A method, device, electronic device, and storage medium for determining an artificial fish reef structure

Abstract

The present application discloses a method for determining the structure of an artificial fish reef. A data set is obtained according to the constructed hydrodynamic simulation model of the artificial fish reef. Model training is performed based on the data set to obtain a hydrodynamic characteristic prediction model of the artificial fish reef, and the hydrodynamic characteristic prediction model of the artificial fish reef is used to predict the corresponding comprehensive index of the wake area according to the structural parameters of the artificial fish reef. Within the value range of the structural parameters of the given artificial fish reef, a set number of groups of structural parameters are randomly generated, and the comprehensive index of the wake area corresponding to each group of initial structural parameters is predicted through the hydrodynamic characteristic prediction model of the artificial fish reef. Each group of the structural parameters is optimized and iterated for a set number of rounds, and the structural parameters corresponding to the optimal comprehensive index of the wake area are determined as the structural parameters of the artificial fish reef. In this solution, the comprehensive index of the artificial fish reef wake area is used as an evaluation index, and the optimal structural parameters of the artificial fish reef can be efficiently searched within the given range of structural parameters.

Description

Technical Field

[0001] This application relates to the field of ocean engineering, and particularly to a method, device, electronic device, and storage medium for determining the structure of an artificial fish reef. Background Art

[0002] An artificial fish reef is a structure placed in the marine or freshwater environment by artificial means, used to improve the aquatic ecological environment, promote the restoration of fishery resources, and protect biodiversity. In order to improve the performance of the wake area of an artificial fish reef, it is necessary to improve the structure and form of the artificial fish reef, and obtain the artificial fish reef with the optimal performance through optimization design.

[0003] In the prior art, the performance optimization of the wake area of artificial fish reefs depends on experience or enumerative experiments, with low efficiency and high cost. At the same time, it is difficult to explore all parameter combinations and it is easy to miss the optimal solution. Summary of the Invention

[0004] This application provides a method, device, electronic device, and storage medium for determining the structure of an artificial fish reef to at least solve the above technical problems existing in the prior art.

[0005] On the one hand, this application provides a method for determining the structure of an artificial fish reef, the method comprising:

[0006] Obtaining a data set according to a constructed hydrodynamic simulation model of an artificial fish reef, the data set including the structural parameters of the artificial fish reef and the corresponding comprehensive index of the wake area, the structural parameters of the artificial fish reef including the volume of the artificial fish reef V , the porosity of the artificial fish reef CR , the shape of the artificial fish reef TYPE , and the height of the artificial fish reef H , and the comprehensive index of the wake area being determined according to the efficiency of the artificial fish reef and the smoothness of the wake area ;

[0007] Performing model training according to the data set to obtain a hydrodynamic characteristic prediction model of the artificial fish reef, the hydrodynamic characteristic prediction model of the artificial fish reef being used to predict the corresponding comprehensive index of the wake area according to the structural parameters of the artificial fish reef;

[0008] Randomly generating a set number of structural parameters within the value range of each structural parameter of the given artificial fish reef, and predicting the comprehensive index of the wake area corresponding to each set of initial structural parameters through the hydrodynamic characteristic prediction model of the artificial fish reef;

[0009] Performing set rounds of optimization iteration on each set of structural parameters, and determining the structural parameters corresponding to the optimal comprehensive index of the wake area as the structural parameters of the artificial fish reef.

[0010] On the other hand, the present application provides a device for determining the structure of an artificial reef, including:

[0011] A simulation module for constructing a hydrodynamic simulation model of the artificial reef and obtaining a data set, where the data set includes the structural parameters of the artificial reef and the corresponding comprehensive index of the wake area. The structural parameters of the artificial reef include the volume of the artificial reef V , the porosity of the artificial reef CR , the shape of the artificial reef TYPE and the height of the artificial reef H . The comprehensive index of the wake area is determined according to the efficiency of the artificial reef and the stability of the wake area ;

[0012] A prediction module for training a model based on the data set to obtain a prediction model of the hydrodynamic characteristics of the artificial reef. The prediction model of the hydrodynamic characteristics of the artificial reef is used to predict the corresponding comprehensive index of the wake area according to the structural parameters of the artificial reef;

[0013] An optimization module for randomly generating a set number of structural parameters within the value range of the structural parameters of the given artificial reef, and predicting the comprehensive index of the wake area corresponding to each set of initial structural parameters through the prediction model of the hydrodynamic characteristics of the artificial reef; and performing a set number of optimization iterations on each set of structural parameters, and determining the structural parameters corresponding to the optimal comprehensive index of the wake area as the structural parameters of the artificial reef.

[0014] On yet another aspect, the present application provides a computer-readable storage medium storing a computer program for executing the method for determining the structure of the artificial reef according to the present application.

[0015] On still another aspect, the present application provides an electronic device, including:

[0016] A processor;

[0017] A memory for storing executable instructions of the processor;

[0018] The processor is configured to read the executable instructions from the memory and execute the instructions to implement the method for determining the structure of the artificial reef according to the present application.

[0019] As can be seen from the above solution, this application uses the comprehensive index of the wake area as the evaluation criterion, selects the optimal artificial reef structure parameters, thereby optimizing the performance of the artificial reef and improving the utilization efficiency of the reef. In the above solution, the hydrodynamic simulation method is used to collect data such as the structure parameters of the artificial reef and the corresponding comprehensive index of the wake area, which can overcome the high cost input required by the traditional flume test and greatly reduce the required time. In addition, the machine learning method is used to predict the comprehensive index of the wake area of the artificial reef, and the artificial reef structure parameters corresponding to the optimal comprehensive index of the wake area are searched within the value range of the given structure parameters. Compared with the traditional enumeration test and empirical design methods, the optimal artificial reef structure can be found quickly and efficiently. Description of the Drawings

[0020] Figure 1 Fig. shows a schematic flow chart of the method for determining an artificial reef in an example of this application;

[0021] Figure 2 Fig. shows a schematic flow chart of the construction of a hydrodynamic simulation model of an artificial reef in an example of this application;

[0022] Figure 3 Fig. shows schematic diagrams of three types of artificial reefs in an example of this application;

[0023] Figure 4 Fig. shows a schematic structural diagram of an artificial reef determination device in an example of this application. Detailed Embodiments

[0024] In order to make the objectives, features, and advantages of this application more obvious and understandable, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of this application.

[0025] An artificial reef is a structure placed in a marine or freshwater environment by artificial means, which is used to improve the aquatic ecological environment, promote the restoration of fishery resources, and protect biodiversity. There are often tidal currents in seawater, and the relatively high tidal current velocity affects the habitats and reproductions of aquatic organisms such as fish, shellfish, and crustaceans. An artificial reef provides a place for aquatic organisms such as fish, shellfish, and crustaceans to inhabit, reproduce, and avoid natural enemies by generating a wake area with a lower flow velocity, promoting the aggregation of biological communities; at the same time, the complex structure of the reef increases the attachment area, which is suitable for the growth of algae and invertebrates and provides a food source for fish. In order to improve the performance of the wake area of an artificial reef, it is necessary to improve the structure and form of the artificial reef, and obtain the artificial reef with the optimal performance through optimized design.

[0026] Therefore, as Figure 1 shown, the present application provides a method for determining the structure of an artificial fish reef, including:

[0027] Operation 101, obtaining a data set according to the constructed hydrodynamic simulation model of the artificial fish reef. The data set includes the structural parameters of the artificial fish reef and the corresponding comprehensive index of the wake area. The structural parameters of the artificial fish reef include the volume of the artificial fish reef V , the porosity of the artificial fish reef CR , the shape of the artificial fish reef TYPE and the height of the artificial fish reef H . The comprehensive index of the wake area is determined according to the efficiency of the artificial fish reef and the stability of the wake area .

[0028] In this example, relevant data of the artificial fish reef are obtained by constructing a hydrodynamic simulation model of the artificial fish reef, which are used to train a prediction model of the hydrodynamic characteristics of the artificial fish reef subsequently to predict the wake area performance of an artificial fish reef with a certain structure.

[0029] The relevant data of the artificial fish reef obtained through the hydrodynamic simulation model of the artificial fish reef mainly include: the structural parameters of the artificial fish reef and the corresponding comprehensive index of the wake area. Assigning the structural parameters of a certain artificial fish reef to the hydrodynamic simulation model of the artificial fish reef and running the hydrodynamic simulation model of the artificial fish reef can obtain the comprehensive index of the wake area under the structural parameters.

[0030] First, construct a hydrodynamic simulation model of the artificial fish reef, as Figure 2 shown, the process is as follows:

[0031] Operation 201, construct a computational domain, which is the spatial area of the water body within a set range around the artificial fish reef.

[0032] The computational domain is a specific spatial area defined during numerical calculation or simulation. Within this area, various physical phenomena or processes can be mathematically modeled and numerically solved according to the physical characteristics and boundary conditions of the specific problem. In the present application, according to the characteristics of the hydrodynamic field of the artificial fish reef, a computational domain is constructed, including the spatial area of the artificial fish reef and the water body within a certain range around it. The length of the computational domain is 15 - 25 times the length of the artificial fish reef to ensure the full development of the wake area of the artificial fish reef. The width is 3 - 4 times the width of the artificial fish reef, and the height is 3 - 4 times the height of the artificial fish reef, ensuring that the influence of the computational domain boundary on the calculation is small.

[0033] It should be particularly noted that the space corresponding to the artificial reef in the computational domain is hollowed out. This means that during the calculation process, the space occupied by the artificial reef is regarded as a special area, and no fluid-related calculations are performed within this area because the artificial reef is a solid structure and does not belong to the fluid. Treating it as a non-fluid area and hollowing it out can focus the calculation on the fluid flow around the artificial reef, more accurately simulate the flow characteristics of the fluid around the solid boundary of the artificial reef, and avoid incorrect results caused by treating the inside of the artificial reef as a fluid. That is, in the present application, the computational domain corresponding to the artificial reef hydrodynamic simulation model is actually the spatial area of the water body within a set range around the artificial reef.

[0034] Operation 202: Perform mesh division on the computational domain. The mesh density of the first area in the computational domain is greater than that of the second area, and the first area is closer to the artificial reef than the second area.

[0035] Based on the computational domain, perform mesh division. When performing mesh division, the closer to the artificial reef, the greater the mesh density. Assume that the size of the mesh at the location with the maximum density is 0.01 - 0.02 times the size of the artificial reef. For example, if the length of the artificial reef is 10 meters, then the mesh size in the length direction will be set between 10×0.01 = 0.1 meters and 10×0.02 = 0.2 meters. Similarly, for the width and height directions of the artificial reef, the mesh sizes in the corresponding directions will be determined according to 0.01 - 0.02 times their respective sizes. The purpose of doing this is to obtain finer meshes in the area close to the artificial reef in order to more accurately capture the complex water flow changes and physical phenomena around the artificial reef and improve the calculation accuracy.

[0036] The first area and the second area in the computational domain of the present application are relative areas. For example, if a random area in the computational domain is determined as the first area for mesh division, then the area adjacent to this area and closer to the artificial reef can be regarded as the second area, and the mesh density of the second area is greater than that of the first area. On the contrary, if a random area in the computational domain is determined as the second area for mesh division, then the area adjacent to this area and farther from the artificial reef can be regarded as the first area, and the mesh density of the first area is smaller than that of the second area.

[0037] Operation 203: Based on the computational domain after mesh division, perform parameter setting for the artificial reef hydrodynamic simulation model to obtain the artificial reef hydrodynamic simulation model.

[0038] The parameter setting of the artificial reef hydrodynamic simulation model at least includes the following aspects: turbulence model, discretization method, solver, operation method, and boundary conditions. Among them:

[0039] Turbulence model. A turbulence model is a tool for mathematically describing and simulating complex flow phenomena such as turbulence. In this application, the Reynolds stress model is used.

[0040] Discretization method. When performing hydrodynamic simulation of artificial reefs, the governing equations are usually continuous partial differential equations and cannot be directly solved. The discretization method is to transform these continuous equations into discrete algebraic equations for numerical solution on a computer. In this application, the finite volume method is used.

[0041] Solver. A program or algorithm used to solve the algebraic equation system obtained by the discretization method. In this application, a pressure - based solver is used (taking pressure as the basic solution variable, obtaining the pressure field by solving the pressure Poisson equation or pressure correction equation, and then calculating other physical quantities such as the velocity field based on the pressure field and other related equations).

[0042] Operation method. It involves various strategies and algorithms in the numerical calculation process to ensure the accuracy, stability, and efficiency of the calculation. In this application, the Semi - Implicit Method for Pressure - Linked Equations (SIMPLE) is used, which is a numerical algorithm widely used in computational fluid dynamics to solve pressure - coupled equation systems, mainly used to solve the coupling problem between velocity and pressure to obtain an accurate flow field solution.

[0043] Boundary conditions. They are the specifications of the fluid flow state on the boundaries of the computational domain, providing necessary constraint information for solving the governing equations. In this application, there are seven surfaces for which boundary conditions need to be set: the inlet boundary is set as a velocity - inlet boundary, the outlet boundary is set as a free - outflow boundary, the left, right, and upper side surfaces are set as symmetric boundary conditions, and the lower plane and the surface of the artificial reef are both set as no - slip boundary conditions.

[0044] Through the above process, the construction of the hydrodynamic simulation model of the artificial reef is completed. Running the hydrodynamic simulation model of the artificial reef can obtain a data set, and the process is as follows:

[0045] First, within the value range of the various structural parameters of the given artificial reef, a set of structural parameters is randomly generated.

[0046] Changing the structural parameters means changing the structure of the artificial reef. The structural parameters at least include:

[0047] Volume of the artificial reef V , and the value range can be 0.1 - 10 m 3 ;

[0048] Porosity of the artificial reef CR , , is the area of the upstream face, is the area of the openings on the upstream face (as shown in Figure 3 ), and its value range is 0 - 1;

[0049] Shape of the artificial reef TYPE , and its value range can be, as shown in Figure 3 : cube, triangular prism, hemispherical;

[0050] Height of the artificial reef can have a value range of 0.1 - 3 m (as shown in Figure 3 exemplarily shows the height of the triangular prism - shaped artificial reef H ).

[0051] Here is just an example, and the value ranges of the above - mentioned various structural parameters can be adjusted as needed.

[0052] Within the given value range, a set of structural parameters (including the specific values of each parameter) is randomly generated, which means a structure of the artificial reef is selected.

[0053] Secondly, according to this set of structural parameters, run the hydrodynamic simulation model of the artificial reef, and capture the hydrodynamic characteristic parameters of the wake area generated by the operation of the hydrodynamic simulation model of the artificial reef, including: the volume of the wake area and the average flow velocity in the wake area .

[0054] Under the selected structure of the artificial reef, run the hydrodynamic simulation model of the artificial reef to simulate the flow characteristics of the fluid around the structure of the artificial reef. Through a custom - defined UDF (user defined functions), capture the hydrodynamic characteristic parameters of the wake area generated by the operation of the hydrodynamic simulation model of the artificial reef, at least including: the volume of the wake area and the average flow velocity in the wake area . Among them:

[0055]

[0056] represents the volume of the th grid in the computational domain, represents the horizontal flow velocity of the th grid, and this formula represents that the volume of the wake area is the sum of the volumes of all grids with horizontal flow velocities less than 0;

[0057]

[0058] This formula represents the average flow velocity in the wake area The sum of the products of the volumes of all grids with horizontal flow velocities less than 0 and their horizontal flow velocities and the volume of the wake region The ratio of

[0059] Then, the efficiency of the artificial fish reef is obtained according to the following formula and the stability of the wake region :

[0060]

[0061]

[0062] Among them, is the inlet flow velocity

[0063] Finally, the product of the efficiency of the artificial fish reef and the stability of the wake region is used as the comprehensive index of the wake region corresponding to this set of structural parameters I , specifically: .

[0064] That is, a set of structural parameters and the corresponding comprehensive index of the wake region are added to the dataset as a sample data. Subsequently, a new set of structural parameters can be randomly generated to repeat the above process, and a new sample data can be obtained. Repeating this way, a dataset with the corresponding data volume can be generated according to needs

[0065] Operation 102: Train a prediction model for the hydrodynamic characteristics of the artificial fish reef based on the dataset. The prediction model for the hydrodynamic characteristics of the artificial fish reef is used to predict the corresponding comprehensive index of the wake region according to the structural parameters of the artificial fish reef

[0066] The dataset obtained through operation 101 can be used to train the prediction model for the hydrodynamic characteristics of the artificial fish reef. Through this model, the comprehensive index of the wake region of an artificial fish reef with a certain structure can be predicted. The input of the model is the structural parameters of the artificial fish reef, and the output is the comprehensive index of the wake region. That is, the correlation function of the input and output data of the model is

[0067]

[0068] The training process of the model can be divided into the following stages

[0069] Stage 1: Data preprocessing. After normalizing the dataset, it is divided into a training set and a test set

[0070] In this application, the normalization is performed according to the following formula

[0071]

[0072] Represents a data set, represents the mean of the data, is the standard deviation of the data, , , represents the total number of data in the data set.

[0073] Assume that the data set contains 4 samples, and the structural parameters of each sample are CR, V, TYPE, H . Taking CR normalization as an example, among the 4 samples CR the values of the parameters are respectively: X 1 = 0.2, X 2 = 0.5, X 3 = 0.3, X 4 = 0.8.

[0074] Calculate the mean , in this example n = 4, then:

[0075]

[0076] Calculate the standard deviation :

[0077] 0.0625;

[0078] 0.0025;

[0079] 0.0225;

[0080] 0.1225;

[0081] .

[0082] Normalization calculation:

[0083] For X 1 = 0.2, ;

[0084] For X 2 = 0.5, ;

[0085] For X 3 = 0.3, ;

[0086] For X 4 = 0.8, .

[0087] According to the above method, each sample in the data set containing various data can achieve normalization processing.

[0088] After normalization, the dataset can be divided into a training set and a test set. In one example, the training set accounts for 80 - 90% of the total data in the dataset for model training, and the test set accounts for 10% - 20% for model testing.

[0089] Phase II, Model Training: Train the initial neural network model with the training set to obtain the model to be tested.

[0090] In one example of this application, according to the complexity of the data in the dataset, the architecture of the initial neural network model can be set as follows: the number of layers is 3 - 4, the number of neurons can be set to 64 - 128, the number of hidden layers is 2 - 3, the output layer, activation function (such as Sigmoid, Thah or Relu), and the learning rate is 0.001 - 0.005.

[0091] Use the training set to train the initial neural network model. A training round can be specified. For example, the specified training round can be selected between 1000 - 10000 rounds. In each round of training, the data in the training set is input into the model sequentially or in batches. The model calculates the input structural parameters based on the current parameters of the neural network and outputs the predicted comprehensive index of the wake area. Then, the difference between the predicted value and the true comprehensive index of the wake area in the training set is calculated through a loss function (such as mean square error, etc.). Based on this difference, the weight parameters of each layer in the neural network are adjusted to make the model prediction value closer to the true value. This process is continuously repeated until the set training round is reached, and then the training stops to obtain the model to be tested.

[0092] Phase III, Model Testing: Test the model to be tested with the test set to determine the coefficient of determination , root mean square error and root mean square relative error .

[0093] If the coefficient of determination , the root mean square error and the root mean square relative error all meet the set conditions (for example ), it indicates that the model is effective, and the model to be tested is used as the prediction model for the hydrodynamic characteristics of artificial fish reefs.

[0094] When using the test set for model testing, the structural parameters in the test set can be sequentially input into the model to be tested, and the model to be tested outputs the corresponding predicted values of the comprehensive index of the wake area. Then, calculate the coefficient of determination , root mean square error and root mean square relative error according to the following formulas:

[0095] , ,

[0096]

[0097]

[0098] is the regression sum of squares, is the regression sum of squares, and n is the number of samples in the test set. is the predicted value of the wake zone comprehensive index output by the model to be tested according to the structural parameters in the i-th sample in the test set. is the true value of the wake zone comprehensive index included in the i-th sample in the test set. is the mean value of the true values of the wake zone comprehensive index included in all samples in the test set.

[0099] Operation 103: Randomly generate a set number of structural parameters within the value range of each structural parameter of the given artificial reef, and predict the wake zone comprehensive index corresponding to each set of initial structural parameters through the artificial reef hydrodynamic characteristics prediction model.

[0100] In this example, the number of sets of structural parameters can be preset, for example, 20 - 100 sets. Assuming 100 sets, then within the value range of each structural parameter of the given artificial reef, 100 sets of structural parameters are randomly generated. The value range of each structural parameter of the given artificial reef is the same as the value range used in the above construction simulation process.

[0101] Input these 100 sets of structural parameters into the artificial reef hydrodynamic characteristics prediction model respectively to obtain the predicted values of the wake zone comprehensive index corresponding to each group. For the convenience of subsequent description, here, these 100 sets of structural parameters are called initial structural parameters, and the predicted values of the corresponding wake zone comprehensive index are called initial wake zone comprehensive indices.

[0102] Operation 104: Perform a set number of optimization iterations on each set of structural parameters, and determine the structural parameters corresponding to the optimal wake zone comprehensive index as the structural parameters of the artificial reef.

[0103] Specifically, based on the multiple sets of initial structural parameters and the corresponding initial wake zone comprehensive indices obtained in Operation 103, the following formula is used to perform the k +1-th round of optimization iteration on the structural parameters:

[0104]

[0105] The value of the i-th set of structural parameters in the k +1-th round of optimization iteration, is the value of the i-th group of structural parameters in the k round of optimization iteration, is the change amount of the i-th group of structural parameters in the k +1 round of optimization iteration;

[0106]

[0107] Among them, is the change amount of the i-th group of structural parameters in the k round of optimization iteration, is the inertia weight, is the cognitive parameter, is the social parameter, , are random numbers generated in the range of [0, 1], is the optimal value of the i-th group of structural parameters in the previous k rounds of optimization iteration, is the optimal value among all structural parameters in the previous k rounds of optimization iteration;

[0108] Among them, according to the comprehensive wake region index corresponding to each round of the i-th group of structural parameters in the previous k rounds of optimization iteration, the value of the structural parameter corresponding to the optimal comprehensive wake region index is used as ;

[0109] According to the comprehensive wake region index corresponding to each round of each group of structural parameters in the previous k rounds of optimization iteration, the value corresponding to the optimal comprehensive wake region index is used as .

[0110] The following takes 100 groups of initial structural parameters as an example to specifically illustrate the above process, where the 100 groups of initial structural parameters are denoted as :

[0111] The 1st round of optimization iteration:

[0112] First, calculate the change amount of each group of structural parameters in the 1st round. According to the formula , since there is no previous iterative optimization information in the 1st round of optimization iteration, therefore, = 0, is the initial structural parameter itself, that is, , is the initial structural parameter corresponding to the optimal comprehensive wake region index among the 100 groups of initial structural parameters. Assume .;

[0113] Then If, then Simplify to .

[0114] Update the value of the structural parameter:

[0115] Based on the value of the new structural parameter , the predicted value of the comprehensive index of the wake area is obtained through the hydrodynamic characteristics prediction model of the artificial reef. At this time, and can be updated:

[0116] Update: For the i-th group of structural parameters, there are currently two values and . Compare the comprehensive indexes of the wake areas of the two. If 's comprehensive index of the wake area is better than 's comprehensive index of the wake area (in one example, the larger the value of the comprehensive index of the wake area, the better the comprehensive index of the wake area), then is updated from to ; It should be noted that the of each group of structural parameters needs to be determined according to the comprehensive indexes of the wake areas of their two rounds of values. The of some structural parameters may be the initial value, and the of some structural parameters may be the value of the first round.

[0117] Update: Compare the values of 100 groups of structural parameters in the first round, and take the value of the first round of the structural parameter with the best comprehensive index of the wake area, assumed to be . If 's comprehensive index of the wake area is better than , then is updated from to .

[0118] The k +1 round of optimization iteration:

[0119] First, calculate the change amount k of each group of structural parameters in the +1 round according to the formula , where has been calculated in the previous round. Assume = , , then substituting into the above formula gives ;

[0120] Update the value of the structural parameter:

[0121] Based on the values of the new structural parameters , the predicted value of the comprehensive index of the wake area is obtained through the hydrodynamic characteristics prediction model of the artificial reef. At this time, and can be updated:

[0122] Update: If the comprehensive index of the wake area of (i.e., ) is better than the comprehensive index of the corresponding wake area of , update to

[0123] Update: Determine the comprehensive index of the wake area corresponding to the values of 100 groups of structural parameters obtained in the k +1 round of optimization iteration. If the optimal comprehensive index of the wake area (assumed to be the comprehensive index of the wake area of ) is better than the comprehensive index of the corresponding wake area of (i.e., ), then update to the structural parameter value corresponding to the optimal comprehensive index of the wake area, that is .

[0124] If the k +1 round of optimization iteration is the set maximum number of rounds, the updated , that is , is determined as the structural parameter of the artificial reef, that is, the value of the 20th group of structural parameters in the k +1 round is used as the final structural parameter of the artificial reef.

[0125] As can be seen from the above solution, this application uses the comprehensive index of the wake area as the evaluation criterion to select the optimal structural parameters of the artificial reef, thereby optimizing the performance of the artificial reef and improving the use efficiency of the reef. In the above solution, the hydrodynamic simulation method is used to collect the training data of the comprehensive index prediction model of the wake area, which can overcome the high cost investment required by the traditional flume test and greatly reduce the required time. In addition, the machine learning method is used to predict the comprehensive index of the wake area of the artificial reef and find the structural parameters of the artificial reef corresponding to the optimal comprehensive index of the wake area. Compared with the traditional enumeration test and empirical design method, it can quickly and efficiently find the optimal artificial reef structure.

[0126] For this reason, as shown in Figure 4 , this application also provides a device for determining the structure of an artificial reef, including:

[0127] The simulation module 10 is used to construct a hydrodynamic simulation model of artificial reefs and obtain a data set, where the data set includes the structural parameters of artificial reefs and the corresponding comprehensive indicators of the wake area. The structural parameters of the artificial reefs include the volume of the artificial reef V , the porosity of the artificial reef CR , the shape of the artificial reef TYPE and the height of the artificial reef H . The comprehensive indicators of the wake area are determined according to the efficiency of the artificial reef and the stability of the wake area ;

[0128] The prediction module 20 is used to train a model according to the data set to obtain a prediction model of the hydrodynamic characteristics of artificial reefs. The prediction model of the hydrodynamic characteristics of artificial reefs is used to predict the corresponding comprehensive indicators of the wake area according to the structural parameters of the artificial reefs;

[0129] The optimization module 30 is used to randomly generate a set number of structural parameters within the value range of the structural parameters of the given artificial reef, and predict the comprehensive indicators of the wake area corresponding to each set of initial structural parameters through the prediction model of the hydrodynamic characteristics of the artificial reef; and perform a set number of optimization iterations on each set of structural parameters, and determine the structural parameters corresponding to the optimal comprehensive indicators of the wake area as the structural parameters of the artificial reef.

[0130] Among them, when constructing the hydrodynamic simulation model of artificial reefs, the simulation module 10 is used for:

[0131] Construct a computational domain, which is the spatial area of the water body within a set range around the artificial reef;

[0132] Perform mesh division on the computational domain. The mesh density of the first area in the computational domain is greater than that of the second area, and the first area is closer to the artificial reef than the second area;

[0133] Based on the computational domain after mesh division, perform parameter setting of the hydrodynamic simulation model of artificial reefs to obtain the hydrodynamic simulation model of artificial reefs.

[0134] When obtaining the data set according to the constructed hydrodynamic simulation model of artificial reefs, the simulation module 10 is also used for:

[0135] Randomly generate a set of structural parameters within the value range of the structural parameters of the given artificial reef;

[0136] Run the hydrodynamic simulation model of artificial reefs according to this set of structural parameters, and capture the hydrodynamic characteristic parameters of the wake area generated by the operation of the hydrodynamic simulation model of artificial reefs, including: the volume of the wake area and the average flow velocity of the wake area ;

[0137] The efficiency of the artificial reef is obtained according to the following formula and the stability of the wake area :

[0138]

[0139]

[0140] The is the inlet flow velocity;

[0141] Multiply the efficiency of the artificial reef and the stability of the wake area as the comprehensive index of the wake area corresponding to this set of structural parameters.

[0142] When performing the set number of optimization iterations on each set of structural parameters, the optimization module 30 is specifically used for:

[0143] Perform the k +1 round of optimization iteration on the structural parameters through the following formula:

[0144]

[0145] The value of the i-th group of structural parameters in the k +1 round of optimization iteration, is the value of the i-th group of structural parameters in the k round of optimization iteration, is the change amount of the i-th group of structural parameters in the k +1 round of optimization iteration;

[0146]

[0147] Among them, is the change amount of the i-th group of structural parameters in the k round of optimization iteration, is the inertia weight, is the cognitive parameter, is the social parameter, 、 are random numbers generated in the range of [0,1], is the optimal value of the i-th group of structural parameters in the previous k rounds of optimization iteration, is the optimal value of all structural parameters in the previous k rounds of optimization iteration;

[0148] Among them, according to the i-th group of structural parameters in the previous kThe comprehensive index of the wake area corresponding to the value of each round in the round-by-round optimization iteration is used to take the value corresponding to the optimal comprehensive index of the wake area as the ;

[0149] According to the comprehensive index of the wake area corresponding to the value of each round of each group of structural parameters in the previous k rounds of optimization iteration, the value corresponding to the optimal comprehensive index of the wake area is used as the .

[0150] After the k +1 round of optimization iteration, the optimization module 30 is further configured to:

[0151] If it is determined that the comprehensive index of the wake area is better than the corresponding comprehensive index of the wake area, the is updated to the ;

[0152] Determine the comprehensive index of the wake area corresponding to the values of all groups of structural parameters obtained in the k +1 round of optimization iteration. If the optimal comprehensive index of the wake area is better than the corresponding comprehensive index of the wake area, then the is updated to the value of the structural parameter corresponding to the optimal comprehensive index of the wake area.

[0153] When performing a set number of rounds of optimization iteration on each group of structural parameters and determining the structural parameters of the artificial fish reef by taking the structural parameters corresponding to the optimal comprehensive index of the wake area, the optimization module 30 is specifically configured to: If the k +1 round of optimization iteration is the set maximum number of rounds, the updated is determined as the structural parameter of the artificial fish reef.

[0154] When predicting the hydrodynamic characteristics of the artificial fish reef by training the model according to the dataset, the prediction module 20 is specifically configured to:

[0155] After normalizing the dataset, divide it into a training set and a test set;

[0156] Train the initial neural network model through the training set to obtain a model to be tested;

[0157] Test the model to be tested through the test set, and determine the coefficient of determination , root mean square error and root mean square relative error of the model to be tested, where:

[0158] , ,

[0159]

[0160]

[0161] The is the regression sum of squares, is the regression sum of squares, and n is the number of samples in the test set. is the wake area comprehensive index predicted by the model to be tested for the structural parameters in the i-th sample in the test set. is the wake area comprehensive index included in the i-th sample in the test set. is the mean value of the wake area comprehensive indexes included in all samples in the test set.

[0162] If the coefficient of determination , the root mean square error and the root mean square relative error all meet the set conditions, then the model to be tested is used as the prediction model for the hydrodynamic characteristics of artificial fish reefs.

[0163] This application also provides an electronic device, including a processor and a memory for storing executable instructions that can be executed by the processor; wherein, the processor is configured to read the executable instructions from the memory and execute the instructions to implement the method for determining the structure of the artificial fish reef disclosed in the foregoing examples.

[0164] In addition to the above methods, devices and equipment, an embodiment of this application may also be a computer program product, which includes computer program instructions. When the computer program instructions are run by a processor, the processor is caused to execute the steps in the methods according to various embodiments of this application described in the "Exemplary Methods" section of this specification.

[0165] The computer program product can be written in any combination of one or more programming languages to write program code for performing the operations of the embodiments of this application. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, executed as an independent software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0166] In addition, an embodiment of the present application may also be a computer-readable storage medium storing computer program instructions, which, when run by a processor, cause the processor to execute the steps in the methods according to various embodiments of the present application described in the above "Exemplary Methods" section of this specification.

[0167] The computer-readable storage medium may adopt any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. The readable storage medium may, for example, include but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0168] The basic principles of the present application have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are only examples and not limitations. It cannot be considered that these advantages, benefits, effects, etc. are essential for each embodiment of the present application. In addition, the above-disclosed specific details are only for illustrative purposes and for ease of understanding, and are not limitations. The above details do not limit the present application to necessarily implement using the above specific details.

[0169] The block diagrams of the devices, apparatuses, equipment, and systems involved in the present application are only illustrative examples and do not intend to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open-ended words, meaning "including but not limited to", and can be used interchangeably with each other. The word "or" and "and" used herein refer to the word "and / or", and can be used interchangeably with each other, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to", and can be used interchangeably with each other.

[0170] It should also be noted that in the devices, equipment, and methods of the present application, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present application.

[0171] The foregoing description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Thus, the present application is not intended to be limited to the aspects shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0172] The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the present application to the form disclosed herein. Although numerous example aspects and embodiments have been discussed above, those skilled in the art will recognize some variations, modifications, alterations, additions, and subcombinations thereof.

Claims

1. A method for determining an artificial fish reef structure, characterized in that, The method includes: Obtain a data set according to the constructed hydrodynamic simulation model of artificial fish reefs. The data set includes the structural parameters of artificial fish reefs and the corresponding comprehensive indicators of the wake area. The structural parameters of the artificial fish reefs include the volume of the artificial fish reef V , the porosity of the artificial fish reef CR , the shape of the artificial fish reef TYPE and the height of the artificial fish reef H . The comprehensive indicator of the wake area is determined according to the efficiency of the artificial fish reef and the stability of the wake area ; Performing model training based on the dataset to obtain a prediction model for the hydrodynamic characteristics of artificial reefs, where the prediction model for the hydrodynamic characteristics of artificial reefs is used to predict the corresponding comprehensive index of the wake area according to the structural parameters of the artificial reef; Randomly generating a set number of groups of structural parameters within the value range of each structural parameter of the given artificial reef, and predicting the comprehensive index of the wake area corresponding to each group of initial structural parameters through the prediction model for the hydrodynamic characteristics of the artificial reef; Performing a set number of rounds of optimization iterations on each group of structural parameters, and determining the structural parameters corresponding to the optimal comprehensive index of the wake area as the structural parameters of the artificial reef; The constructing of the hydrodynamic simulation model of the artificial reef includes: Constructing a computational domain, where the computational domain is the spatial area of the water body within a set range around the artificial reef; Performing mesh division on the computational domain, where the mesh density of the first area in the computational domain is greater than that of the second area, and the first area is closer to the artificial reef than the second area; Based on the computationally divided domain, performing parameter setting of the hydrodynamic simulation model of the artificial reef to obtain the hydrodynamic simulation model of the artificial reef; The obtaining of the dataset according to the constructed hydrodynamic simulation model of the artificial reef includes: Randomly generating a set of structural parameters within the value range of each structural parameter of the given artificial reef; Run the hydrodynamic simulation model of the artificial fish reef according to the set of structural parameters, and capture the hydrodynamic characteristic parameters of the wake area generated by the operation of the hydrodynamic simulation model of the artificial fish reef, including: the volume of the wake area and the average flow velocity in the wake area ; The efficiency of the artificial fish reef is obtained according to the following formula and the stability of the wake area : The said is the inlet flow velocity; Take the product of the efficiency of the artificial fish reef and the stability of the wake area as the comprehensive index of the wake area corresponding to this set of structural parameters.

2. The method according to claim 1, wherein The performing of a set number of rounds of optimization iterations on each group of structural parameters includes: The k +1 round of optimization iteration is performed on the structural parameters through the following formula: The is the value of the i-th group of structural parameters in the k +1-th round of optimization iteration, is the value of the i-th group of structural parameters in the k round of optimization iteration, is the change amount of the i-th group of structural parameters in the k +1-th round of optimization iteration; Among them, is the change amount of the i-th group of structure parameters in the k round of optimization iteration, is the inertia weight, is the cognitive parameter, is the social parameter, 、 are random numbers generated within the range of [0, 1], is the optimal value of the i-th group of structure parameters in the previous k round of optimization iteration, is the optimal value among all structure parameters in the previous k rounds of optimization iteration; Among them, according to the values corresponding to each round in the k round of optimization iteration of the i-th group of structural parameters, the comprehensive index of the wake region is used, and the value corresponding to the optimal comprehensive index of the wake region is used as the ; According to the comprehensive index of the wake region corresponding to the value of each set of structural parameters in each round of optimization iteration in the previous round, the value corresponding to the optimal comprehensive index of the wake region is used as the k 。 。 3. The method according to claim 2, wherein After the k +1 round of optimization iteration, the method further includes: If the comprehensive index of the wake region of is better than the comprehensive index of the corresponding wake region of , update the to the ; Determine the k corresponding wake region comprehensive index for the values of all group structure parameters obtained through the +1 round of optimization iteration. If the optimal wake region comprehensive index among them is better than the corresponding wake region comprehensive index, then the is updated to the value of the structure parameter corresponding to the optimal wake region comprehensive index.

4. The method according to claim 3, characterized in that, The performing of a set number of rounds of optimization iterations on each group of structural parameters and determining the structural parameters corresponding to the optimal comprehensive index of the wake area as the structural parameters of the artificial reef includes: If the k +1 round of optimization iteration reaches the set maximum number of rounds, the updated is determined as the structural parameters of the artificial fish reef.

5. The method according to claim 1, characterized in that The performing of model training based on the dataset to obtain a prediction model for the hydrodynamic characteristics of artificial reefs includes: After normalizing the dataset, dividing it into a training set and a test set; Training an initial neural network model through the training set to obtain a model to be tested; Testing the model to be tested with the test set to determine the coefficient of determination of the model to be tested , root mean square error and root mean square relative error , where: ， ， The is the regression sum of squares, is the regression sum of squares, n is the number of samples in the test set, is the wake region comprehensive index predicted by the model to be tested for the structural parameters in the i-th sample in the test set, is the wake region comprehensive index included in the i-th sample in the test set, is the mean of the wake region comprehensive indexes included in all samples in the test set; If the coefficient of determination , the root mean square error and the root mean square relative error all meet the set conditions, the model to be tested is used as a prediction model for the hydrodynamic characteristics of artificial reefs.

6. An apparatus for determining an artificial fish reef structure, characterized in that Includes: A simulation module, which is used to construct a hydrodynamic simulation model of artificial fish reefs and obtain a data set. The data set includes the structural parameters of the artificial fish reefs and the corresponding comprehensive indexes of the wake area. The structural parameters of the artificial fish reefs include the volume of the artificial fish reef V , the porosity of the artificial fish reef CR , the shape of the artificial fish reef TYPE and the height of the artificial fish reef H . The comprehensive indexes of the wake area are determined according to the efficiency of the artificial fish reef and the stability of the wake area ; A prediction module for performing model training based on the dataset to obtain a prediction model for the hydrodynamic characteristics of artificial reefs, where the prediction model for the hydrodynamic characteristics of artificial reefs is used to predict the corresponding comprehensive index of the wake area according to the structural parameters of the artificial reef; An optimization module for randomly generating a set number of groups of structural parameters within the value range of each structural parameter of the given artificial reef, and predicting the comprehensive index of the wake area corresponding to each group of initial structural parameters through the prediction model for the hydrodynamic characteristics of the artificial reef; and performing a set number of rounds of optimization iterations on each group of structural parameters, and determining the structural parameters corresponding to the optimal comprehensive index of the wake area as the structural parameters of the artificial reef; The constructing of the hydrodynamic simulation model of the artificial reef includes: Constructing a computational domain, where the computational domain is the spatial area of the water body within a set range around the artificial reef; Performing mesh division on the computational domain, where the mesh density of the first area in the computational domain is greater than that of the second area, and the first area is closer to the artificial reef than the second area; Based on the computationally divided domain, performing parameter setting of the hydrodynamic simulation model of the artificial reef to obtain the hydrodynamic simulation model of the artificial reef; The data set obtained according to the constructed hydrodynamic simulation model of artificial fish reefs includes: Randomly generate a set of structural parameters within the value range of the various structural parameters of the given artificial fish reef; Run the hydrodynamic simulation model of the artificial fish reef according to the set of structural parameters, and capture the hydrodynamic characteristic parameters of the wake area generated by the operation of the hydrodynamic simulation model of the artificial fish reef, including: the volume of the wake area and the average flow velocity in the wake area ; The efficiency of the artificial fish reef is obtained according to the following formula and the stability of the wake area : The said is the inlet flow velocity; Take the efficiency of the artificial reef and the stability of the wake area as the comprehensive index of the wake area corresponding to this set of structural parameters.

7. A computer-readable storage medium storing a computer program for executing the method for determining the structure of an artificial fish reef according to any one of claims 1-5 above.

8. An electronic device, comprising: A processor; A memory for storing executable instructions of the processor; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the method for determining the structure of an artificial fish reef according to any one of claims 1-5 above.

Images