Net cage state simulation method and system and mobile storage medium

Abstract

The invention discloses a net cage state simulation method and system and a removable storage medium, and relates to the technical field of aquaculture, and the net cage state simulation method comprises the following steps: employing a two-dimensional plate unit to carry out simulation analysis on a thin-wall structure of a net cage body, and simulating the acting force of waves and ocean currents on the thin-wall structure; performing simulation analysis on the net rack structure and the netting structure of the net cage body by adopting a Morison model, and simulating the acting force of waves and ocean currents on the net rack structure and the netting structure; and utilizing a neural network model to evaluate local area deformation conditions and stress conditions of the net cage body under different sea conditions. According to the technical scheme, by combining different simulation methods and cooperating with the neural network model, the accuracy of net cage state simulation is improved, the performance of the net cage body under different marine environment conditions is evaluated, the method has the advantages of being short in period and low in cost, and the net cage body can be optimized and iterated in the design stage.

Description

Technical Field

[0001] The present invention relates to the technical field of aquaculture, and in particular to a cage state simulation method, system and removable storage medium. Background Art

[0002] Large-scale deep-sea aquaculture cages are a relatively new type of aquaculture equipment. Their structural forms are mostly hexagonal or rectangular, and are composed of structural components such as buoys, columns, struts, spokes, net struts, and nets. The emergence of this new type of aquaculture equipment is intended to adapt to the complex sea conditions in deep-sea areas and improve aquaculture efficiency and output. However, due to the high construction cost and long construction period of large deep-sea aquaculture cages, it is particularly important to quickly evaluate and judge the structural safety of aquaculture cages. At present, there are relatively few methods for evaluating the structural safety of aquaculture equipment. Evaluation and judgment through model experiments face high costs and long periods, and simulations under laboratory conditions are difficult to ensure the accuracy of their boundary conditions. These factors will affect the accuracy of the experimental results. Summary of the invention

[0003] The main purpose of the present invention is to propose a cage state simulation method, system and removable storage medium, aiming to solve the problems of high cost, long cycle and low accuracy of the model experiment adopted in the current aquaculture cage.

[0004] To achieve the above object, the cage state simulation method proposed in the present invention comprises the following steps:

[0005] A two-dimensional plate unit is used to simulate and analyze the thin-walled structure of the cage body to simulate the forces of waves and currents on the thin-walled structure;

[0006] The Morison model is used to simulate and analyze the grid structure and the net structure of the cage body, simulating the forces of waves and currents on the grid structure and the net structure;

[0007] The neural network model is used to evaluate the local deformation and stress conditions of the cage body under different sea conditions.

[0008] In one embodiment, the step of using two-dimensional plate units to simulate and analyze the thin-walled structure of the cage body to simulate the force of waves and currents on the thin-walled structure includes the following steps:

[0009] The net cage body comprises a thin-walled structure, a frame structure buoyancy box and a net structure, wherein the thin-walled structure comprises a buoyancy box and a column, and the frame structure comprises a strut and a spoke; and / or,

[0010] The step of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body, and simulating the force of waves and currents on the grid structure and the net structure, includes the following steps:

[0011] The ratio of the characteristic size of the structures in the grid structure and the net structure to the wave wavelength is less than or equal to 0.2, that is, Where D is the characteristic size of the structure, λ is the wave wavelength, and the characteristic size D is selected based on the geometric characteristics of the structure and the main direction of the wave action.

[0012] In one embodiment, the step of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body, and simulating the force of waves and currents on the grid structure and the net structure, comprises the following steps:

[0013] The net structure is simplified into a rod structure by mesh grouping, and then the Morison model is used to simulate the forces of waves and currents on the rod structure.

[0014] The mesh grouping method is used to convert the net cloth into an equivalent model of a large-size grid. The prerequisites for establishing the equivalent model are:

[0015] The net structure is assumed to be an indeformable rigid body, the shielding effect between the net structures and the inertia force of the net structure are ignored, and only the drag force of the net structure is considered.

[0016] In one embodiment, the step of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body, and simulating the force of waves and currents on the grid structure and the net structure, includes the following steps:

[0017] The Morison model is used to calculate the forces acting on the cage body, which include horizontal drag force and horizontal inertial force. The horizontal drag force is caused by the relative motion speed between the wave water particles and the grid structure and the net structure, and the horizontal inertial force is caused by the relative acceleration between the wave water particles and the grid structure. The calculation formula of the Morison model is as follows:

[0018]

[0019] Among them, C d Indicates the drag coefficient; C M represents the inertia coefficient; A represents the effective projected area; V0 represents the volume of displaced water; u represents the velocity of the water particle; du / dt represents the acceleration of the water particle.

[0020] In one embodiment, the step of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body, and simulating the force of waves and currents on the grid structure and the net structure, includes the following steps:

[0021] In deep sea areas, the grid structure and net structure are subject to not only horizontal drag force and horizontal inertia force, but also the combined drag force f under the combined action of the currents in the sea area. -D ., where the combined drag force f -D The calculation formula is:

[0022]

[0023] f -D represents the drag force on small-scale components under the combined action of waves and currents; ρ represents the density of seawater; C d represents the drag coefficient; A represents the effective projected area; u represents the velocity of water particles under the action of waves; u c It represents the velocity of water particles under the action of ocean current.

[0024] In one embodiment, the step of using a neural network to evaluate the deformation and stress of a local area of ​​the cage body under different sea conditions comprises the following steps:

[0025] Collecting monitoring data of deformation of a local area of ​​the cage body and preprocessing the monitoring data;

[0026] Build a neural network model using processed monitoring data;

[0027] The preprocessed data is divided into a training set and a test set, and the training set data is used to train the neural network model;

[0028] Use the test set data to verify the trained neural network model and evaluate the stability and accuracy of the neural network model;

[0029] The structural deformation is analyzed and evaluated based on the prediction results of the neural network.

[0030] In one embodiment, the step of collecting the monitoring data of the deformation of the cage body and preprocessing the monitoring data comprises the following steps:

[0031] The monitoring data includes displacement, stress and strain of a local area of ​​the cage body, and the monitoring data is cleaned, denoised and normalized; and / or,

[0032] The step of constructing a neural network model using the processed monitoring data comprises the following steps:

[0033] Using the processed monitoring data, obtaining the network parameters related to the number of layers, number of nodes, and activation functions, and constructing a feedforward neural network model; and / or,

[0034] The step of using the test set data to verify the trained neural network model and evaluate the stability and accuracy of the neural network model includes the following steps:

[0035] The prediction accuracy and stability of the model are evaluated by comparing the differences between the predicted results and the actual monitoring data.

[0036] The step of collecting the monitoring data of the deformation of the cage body and preprocessing the monitoring data includes the following steps, and the calculation formula is as follows:

[0037]

[0038] h j =f(net j )

[0039]

[0040] Among them, I m represents the input parameter, ω jm is the weight between the input layer neurons and the hidden layer neurons, θ j is the corresponding threshold, M is the number of neurons in the input layer, and net j represents the hidden layer input parameter; f is the corresponding activation function, h j is the output data of the hidden layer; ω nk Represents the weight between the hidden layer neurons and the output layer neurons, θ n is the corresponding threshold, K is the number of neurons in the hidden layer, F is the corresponding activation function, O n is the final output data.

[0041] The present invention also proposes a cage state simulation system, comprising a memory, a processor, and a cage state simulation program stored in the memory and executable on the processor, wherein the cage state simulation program implements the steps of the cage state simulation method described above when executed by the processor.

[0042] The present invention further provides a removable storage medium, on which a network cage state simulation program is stored. When the network cage state simulation program is executed by a processor, the steps of the network cage state simulation method described above are implemented.

[0043] The technical solution of the present invention uses two-dimensional plate units to simulate and analyze the thin-walled structure of the cage body, simulating the forces of waves and currents on the thin-walled structure; uses the Morison model to simulate and analyze the grid structure and net structure of the cage body, simulating the forces of waves and currents on the grid structure and net structure; and uses a neural network model to evaluate the deformation and stress of the local area of ​​the cage body under different sea conditions. By using two-dimensional plate units in finite element analysis software to simulate the thin-walled structure of the cage body, the bending and torsion behavior of the structure when subjected to external forces can be effectively captured. In the simulation analysis, the Morison model can be used to calculate the forces on the grid structure and net structure of the cage under the action of waves and currents, which is crucial for evaluating the stability and durability of the cage and helps to understand and predict the changes of the cage body under different sea conditions. By collecting monitoring data on cage deformation under different sea conditions, the neural network model established using the monitoring data can quickly evaluate the deformation and stress conditions of the cage under various sea conditions, thereby ensuring the safety of the cage and facilitating the optimization of its design; by combining different simulation methods, the accuracy of cage state simulation is improved, and a comprehensive evaluation of the performance of the cage under different marine environmental conditions is achieved. This method has the advantages of short cycle and low cost, so that the cage body can be optimized and iterated during the design stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 the structures shown in these drawings without paying creative work.

[0045] Figure 1 A schematic diagram of the steps of an embodiment of a cage state simulation method provided by the present invention;

[0046] Figure 2 A schematic diagram of step S300 of an embodiment of a cage state simulation method provided by the present invention;

[0047] Figure 3 A simplified schematic diagram of a net structure in step S200 of an embodiment of a net cage state simulation method provided by the present invention;

[0048] Figure 4 A schematic diagram of the deformation of the cage structure according to an embodiment of the cage state simulation method provided by the present invention;

[0049] Figure 5 A schematic diagram of the cage structure stress condition of an embodiment of the cage state simulation method provided by the present invention.

[0050] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components in a certain specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0053] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features limited to "first" and "second" may explicitly or implicitly include at least one of the features. In addition, if "and / or" or "and / or" appears in the full text, its meaning includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or solutions that satisfy both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in this field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0054] Large-scale deep-sea aquaculture cages are a relatively new type of aquaculture equipment. Their structural forms are mostly hexagonal or rectangular, and are composed of structural components such as buoys, columns, struts, spokes, net struts, and nets. The emergence of this new type of aquaculture equipment is intended to adapt to the complex sea conditions in deep-sea areas and improve aquaculture efficiency and output. However, due to the high construction cost and long construction period of large deep-sea aquaculture cages, it is particularly important to quickly evaluate and judge the structural safety of aquaculture cages. At present, there are relatively few methods for evaluating the structural safety of aquaculture equipment. Evaluation and judgment through model experiments face high costs and long periods, and simulations under laboratory conditions are difficult to ensure the accuracy of their boundary conditions. These factors will affect the accuracy of the experimental results.

[0055] In order to solve the above problems, the present invention proposes a cage state simulation method.

[0056] See also Figure 1 In one embodiment of the present invention, the cage state simulation method comprises the following steps:

[0057] S100: using a two-dimensional plate unit to simulate and analyze the thin-wall structure of the cage body, simulating the force of waves and currents on the thin-wall structure;

[0058] S200: using the Morison model to simulate and analyze the grid structure and the net structure of the cage body, simulating the force of waves and currents on the grid structure and the net structure;

[0059] S300: Using a neural network model to evaluate the deformation and stress conditions of the local area of ​​the cage body under different sea conditions.

[0060] By using the two-dimensional plate unit in the finite element analysis software to simulate the thin-walled structure of the cage body, the bending and torsion behavior of the structure when subjected to external forces can be effectively captured. In the simulation analysis, the Morison model can be used to calculate the forces on the cage's grid structure and net structure under the action of waves and currents, which is crucial for evaluating the stability and durability of the cage and helps to understand and predict the changes of the cage body under different sea conditions. By collecting monitoring data on cage deformation under different sea conditions, the neural network model established using the monitoring data can quickly evaluate the deformation and stress of the cage under various sea conditions, and then improve the structural strength of the area according to the predicted situation; by combining different simulation methods, the accuracy of cage state simulation is improved, and the performance of the cage under different marine environmental conditions is fully evaluated. It has the advantages of short calculation cycle and low calculation cost, so that the cage body can be optimized and iterated during the design stage.

[0061] In one embodiment, the thin-walled structure of the cage body is simulated and analyzed using a two-dimensional plate unit, and before the step S100 of simulating the force of waves and currents on the thin-walled structure, the following steps are included:

[0062] The net cage body comprises a thin-wall structure, a frame structure buoyancy box and a net structure, wherein the thin-wall structure comprises a buoyancy box and a column, and the frame structure comprises a support rod and a spoke rod.

[0063] Determine the structural layout of the cage body, including thin-walled structure, grid structure and mesh structure, and determine the composition of thin-walled structure and grid structure, select suitable materials for each part of the cage, consider factors such as corrosion resistance, strength, weight and cost, use computer-aided design software to create detailed geometric models of each part of the cage, define physical and mechanical properties such as density, elastic modulus, yield strength, etc. for each part of the cage body, mesh the geometric model, generate finite element mesh, and for thin-walled structures, use two-dimensional plate units for meshing, and determine the boundary conditions of the model, such as support points, fixed points, etc. Define the external loads acting on the cage, including environmental loads such as waves, currents, and wind. If there are existing experimental data or similar cases, verify the model to ensure the accuracy of the simulation. Then, use two-dimensional plate units to simulate and analyze the thin-walled structure to simulate the forces of waves and currents on the thin-walled structure. The Morison model can also be used to simulate and analyze the grid structure and the net structure, and the neural network model can be used to evaluate the deformation and stress of the local area of ​​the cage body under different sea conditions. The structural strength of the area can be improved according to the predicted situation, providing an important basis for the design optimization and performance evaluation of the cage body.

[0064] In one embodiment, the step S200 of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body and simulating the force of waves and currents on the grid structure and the net structure includes the following steps:

[0065] The net structure is simplified into a rod structure by mesh grouping method, and then the Morison model is used to simulate the forces of waves and currents on the rod structure.

[0066] Before conducting Morison model simulation analysis, the mesh structure needs to be preprocessed to simplify the model and reduce the amount of calculation. The mesh structure is usually composed of a large number of meshes, and directly simulating each mesh will result in too much calculation. Figure 3 , the mesh grouping method is used to simplify multiple meshes into equivalent rod structures; this equivalence is based on the physical properties of the net, such as wire diameter, mesh foot length and material properties, to ensure that the equivalent rod structure can represent the mechanical behavior of the actual net. The necessary physical properties such as density, elastic modulus, cross-sectional area and shape are defined for the simplified rod structure, which will be used for subsequent Morison model simulation analysis; the Morison model is used to calculate the forces of waves and currents on the rod structure.

[0067] In one embodiment, the step of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body, and simulating the force of waves and currents on the grid structure and the net structure, comprises the following steps:

[0068] The ratio of the characteristic size of the structures in the grid structure and the net structure to the wave wavelength is less than or equal to 0.2, that is, Where D is the characteristic size of the structure, λ is the wave wavelength, and the characteristic size D is selected based on the geometric characteristics of the structure and the main direction of the wave action.

[0069] The Morison model is an empirical formula for calculating the force of waves on marine structures. The Morison model is applicable to small-scale components, where the characteristic size of the structure is much smaller than the wavelength. The characteristic size of the structure can be determined based on the geometric characteristics of the structure and the main direction of wave action. For circular or cylindrical grid structures, the characteristic size can be its diameter, radius or length. For grid structures with non-circular cross-sections, the characteristic size can be expressed by an equivalent diameter, which refers to the diameter of a circle with the same cross-sectional area. For thin plate structures, the characteristic size can be its width or height.

[0070] In one embodiment, the net structure is simplified into a rod structure by a mesh grouping method, and then the step of simulating the force of waves and currents on the rod structure by using the Morison model includes the following steps:

[0071] The mesh grouping method is used to convert the net into an equivalent model of a large-size grid. The prerequisite for establishing the equivalent model is as follows: the net structure is assumed to be an undeformable rigid body, the shielding effect between the net structures and the inertia force of the net structure are ignored, and only the drag force of the net structure is considered.

[0072] The hydrodynamic load of the net structure is simulated by equivalent Morrison unit, that is, a single net line is simulated by Morrison unit. In order to improve the calculation efficiency and take into account the limitation of the number of Morrison units by the calculation software, the mesh grouping method is used to convert the net into an equivalent model of large mesh grid. The prerequisites for establishing the equivalent model are as follows: in the simulation, it is assumed that the net structure will not deform, which simplifies the model and reduces the complexity of the calculation; in the net structure, denser meshes may affect the water flow of adjacent meshes, and this effect is called shielding effect. In this model, in order to further simplify the calculation, this effect is ignored; because the inertia force of the net is much smaller than the drag force caused by waves and currents, only the drag force is considered in the model, and the inertia force is ignored.

[0073] By using an equivalent model and calculating using the Morison model, where the force in the Morison model only includes the drag force, the dynamic load caused by waves and currents is calculated. The net structure may include a breeding net and a protective net. For example, the following are the actual parameters of the main parameters of the breeding net in a specific embodiment and the equivalent parameters corresponding to the equivalent model;

[0074]

[0075] The following are the actual parameters of the main parameters of the protection net in a specific embodiment and the equivalent parameters of the corresponding equivalent model:

[0076]

[0077] In one embodiment, the step S300 of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body to simulate the force of waves and currents on the grid structure and the net structure includes the following steps:

[0078] The Morison model is used to calculate the forces acting on the cage body, which include horizontal drag force and horizontal inertial force. The horizontal drag force is caused by the relative motion speed between the wave water particles and the grid structure and the net structure, and the horizontal inertial force is caused by the relative acceleration between the wave water particles and the grid structure.

[0079] The calculation formula of the Morison model is as follows:

[0080]

[0081] Among them, C d Indicates the drag coefficient; C M represents the inertia coefficient; A represents the effective projected area; V0 represents the volume of water displaced; u represents the velocity of the water particle; Represents the acceleration of the water particle.

[0082] By analyzing the forces of waves and currents on the cage structure, it is determined that the types of forces include horizontal drag force and horizontal inertia force. Through the above formula, the forces on the cage structure under different wave and current conditions can be calculated. The calculation results are crucial for evaluating the stability, durability and safety of the cage structure. Designers and engineers can use this data to optimize the design of the cage to ensure its performance and survivability in harsh marine environments.

[0083] In one embodiment, the step S300 of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body to simulate the force of waves and currents on the grid structure and the net structure includes the following steps:

[0084] In deep sea areas, the grid structure and net structure are subject to not only horizontal drag force and horizontal inertia force, but also the combined drag force f under the combined action of the currents in the sea area. -D ., where the combined drag force f -D The calculation formula is:

[0085]

[0086] f -D represents the drag force on small-scale components under the combined action of waves and currents; ρ represents the density of seawater; C d represents the drag coefficient; A represents the effective projected area; u represents the velocity of water particles under the action of waves; u c It represents the velocity of water particles under the action of ocean current.

[0087] In deep sea areas, in addition to considering the horizontal drag force and horizontal inertial force on the cage body, it is also necessary to consider the combined drag force caused by the ocean current, calculate the combined drag force on the grid structure and the net structure under the combined action of waves and currents, and use the extended Morison model to calculate the magnitude and direction of the combined drag force; the calculation of this combined force is crucial for evaluating the stability and durability of the cage structure in a complex marine environment. Designers and engineers can use this data to optimize the design of the cage to ensure its performance and survivability in the deep sea environment; it helps to improve the efficiency and reliability of the cage design. By considering the combined effects of waves and currents, the behavior of the cage in the actual marine environment can be more accurately predicted.

[0088] In one embodiment, reference Figure 2 The step S300 of using a neural network to evaluate the deformation and stress of a local area of ​​the cage body under different sea conditions comprises the following steps:

[0089] S31: collecting monitoring data of deformation of a local area of ​​the cage body, and preprocessing the monitoring data;

[0090] S32: constructing a neural network model using the processed monitoring data;

[0091] S33: dividing the preprocessed data into a training set and a test set, and using the training set data to train the neural network model;

[0092] S34: Use the test set data to verify the trained neural network model and evaluate the stability and accuracy of the neural network model;

[0093] S35: Analyze the structural deformation according to the prediction results of the neural network.

[0094] Collect deformation data of local areas of the cage body under different sea conditions. These data may include parameters such as displacement, stress, strain, etc., which can be obtained through actual monitoring, experiments or detailed numerical simulations. Preprocess the collected data to improve data quality, eliminate outliers and noise, and convert the data into a format suitable for neural network processing. Select a suitable neural network structure according to the nature of the problem, such as a feedforward neural network or a convolutional neural network or a recurrent neural network; after selecting a suitable neural network structure, it is necessary to set parameters such as the number of layers of the network, the number of nodes per layer, and the activation function; divide the preprocessed data into a training set and a test set. Usually, the training set is used to train the model, and the test set is used to evaluate the generalization ability of the model; the neural network model is trained using the training set data. During the training process, the prediction error is minimized by adjusting the network weights and biases. Commonly used optimization algorithms include gradient descent, Adam, etc. Use the test set data to verify the trained neural network model. Evaluate the prediction accuracy, stability and generalization ability of the model; analyze the deformation and stress of the cage body under different sea conditions based on the prediction results of the neural network; exemplarily, refer to Figure 4 , is to obtain the deformation of the cage structure under the preset sea conditions; refer to Figure 5 , is to obtain the stress of the cage structure under the preset sea conditions; identify the area with the most serious deformation, evaluate the safety and durability of the structure; based on the analysis results, optimize the cage design to improve its performance and safety under severe sea conditions. The neural network model can provide an efficient and accurate method to predict and evaluate the deformation and stress of the cage under different sea conditions, improve the structural strength of the area according to the predicted situation, and provide a scientific basis for the design and maintenance of the cage.

[0095] In one embodiment, the step of collecting the monitoring data of the deformation of the cage body and preprocessing the monitoring data comprises the following steps:

[0096] The monitoring data includes displacement, stress and strain of the local area of ​​the cage body, and the monitoring data is cleaned, denoised and normalized. The displacement, stress and strain data of the local area of ​​the cage body are collected, and the above data can come from actual on-site monitoring equipment, such as sensors, or from detailed computer numerical simulation or laboratory testing. By cleaning, denoising and normalizing the monitoring data, the quality of the monitoring data can be ensured, providing a solid foundation for establishing an accurate and reliable neural network model.

[0097] In one embodiment, the step of constructing a neural network model using the processed monitoring data comprises the following steps:

[0098] Using the processed monitoring data, we obtain the network parameters related to the number of layers, number of nodes, and activation function, and build a feedforward neural network model. The number of layers and nodes can determine the network structure. The number of layers of the neural network includes the input layer, hidden layer, and output layer, and we select the appropriate activation function for the hidden layer.

[0099] In one embodiment, the step of using the test set data to verify and evaluate the trained neural network model includes the following steps:

[0100] The prediction accuracy and stability of the model are evaluated by comparing the difference between the prediction results and the actual monitoring data. The test set data is prepared and preprocessed, and the preprocessed test data is input into the trained neural network model. The test data of the test set is predicted by the neural network model, and the deformation prediction results of the local area of ​​the cage body are output. The difference between the model prediction results and the actual monitoring data of the test set is calculated to obtain the error metric, and the error metric is used to evaluate the prediction accuracy and stability of the model to ensure the reliability and effectiveness of the neural network model.

[0101] The step of collecting the monitoring data of the deformation of the cage body and preprocessing the monitoring data includes the following steps, and the calculation formula is as follows:

[0102]

[0103] h j =f(net j );

[0104]

[0105] h j =f(net j );

[0106]

[0107] Among them, I m represents the input parameter, ω jm is the weight between the input layer neurons and the hidden layer neurons, θ j is the corresponding threshold, M is the number of neurons in the input layer, and net j represents the hidden layer input parameter; f is the corresponding activation function, h j is the output data of the hidden layer; ω nk Represents the weight between the hidden layer neurons and the output layer neurons, θ n is the corresponding threshold, K is the number of neurons in the hidden layer, F is the corresponding activation function, O n is the final output data.

[0108] The technical solution of the present invention uses two-dimensional plate units to simulate and analyze the thin-walled structure of the cage body, simulating the forces of waves and currents on the thin-walled structure; uses the Morison model to simulate and analyze the grid structure and net structure of the cage body, simulating the forces of waves and currents on the grid structure and net structure; and uses a neural network model to evaluate the deformation and stress of the local area of ​​the cage body under different sea conditions. By using two-dimensional plate units in finite element analysis software to simulate the thin-walled structure of the cage body, the bending and torsion behavior of the structure when subjected to external forces can be effectively captured. In the simulation analysis, the Morison model can be used to calculate the forces on the grid structure and net structure of the cage under the action of waves and currents, which is crucial for evaluating the stability and durability of the cage and helps to understand and predict the changes of the cage body under different sea conditions. By collecting monitoring data on cage deformation under different sea conditions, the neural network model established using the monitoring data can quickly evaluate the deformation and stress conditions of the cage under various sea conditions, and then improve the structural strength of the area based on the predictions; by combining different simulation methods, the accuracy of cage state simulation can be improved, and a comprehensive evaluation of the performance of the cage under different marine environmental conditions can be achieved. This has the advantages of short calculation cycle and low calculation cost, so that the cage body can be optimized and iterated during the design phase.

[0109] The present invention also proposes a cage state simulation system, comprising a memory, a processor, and a cage state simulation program stored in the memory and executable on the processor, wherein the cage state simulation program implements the steps of the cage state simulation method described above when executed by the processor.

[0110] The present invention further provides a removable storage medium, on which a network cage state simulation program is stored. When the network cage state simulation program is executed by a processor, the steps of the network cage state simulation method described above are implemented.

[0111] The above description is only an exemplary embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the technical concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. A cage state simulation method, characterized in that: The following steps are involved: A two-dimensional plate unit is used to simulate and analyze the thin-walled structure of the cage body to simulate the forces of waves and currents on the thin-walled structure; The Morison model is used to simulate and analyze the grid structure and the net structure of the cage body, simulating the forces of waves and currents on the grid structure and the net structure; The neural network model is used to evaluate the local deformation and stress conditions of the cage body under different sea conditions.

2. The cage state simulation method according to claim 1, characterized in that: Before the step of using two-dimensional plate units to simulate and analyze the thin-walled structure of the cage body to simulate the force of waves and currents on the thin-walled structure, the following steps are included: The net cage body comprises a thin-walled structure, a frame structure buoyancy box and a net structure, wherein the thin-walled structure comprises a buoyancy box and a column, and the frame structure comprises a strut and a spoke; and / or, The step of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body, and simulating the force of waves and currents on the grid structure and the net structure, includes the following steps: The ratio of the characteristic size of the structures in the grid structure and the net structure to the wave wavelength is less than or equal to 0.2, that is, Where D is the characteristic size of the structure, λ is the wave wavelength, and the characteristic size D is selected based on the geometric characteristics of the structure and the main direction of the wave action.

3. The cage state simulation method according to claim 2, characterized in that: The step of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body, and simulating the force of waves and currents on the grid structure and the net structure, includes the following steps: The net structure is simplified into a rod structure by mesh grouping, and then the Morison model is used to simulate the forces of waves and currents on the rod structure. The mesh grouping method is used to convert the net cloth into an equivalent model of a large-size grid. The prerequisites for establishing the equivalent model are: The net structure is assumed to be an indeformable rigid body, the shielding effect between the net structures and the inertia force of the net structure are ignored, and only the drag force of the net structure is considered.

4. The cage state simulation method according to any one of claims 1 to 3, characterized in that: The steps of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body and simulating the force of waves and currents on the grid structure and the net structure include the following steps: The Morison model is used to calculate the forces acting on the cage body, which include horizontal drag force and horizontal inertial force. The horizontal drag force is caused by the relative motion speed between the wave water particles and the grid structure and the net structure, and the horizontal inertial force is caused by the relative acceleration between the wave water particles and the grid structure. The calculation formula of the Morison model is as follows: Among them, C d Indicates the drag coefficient; C M represents the inertia coefficient; A represents the effective projected area; V0 represents the volume of water displaced; u represents the velocity of the water particle; Represents the acceleration of the water particle.

5. The cage state simulation method according to any one of claims 1 to 3, characterized in that: The steps of using the Morison model to simulate and analyze the grid structure and the net structure of the cage body and simulating the force of waves and currents on the grid structure and the net structure include the following steps: In deep sea areas, the grid structure and net structure are subject to not only horizontal drag force and horizontal inertia force, but also the combined drag force f under the combined action of the currents in the sea area. -D. , where the combined drag force f -D. The calculation formula is: f -D represents the drag force on small-scale components under the combined action of waves and currents; ρ represents the density of seawater; C d represents the drag coefficient; A represents the effective projected area; u represents the velocity of water particles under the action of waves; u c It represents the velocity of water particles under the action of ocean current.

6. The cage state simulation method according to any one of claims 1 to 3, characterized in that: The step of using a neural network to evaluate the deformation and stress conditions of a local area of ​​the cage body under different sea conditions comprises the following steps: Collecting monitoring data of deformation of a local area of ​​the cage body and preprocessing the monitoring data; Build a neural network model using processed monitoring data; The preprocessed data is divided into a training set and a test set, and the training set data is used to train the neural network model; Use the test set data to verify the trained neural network model and evaluate the stability and accuracy of the neural network model; The structural deformation is analyzed based on the prediction results of the neural network.

7. The cage state simulation method according to claim 6, characterized in that: The step of collecting the monitoring data of the deformation of the cage body and preprocessing the monitoring data comprises the following steps: The monitoring data includes displacement, stress and strain of a local area of ​​the cage body, and the monitoring data is cleaned, denoised and normalized; and / or, The step of constructing a neural network model using the processed monitoring data comprises the following steps: Using the processed monitoring data, obtaining the network parameters related to the number of layers, number of nodes, and activation functions, and constructing a feedforward neural network model; and / or, The step of using the test set data to verify the trained neural network model and evaluate the stability and accuracy of the neural network model includes the following steps: The prediction accuracy and stability of the model are evaluated by comparing the differences between the predicted results and the actual monitoring data.

8. The cage state simulation method according to claim 6, characterized in that: The step of collecting the monitoring data of the deformation of the cage body and preprocessing the monitoring data includes the following steps, and the calculation formula is as follows: h j =f(not j ) h j =f(not j ) Among them, I m represents the input parameter, ω jm is the weight between the input layer neurons and the hidden layer neurons, θ j is the corresponding threshold, M is the number of neurons in the input layer, and net j represents the hidden layer input parameter; f is the corresponding activation function, h j is the output data of the hidden layer; ω nk Represents the weight between the hidden layer neurons and the output layer neurons, θ n is the corresponding threshold, K is the number of neurons in the hidden layer, F is the corresponding activation function, O n is the final output data.

9. A cage state simulation system, characterized in that: The invention comprises a memory, a processor and a network cage state simulation program stored in the memory and executable on the processor, wherein the network cage state simulation program implements the steps of the network cage state simulation method according to any one of claims 1 to 8 when executed by the processor.

10. A removable storage medium, characterized in that: The removable storage medium stores a network cage state simulation program, and when the network cage state simulation program is executed by the processor, the steps of the network cage state simulation method according to any one of claims 1 to 8 are implemented.

Images