and analyze new finite element data.
[0175] Model monitoring and maintenance: Continuously monitor and maintain the deployed model to ensure its performance and stability.
[0176] Please refer to the above steps for training to obtain a cabin underwater impact simulation prediction model P.
[0177] In order to improve the recognition accuracy of the cabin underwater impact simulation prediction model P, this embodiment also performs enhanced training.
[0178] Optionally, S4, using the labeled training data, performing model optimization iterative training on the cabin underwater impact simulation prediction model P, and deploying the qualified cabin underwater impact simulation prediction model P on the background server, including:
[0179] Preset model iteration stop conditions;
[0180] Construct labeled training data for several underwater impact scenarios x;
[0181] The labeled training data is used to conduct supervised training learning on the cabin underwater impact simulation prediction model P until the preset model iteration stop condition is reached, and a qualified cabin underwater impact simulation prediction model P is obtained;
[0182] The qualified cabin underwater impact simulation prediction model P is deployed on the background server for cabin underwater impact simulation prediction.
[0183] According to the data structure of the response data set T, the administrator can construct several groups of labeled training data of corresponding data structures based on experiments, so as to conduct targeted training on the cabin underwater impact simulation prediction model P and improve the recognition accuracy of the cabin underwater impact simulation prediction model P.
[0184] Supervised training learning can be performed according to the training mode selected above until the model reaches the model iteration stop condition set by the administrator, such as the recognition accuracy of the cabin underwater impact simulation prediction model P on the verification set reaches 90%.
[0185] Supervised training is a basic training method in machine learning. Its core is to use a data set with known labels or results to train the model. In this process, the model will try to learn the mapping relationship from input data to corresponding labels or results.
[0186] In supervised training, the dataset is usually divided into a training set and a test set. The training set is used to train the model, while the test set is used to evaluate the model's performance. During the training process, the model continuously adjusts its internal parameters based on the data in the training set to minimize the difference between the predicted results and the true labels. This difference is usually measured by a loss function, and the goal of the model is to minimize the value of the loss function.
[0187] Once the model achieves satisfactory performance on the training set, the test set can be used to evaluate its generalization ability, that is, the model's ability to predict new data. If the model performs well on the test set, then it can be considered that the model has successfully learned the laws and patterns in the data and can be used in actual application scenarios.
[0188] Supervised training has a wide range of applications in many fields, such as image recognition, speech recognition, natural language processing, etc. Through supervised training, we can train models that can accurately recognize images, understand speech, or generate text, providing intelligent support for various applications.
[0189] This solution performs supervised training on the underwater impact simulation prediction model P of the cabin according to the labeled training data, so as to improve the prediction ability of the model. The specific operation can be combined with the supervised training model.
[0190] S5. Obtain the structure and performance data of the target cabin and input them into the backend server. The backend server calls the cabin underwater impact simulation prediction model P to identify and predict the structure and performance data of the target cabin, identify and output the prediction results of the maximum impact velocity and / or acceleration of the corresponding target cabin subjected to underwater impact.
[0191] The constructed cabin underwater impact simulation prediction model P can be deployed on a background server to predict the maximum impact velocity and / or acceleration of the target cabin.
[0192] The target cabin, i.e., the cabin to be subjected to the underwater explosion experiment, can be designed by researchers, and the structural and performance data (c) of the target cabin can be input into the background server, and the corresponding finite element simulation parameters (a and b) can be configured. The cabin underwater impact simulation prediction model P can identify the data characteristics of the structural and performance data of the target cabin, and identify and output the prediction results of the maximum impact velocity and / or acceleration that match the data characteristics of the structural and performance data of the target cabin according to the relationship between the identified response data characteristics and the data characteristics of the structural and performance data. The data acquisition method can be to transmit the data to the background server through the USB interface of the R&D design platform, such as the finite element design platform.
[0193] Therefore, this solution uses machine learning technology to predict the maximum impact velocity and / or acceleration suffered by the surface simulation cabin, and uses finite element simulation analysis to calculate the maximum impact velocity of the surface simulation cabin under the shock wave generated by the underwater TNT explosion, thereby improving the efficiency of experiments and cabin design.
[0194] On the other hand, a cabin underwater impact simulation and prediction device based on machine learning and finite element simulation is provided.
[0195] Figure 3 The present invention is a device block diagram of a device for simulating and predicting underwater impact of a cabin body based on machine learning and finite element simulation according to an exemplary embodiment. The device is used in a method for simulating and predicting underwater impact of a cabin body based on machine learning and finite element simulation. Figure 3 The device includes a data acquisition unit 310 and a background server 320. Wherein:
[0196] The data acquisition unit 310 is used to obtain the structure and performance data of the target cabin and input it into the backend server;
[0197] The backend server 320 is used to call the underwater impact simulation prediction model P of the hull to identify and predict the structure and performance data of the target hull, match and output the prediction results of the maximum impact velocity and / or acceleration of the target hull subjected to underwater impact and output them;
[0198] The data acquisition unit is in communication connection with the background server.
[0199] The functions and interactions of the data acquisition unit 310 and the background server 320 can be understood in conjunction with the above-mentioned “cabin underwater impact simulation prediction method based on machine learning and finite element simulation” and will not be elaborated in this embodiment.
[0200] Figure 4 is a schematic diagram of a structure of a prediction device provided by an embodiment of the present invention, such as Figure 4 As shown, the prediction device may include the above Figure 3 The device for predicting underwater impact simulation of a cabin based on machine learning and finite element simulation is shown. Optionally, the prediction device 410 may include a first processor 2001 .
[0201] Optionally, the prediction device 410 may further include a memory 2002 and a transceiver 2003 .
[0202] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.
[0203] Combine the following Figure 4 The components of the prediction device 410 are described in detail:
[0204] The first processor 2001 is the control center of the prediction device 410, and may be a processor or a general term for multiple processing elements. For example, the first processor 2001 is one or more central processing units (CPUs), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (FPGAs).
[0205] Optionally, the first processor 2001 may perform various functions of the prediction device 410 by running or executing a software program stored in the memory 2002 and calling data stored in the memory 2002 .
[0206] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 4 CPU0 and CPU1 are shown in FIG.
[0207] In a specific implementation, as an embodiment, the prediction device 410 may also include multiple processors, such as Figure 4 The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).
[0208] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled to be executed by the first processor 2001. The specific implementation method can refer to the above method embodiment, which will not be repeated here.
[0209] Optionally, the memory 2002 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently and access the first processor 2001 through an interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.
[0210] The transceiver 2003 is used to communicate with a network device or a terminal device.
[0211] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 4 The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.
[0212] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently, and may be connected to the first processor 2001 through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.
[0213] It should be noted that Figure 4 The structure of the prediction device 410 shown in the figure does not constitute a limitation on the router, and the actual knowledge structure recognition device may include more or fewer components than those shown in the figure, or combine certain components, or arrange the components differently.
[0214] In addition, the technical effects of the prediction device 410 can refer to the technical effects of the cabin underwater impact simulation prediction method based on machine learning and finite element simulation described in the above method embodiment, which will not be repeated here.
[0215] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0216] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).
[0217] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0218] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.
[0219] In the present invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.
[0220] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0221] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0222] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0223] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0224] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0225] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0226] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0227] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A method for simulating and predicting underwater impact of a cabin, characterized in that: The method comprises: S1. Constructing the underwater explosion finite element simulation model Q of the water surface simulation cabin; including: Construct a water surface simulation cabin model; Based on the water surface simulation cabin model, constructing an underwater explosion finite element simulation model Q of the water surface simulation cabin model by simulation; The underwater explosion finite element simulation model Q is configured with corresponding underwater explosion finite element simulation parameters C(x), where C(x) includes the following parameters for different underwater impact scenarios x: C(x)=(a(x), b(x), c(x)) Wherein, a represents the explosive equivalent; b represents the explosive position; c represents the structure and performance parameters of the surface simulation cabin; a(x) represents the explosive equivalent under underwater impact scenario x, b(x) represents the explosive position under underwater impact scenario x, and c(x) represents the structure and performance parameters of the surface simulation cabin under underwater impact scenario x; Using a finite element simulation application, based on the corresponding underwater explosion finite element simulation parameters C(x) configured by the underwater explosion finite element simulation model Q, a corresponding underwater explosion finite element simulation model Q is generated; S2. Perform finite element dynamic simulation on the underwater explosion finite element simulation model Q to simulate the impact of the underwater shock wave on the surface simulation cabin, and obtain a response data set T of the surface simulation cabin under the impact load, wherein: T={T(Q),C(x)}, Where: T(Q) = ΠT{Q(1), Q(2) ... Q(x)}; C(x) represents the underwater explosion finite element simulation parameters based on the underwater impact scenario x, T(Q) represents the response data set made by the underwater explosion finite element simulation model Q under the underwater shock wave simulation, and the response data set is composed of the response data accumulated by each sub-model Q(x) in the corresponding underwater impact scenario x, wherein: ∏T{Q(1),Q(2)...Q(x)}=Π{T(Q(1)),T(Q(2))...T(Q(x))}; T(Q(x)) represents the response data of the sub-model Q(x) under the configured finite element simulation calculation entity unit formula LS-DYNAID(n), where n is the formula address; S3, using the response data set T as input, training, learning and identifying the machine learning model to generate a cabin underwater impact simulation prediction model P; S4, labeling training data, performing model optimization iterative training on the cabin underwater impact simulation prediction model P through the labeled training data, and deploying the qualified cabin underwater impact simulation prediction model P on the background server; S5. Obtain the structure and performance data of the target cabin and input them into the background server; the background server calls the cabin underwater impact simulation prediction model P to identify and predict the structure and performance data of the target cabin, identify and output the prediction result of the maximum impact velocity and / or acceleration corresponding to the underwater impact of the target cabin.
2. The method for simulating and predicting underwater impact of a cabin according to claim 1, characterized in that: Generating the corresponding underwater explosion finite element simulation model Q also includes: Randomly generate N underwater explosion finite element simulation models Q according to the a, b and c parameters in C(x): Q=polyfit{Q(1), Q(2)...Q(x)}, Where polyfit{Q(x)} represents the model constructed based on the parameters defined by the model simulation under the underwater impact scenario x.
3. The method for simulating and predicting underwater impact of a cabin according to claim 2, characterized in that: In step S2, a finite element dynamic simulation is performed on the underwater explosion finite element simulation model Q, including: Perform grid division on each sub-model Q(x) constructed under different underwater impact scenarios x to obtain a discrete unit model of each sub-model Q(x); According to the model attributes of each discrete unit model, a finite element simulation calculation entity unit formula LS-DYNA ID (n) matching the model attributes is selected from ANSYS LS-DYNA; The finite element simulation calculation entity unit formula LS-DYNA ID (n) is bound to the corresponding discrete unit model.
4. The method for simulating and predicting underwater impact of a cabin according to claim 3, characterized in that: The finite element simulation calculation entity unit formula LS-DYNA ID(n) is used to calculate the response data corresponding to the discrete unit model under the underwater impact scenario x.
5. The method for simulating and predicting underwater impact of a cabin according to claim 4, characterized in that: In S4, the training data is labeled, and the cabin underwater impact simulation prediction model P is subjected to model optimization iterative training through the labeled training data, and the qualified cabin underwater impact simulation prediction model P is deployed on the background server, including: Preset model iteration stop conditions; Construct labeled training data for several underwater impact scenarios x; The hull underwater impact simulation prediction model P is supervisedly trained and learned using the labeled training data until a preset model iteration stop condition is reached to obtain a qualified hull underwater impact simulation prediction model P; The qualified cabin underwater impact simulation prediction model P is deployed on the background server for cabin underwater impact simulation prediction.
6. A device for simulating and predicting underwater impact of a cabin, the device being used to implement the method for simulating and predicting underwater impact of a cabin as claimed in any one of claims 1 to 5, characterized in that: The device comprises: A data acquisition unit is used to obtain the structure and performance data of the target cabin and input it into the backend server; The backend server is used to call the underwater impact simulation prediction model P of the cabin to identify and predict the structure and performance data of the target cabin, match and output the prediction results of the maximum impact velocity and / or acceleration of the target cabin subjected to underwater impact and output them; The data acquisition unit is in communication connection with the background server.
7. A prediction device, characterized in that The prediction device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method for simulating and predicting underwater impact of a cabin as described in any one of claims 1 to 5 is implemented.
8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, which can be called by a processor to execute the method for simulating and predicting underwater impact of a cabin as described in any one of claims 1 to 5.
Citation Information
Patent Citations
Rapid prediction method for explosive shock wave load in ship cabin considering bulkhead shielding
CN115081110A
Fan blade dynamic impact detection method and system based on digital twin modeling
CN115859837A
Method for rapidly calculating pressure load of adjacent damaged cabin of internal explosion main cabin
CN115879308A
Deepwater explosion shock wave loading method and system based on elliptical reflection principle
CN115906320A
Topological optimization method for damping arrangement of launching cabin structure
CN117669026A
