A fiber hybrid composite material explosion-resistant structure optimization method and system

Abstract

The present application relates to a method and system for optimizing the explosion-resistant structure of a fiber hybrid composite material, wherein the method comprises: obtaining geometric parameters of the fiber through a Latin hypercube sampling method, establishing a three-dimensional mesoscopic model based on the geometric parameters, and meshing the three-dimensional mesoscopic model to obtain a mesh file; performing finite element calculations through a finite element analysis method based on the mesh file and pre-configured simulation parameters, and obtaining explosion response evaluation indicators according to the results of the finite element calculations; determining the number of neurons of a back propagation neural network based on the number of geometric parameters and explosion response evaluation indicators, using the geometric parameters and the corresponding explosion response evaluation indicators as training data, and training the back propagation neural network based on the training data to obtain an agent model; searching for genetic geometric parameters from a preset fiber geometric parameter range through a genetic algorithm, predicting the fitness of the genetic geometric parameters through a proxy model, and iteratively solving the target geometric parameters based on the fitness.

Description

Technical Field

[0001] The present application relates to the field of fiber hybrid composite explosion-proof structure design, and in particular to a fiber hybrid composite explosion-proof structure optimization method and system. Background Art

[0002] Due to the reinforcement effect of fibers, fiber hybrid composites have higher strength and better impact toughness, and the research on their explosion resistance is of great significance. At present, fiber hybrid composites have been widely used in highway bridges, tunnels, subways and other fields. The parameters such as fiber shape, diameter, aspect ratio, volume content, etc. will affect the mechanical properties of composite materials such as compressive strength and shear strength.

[0003] Currently, the determination of fiber geometric parameters used in composite materials mostly relies on engineering experience. There is a lack of research on geometric parameters applicable to protective engineering, and the potential of fiber hybrid composite materials in the field of explosion resistance cannot be fully realized.

[0004] Therefore, how to improve the performance of complex fiber hybrid materials in the explosion-proof field by optimizing the geometric parameters of the fibers is a key technical problem that needs to be solved urgently. Summary of the invention

[0005] The embodiments of the present application provide a method and system for optimizing the explosion-proof structure of a fiber hybrid composite material, so as to at least solve the problem in the related art of how to improve the performance of the fiber complex hybrid material in the explosion-proof field by optimizing the geometric parameters of the fiber.

[0006] In a first aspect, an embodiment of the present application provides a method for optimizing an explosion-proof structure of a fiber hybrid composite material, comprising:

[0007] The geometric parameters of the fiber are obtained by a Latin hypercube sampling method, a three-dimensional mesoscopic model is established based on the geometric parameters, and the three-dimensional mesoscopic model is meshed to obtain a mesh file;

[0008] Based on the grid file and the pre-configured simulation parameters, a finite element calculation is performed by a finite element analysis method, and an explosion response evaluation index is obtained according to the result of the finite element calculation;

[0009] Determine the number of neurons of the back propagation neural network based on the geometric parameters and the number of the explosion response evaluation indicators, use the geometric parameters and the corresponding explosion response evaluation indicators as training data, and train the back propagation neural network based on the training data to obtain a proxy model;

[0010] Genetic geometric parameters are searched from a preset fiber geometric parameter range by using a genetic algorithm, the fitness of the genetic geometric parameters is predicted by the proxy model, and the target geometric parameters are iteratively solved based on the fitness.

[0011] In one embodiment, the back propagation neural network includes: an input layer, a hidden layer and an output layer, and the determining of neuron data of the back propagation neural network based on the geometric parameters and the number of the explosion response evaluation indexes includes:

[0012] Determine the number of neurons in the input layer according to the number of the geometric parameters;

[0013] Determining the number of neurons in the output layer according to the number of explosion response evaluation indicators corresponding to the geometric parameters;

[0014] The number of neurons in the hidden layer is determined based on the number of neurons in the input layer and the number of neurons in the output layer. The specific formula is as follows:

[0015]

[0016] Among them, h represents the number of neurons in the hidden layer, m represents the number of neurons in the input layer, and n represents the number of neurons in the output layer. Represents a constant between [1,10].

[0017] In one embodiment, the step of training the back propagation neural network based on the training data to obtain the proxy model includes:

[0018] Normalizing the training data, and generating prediction results of the normalized data through the back propagation neural network;

[0019] Denormalizing the prediction results to restore the prediction results of the normalized data to the scale of the training data;

[0020] The neuron parameters of the back propagation network are iteratively updated based on the LM algorithm to obtain a proxy model.

[0021] In one embodiment, searching genetic geometric parameters from a preset fiber geometric parameter range by using a genetic algorithm, predicting the fitness of the genetic geometric parameters by using the proxy model, and iteratively solving the target geometric parameters based on the fitness include:

[0022] Obtaining a preset fiber geometric parameter range, randomly generating an initial population satisfying the fiber geometric parameter range, wherein each individual in the population represents a genetic geometric parameter of the fiber;

[0023] Calculating the fitness of each individual through the agent model, wherein the fitness represents the explosion response evaluation index;

[0024] Iteratively updating the individuals through a selection operator, a crossover operator, and a mutation operator based on the fitness to form an updated population;

[0025] In response to meeting a preset iteration condition, the iteration is stopped, and the individual with the highest fitness is used as the target geometric parameter.

[0026] In one embodiment, the establishing of the three-dimensional mesoscopic model based on the geometric parameters and meshing the three-dimensional mesoscopic model to obtain a mesh file includes:

[0027] The three-dimensional mesoscopic model is established based on the geometric parameters by modeling software, and the three-dimensional mesoscopic model is meshed by the meshing software to obtain the mesh file. The geometric parameters include: fiber shape, length, diameter, and volume content.

[0028] In one embodiment, the explosion response evaluation index includes: the deformation displacement, the force exerted, and the absorbed energy of the center of the rear panel of the fiber hybrid composite material.

[0029] In one embodiment, the simulation parameters include: material properties, boundary conditions, mutual contact conditions and explosion loads.

[0030] In a second aspect, an embodiment of the present application provides a fiber hybrid composite material explosion-resistant structure optimization system, comprising:

[0031] Geometric parameter module: used to obtain geometric parameters of the fiber by Latin hypercube sampling method, establish a three-dimensional mesoscopic model based on the geometric parameters, and mesh the three-dimensional mesoscopic model to obtain a mesh file;

[0032] Response evaluation module: used for performing finite element calculation by finite element analysis method based on the grid file and pre-configured simulation parameters, and obtaining explosion response evaluation index according to the result of the finite element calculation;

[0033] Training module: used to determine the number of neurons of the back propagation neural network based on the number of the geometric parameters and the explosion response evaluation index, use the geometric parameters and the corresponding explosion response evaluation index as training data, and train the back propagation neural network based on the training data to obtain a proxy model;

[0034] Solving module: used for searching genetic geometric parameters from a preset fiber geometric parameter range through a genetic algorithm, predicting the fitness of the genetic geometric parameters through the proxy model, and iteratively solving the target geometric parameters based on the fitness.

[0035] In a third aspect, an embodiment of the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for optimizing the explosion-proof structure of a fiber hybrid composite material as described in the first aspect above is implemented.

[0036] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for optimizing the explosion-proof structure of a fiber hybrid composite material as described in the first aspect above.

[0037] The embodiments of the present application provide a method and system for optimizing explosion-proof structures of fiber hybrid composite materials, which have at least the following technical effects.

[0038] The mesh file is obtained through modeling and meshing to ensure that each area is properly divided, so as to accurately capture the effect of fiber on the mechanical properties of the hybrid material. The proxy model established based on the back propagation neural network has a strong nonlinear solving ability and can accurately predict the explosion resistance of fiber hybrid composites, solving the problem that the finite element calculation of the explosion resistance of composite materials takes a lot of time. The genetic algorithm simulates the biological natural selection process and selects the target geometric parameters with the highest fitness from the preset fiber geometric parameter range. In this way, by introducing the proxy model to calculate the fitness value of the genetic individual, the optimal target geometric parameters are obtained after cyclic iteration, which improves the explosion resistance of the optimized fiber hybrid composite.

[0039] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0041] Figure 1 It is a flow chart of a method for optimizing explosion-proof structure of fiber hybrid composite material shown in an embodiment of the present application;

[0042] Figure 2 is a schematic diagram of a three-dimensional straight steel fiber concrete mesoscopic model according to an exemplary embodiment;

[0043] Figure 3 is a neuron structure diagram of a back propagation neural network according to an exemplary embodiment;

[0044] Figure 4is a schematic diagram showing a comparison of damage areas of straight steel fiber concrete under explosion load before and after optimization according to an exemplary embodiment;

[0045] Figure 5 is a schematic diagram showing the central deformation displacement results of a straight steel fiber reinforced concrete rear panel according to an exemplary embodiment;

[0046] Figure 6 It is a structural block diagram of a fiber hybrid composite material explosion-proof structure optimization system according to an embodiment of the present application;

[0047] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0049] Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. In addition, it can also be understood that although the efforts made in this development process may be complicated and lengthy, for ordinary technicians in this field related to the content disclosed in this application, some changes in design, manufacturing or production based on the technical content disclosed in this application are just conventional technical means, and should not be understood as insufficient content disclosed in this application.

[0050] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0051] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantity limitation, and may indicate the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0052] In a first aspect, the present application provides a method for optimizing the explosion-proof structure of a fiber hybrid composite material. Figure 1 is a flow chart of a method for optimizing explosion-proof structure of fiber hybrid composite material shown in an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0053] Step S101, obtaining geometric parameters of the fiber by a Latin hypercube sampling method, establishing a three-dimensional mesoscopic model based on the geometric parameters, and meshing the three-dimensional mesoscopic model to obtain a mesh file.

[0054] Optionally, Latin hypercube sampling is used to select geometric parameters of the fiber, a three-dimensional microscopic model is established through three-dimensional modeling software, and the established three-dimensional model is meshed to obtain a mesh file. Among them, Latin hypercube sampling can perform stratified random sampling in a multivariate parameter distribution to ensure that the sample is representative in the entire parameter space and improve the parameter quality.

[0055] In one example, step S101 includes: establishing a three-dimensional mesoscopic model based on geometric parameters using modeling software, and meshing the three-dimensional mesoscopic model using meshing software to obtain a mesh file, wherein the geometric parameters include: fiber shape, length, diameter, and volume content.

[0056] Optionally, the geometric parameters are input into the modeling software, and the modeling is performed by the built-in algorithm of the software. The modeling software includes but is not limited to any one of Digimat, COMSOL Multiphysics, HyperMesh and OpenFOAM. In this application, Digimat software is used for modeling and automatic meshing to obtain a mesh file to ensure that each area is properly divided, so as to accurately capture the effect of the fiber on the mechanical properties of the hybrid material. The mesh file is a *.K file.

[0057] Taking the fiber hybrid composite material as straight steel fiber concrete as an example, the optimization parameters include: diameter, length and volume content of the straight steel fiber. Before modeling, the size of the straight steel fiber concrete needs to be set. Assuming the total length is 800mm, the total width is 200mm, and the total height is 80mm, the optimized size of the straight steel fiber obtained is 35mm in length, 0.4mm in diameter, and 2% in volume content. Figure 2 It is a schematic diagram of a three-dimensional straight steel fiber concrete microscopic model according to an exemplary embodiment. It should be noted that the straight steel fiber concrete and its corresponding optimization parameters are only used as an example of a fiber hybrid composite material, which can be set according to actual conditions and are not limited to the examples listed in this application.

[0058] Step S102, based on the grid file and the pre-configured simulation parameters, finite element calculation is performed by a finite element analysis method, and an explosion response evaluation index is obtained according to the result of the finite element calculation.

[0059] Optionally, the simulation parameters include material properties, boundary conditions, mutual contact conditions and explosion loads. The finite element analysis method includes performing finite element calculations using finite element analysis software such as LS-DYNA, ANSYS and ABAQUS. The explosion response evaluation index includes: deformation displacement, force and absorbed energy at the center of the rear panel of the fiber hybrid composite material.

[0060] Taking the fiber hybrid composite material as straight steel fiber concrete as an example, the *.K file obtained in step S101 is imported into the Ls-dyna software, and the material properties, boundary conditions, mutual contact conditions and explosion loads are set. And the finite element calculation is performed by the ConWep algorithm, and based on the calculation results, the central deformation displacement of the rear panel of the straight steel fiber concrete is used as the explosion response evaluation index. Among them, the ConWep algorithm can calculate the explosion load according to the position of the explosion source and the explosion parameters, and simulate the shock wave generated by the explosion. Assume that the explosive equivalent is 1.3kg and the distance from the straight steel fiber concrete is 40cm to simulate the explosion shock wave. It should be noted that the parameters such as the finite element calculation software, the finite element calculation algorithm, the position of the explosion source, the explosion parameters and the explosion response evaluation index involved in this example are only used as an example, which can be set according to the actual situation and are not limited to the examples listed in this application.

[0061] In one example, the explosion response evaluation index in step S102 includes: the deformation displacement, the force and the absorbed energy of the center of the rear panel of the fiber hybrid composite material. Optionally, the explosion response evaluation index can be at least one of the deformation displacement, the force and the absorbed energy of the center of the rear panel of the fiber hybrid composite material.

[0062] In one example, the simulation parameters in step S102 include: material properties, boundary conditions, mutual contact conditions and explosion load. Optionally, the material properties include fiber material properties and properties of each material in the mixed material. According to the specific physical properties of the material (such as elastic modulus, yield strength, Poisson's ratio, etc.), corresponding parameters are defined for each material. The mutual contact conditions generally include node-to-surface contact and friction coefficient. The parameters involved in the explosion load in this application generally include: explosion source location and explosion parameters, and the explosion parameters include explosion type (such as air explosion) or explosion energy, etc.

[0063] Step S103, based on the number of geometric parameters and explosion response evaluation indicators, the number of neurons of the back propagation neural network is determined, the geometric parameters and the corresponding explosion response evaluation indicators are used as training data, and the back propagation neural network is trained based on the training data to obtain a proxy model. Optionally, the proxy model established based on the back propagation neural network has a strong nonlinear solving ability, can accurately predict the explosion resistance of the fiber hybrid composite material, and solves the problem that the finite element calculation of the explosion resistance of the composite material takes a lot of time.

[0064] In one example, the back propagation neural network includes: an input layer, a hidden layer and an output layer, and step S103 includes:

[0065] Step S1031, determining the number of neurons in the input layer according to the number of geometric parameters.

[0066] Step S1032, determining the number of neurons in the output layer according to the number of explosion response evaluation indicators corresponding to the geometric parameters.

[0067] Step S1033, determining the number of neurons in the hidden layer based on the number of neurons in the input layer and the number of neurons in the output layer, the specific formula is as follows:

[0068]

[0069] Among them, h represents the number of neurons in the hidden layer, m represents the number of neurons in the input layer, and n represents the number of neurons in the output layer. Represents a constant between [1,10].

[0070] Optionally, taking the fiber hybrid composite material as straight steel fiber concrete as an example, the optimization parameters include: diameter, length and volume content of the straight steel fiber. The corresponding central deformation displacement of the rear panel of the straight steel fiber concrete is used as an explosion response evaluation index. Take the constant 3. Figure 3 is a neuron structure diagram of a back propagation neural network according to an exemplary embodiment, such as Figure 3 As shown, the number of neurons in the hidden layer is 5. It should be noted that the number of neurons in this example is only an example, which can be set according to actual conditions and is not limited to the examples listed in this application.

[0071] In one example, step S103 includes:

[0072] Step S1034, normalize the training data, and generate prediction results of the normalized data through a back propagation neural network.

[0073] Optionally, the training data is normalized, with the minimum and maximum values ​​in the input training data as the benchmarks 0 and 1 respectively, and the remaining data is scaled to fall within [0,1]. The normalization formula is as follows:

[0074]

[0075] Among them, x(i) represents the i-th data in the training data, x min represents the minimum value in the training data, x max represents the maximum value in the training data, and y(i) represents the i-th data after normalization.

[0076] Step S1035 , performing denormalization processing on the prediction results to restore the prediction results of the normalized data to the scale of the training data.

[0077] Optionally, the back propagation neural network obtains normalized values, so the prediction results need to be denormalized to restore the prediction results of the normalized data to the scale of the training data. The denormalization formula is as follows:

[0078]

[0079] Among them, x(i) represents the result of the denormalization of the i-th data, y(i) represents the i-th prediction result, and x min represents the minimum value in the training data, x max represents the maximum value in the training data,

[0080] Step S1036, iteratively update the neuron parameters of the back propagation network based on the LM algorithm to obtain a proxy model.

[0081] Optionally, the formula for updating neuron parameters by the LM algorithm is as follows:

[0082]

[0083] Among them, x k+1 Represents the weight and bias after iteration, x k Represents the current weight and bias, H represents the Hessian matrix, I represents the unit matrix, and J T represents the Jacobian matrix, μ represents the scalar factor, and e represents the error vector. The Hessian matrix is ​​approximated by the Jacobian matrix as follows:

[0084]

[0085] Where H represents the Hessian matrix, J T represents the Jacobian matrix.

[0086] Step S104, searching genetic geometric parameters from a preset fiber geometric parameter range by using a genetic algorithm, predicting the fitness of the genetic geometric parameters by using a proxy model, and iteratively solving the target geometric parameters based on the fitness.

[0087] Optionally, a genetic algorithm is used to simulate the biological natural selection process, and the target geometric parameter with the highest fitness is selected from the preset fiber geometric parameter range. In this way, by introducing an agent model to calculate the fitness value of the genetic individual, the optimal individual is obtained after cyclic iteration, thereby achieving the goal of optimizing the anti-explosion performance of the fiber hybrid composite material.

[0088] In one example, step S104 includes:

[0089] Step S1041, obtaining a preset fiber geometric parameter range, randomly generating an initial population that meets the fiber geometric parameter range, and each individual in the population represents the genetic geometric parameters of the fiber.

[0090] Optionally, an initial population is randomly generated, and each individual in the population represents a potential solution in the solution space, that is, the geometric parameters of the front position, and each individual is encoded as a chromosome using binary coding. Taking the fiber hybrid composite material as straight steel fiber concrete as an example, the optimization parameters include: the diameter, length and volume content of the straight steel fiber. Assume that the length of the straight steel fiber is between 20 and 60 mm, the diameter is between 0.3 and 0.9 mm, and the volume content is between 0 and 4%. It should be noted that the optimization parameters of the straight steel fiber in this example are only used as an example, and can be set according to actual conditions, and are not limited to the examples listed in this application.

[0091] Step S1042, calculating the fitness of each individual through the agent model, and the fitness represents the explosion response evaluation index.

[0092] Optionally, the fitness of each individual in the population is calculated by the proxy model, and the fitness value is predicted by the proxy model. The fitness value is used to evaluate the quality of individuals. According to the survival law of the fittest, individuals with large fitness values ​​will have a greater probability of being used to generate new individuals in subsequent calculations, and individuals with small fitness values ​​will be eliminated. Taking the fiber hybrid composite material as straight steel fiber concrete as an example, the fitness value represents the central deformation displacement of the rear panel of the straight steel fiber concrete.

[0093] Step S1043, iteratively updating individuals through selection operators, crossover operators, and mutation operators based on fitness to form an updated population.

[0094] Optionally, a selection operator is used to select individuals from the current population to determine which individuals can pass on their genetic information to the next generation. Selection operators include: roulette selection, tournament selection, ranking selection, and elite selection. The present application uses a roulette selection operator, wherein the individual with greater fitness has a greater probability of being selected. The crossover operator selects a crossover point from the individuals selected by random pairing and exchanges the gene fragments at the crossover point. The mutation operator randomly changes the genetic information of certain offspring individuals with a certain small probability, introducing new genetic information to increase the diversity of the population.

[0095] Step S1044, in response to satisfying the preset iteration condition, the iteration is stopped, and the individual with the highest fitness is used as the target geometric parameter.

[0096] Optionally, when the number of iterations reaches a preset number of iterations, or the fitness no longer significantly improves, the iteration is stopped and the optimal individual, i.e., the target geometric parameters, are output. Otherwise, the iteration continues to search for individuals with higher fitness. In this way, by introducing a proxy model to calculate the fitness value of the genetic individual, the optimal target geometric parameters are obtained after cyclic iterations, thereby improving the anti-explosion performance of the optimized fiber hybrid composite material.

[0097] Figure 4 is a schematic diagram showing a comparison of the damage area of ​​straight steel fiber concrete under explosion load before and after optimization according to an exemplary embodiment. Figure 4 As shown, after being optimized by the method of the present application, the damaged area of ​​the straight steel fiber concrete is smaller. Figure 5 FIG. 1 is a schematic diagram showing the central deformation displacement results of a straight steel fiber reinforced concrete rear panel according to an exemplary embodiment. Figure 5 As shown, after being optimized by the method of this application, the explosion resistance of straight steel fiber concrete is higher. Figure 4 and Figure 5 V f It represents volume content, L represents length, and D represents diameter.

[0098] In summary, the modeling software is used to perform modeling and automatically divide the mesh to obtain the mesh file to ensure that each area is properly divided, so as to accurately capture the effect of fiber on the mechanical properties of the hybrid material. The proxy model established based on the back propagation neural network has a strong nonlinear solving ability, which can accurately predict the explosion resistance of fiber hybrid composites and solve the problem that the finite element calculation of the explosion resistance of composite materials takes a lot of time. The genetic algorithm is used to simulate the biological natural selection process and select the target geometric parameters with the highest fitness from the preset fiber geometric parameter range. In this way, by introducing the proxy model to calculate the fitness value of the genetic individual, the optimal target geometric parameters are obtained after cyclic iteration, which improves the explosion resistance of the optimized fiber hybrid composite.

[0099] In a second aspect, the present application provides a fiber hybrid composite material explosion-proof structure optimization system. Figure 6 is a structural block diagram of a fiber hybrid composite material explosion-proof structure optimization system according to an embodiment of the present application, such as Figure 6 As shown, the system includes:

[0100] The geometric parameter module 100 is used to obtain geometric parameters of the fiber by using the Latin hypercube sampling method, establish a three-dimensional mesoscopic model based on the geometric parameters, and mesh the three-dimensional mesoscopic model to obtain a mesh file.

[0101] The response evaluation module 200 is used to perform finite element calculations by a finite element analysis method based on a grid file and pre-configured simulation parameters, and obtain explosion response evaluation indicators according to the results of the finite element calculations.

[0102] Training module 300: used to determine the number of neurons of the back propagation neural network based on the number of geometric parameters and explosion response evaluation indicators, use the geometric parameters and the corresponding explosion response evaluation indicators as training data, and train the back propagation neural network based on the training data to obtain a proxy model.

[0103] Solving module 400: used to search genetic geometric parameters from a preset fiber geometric parameter range through a genetic algorithm, predict the fitness of the genetic geometric parameters through an agent model, and iteratively solve the target geometric parameters based on the fitness.

[0104] In one example, the back propagation neural network in the training module 300 includes: an input layer, a hidden layer, and an output layer. The training module 300 includes:

[0105] The number of neurons in the input layer is determined according to the number of geometric parameters.

[0106] The number of neurons in the output layer is determined according to the number of explosion response evaluation indicators corresponding to the geometric parameters.

[0107] The number of neurons in the hidden layer is determined based on the number of neurons in the input layer and the number of neurons in the output layer. The specific formula is as follows:

[0108]

[0109] Among them, h represents the number of neurons in the hidden layer, m represents the number of neurons in the input layer, and n represents the number of neurons in the output layer. Represents a constant between [1,10].

[0110] In one example, the training module 300 includes:

[0111] The training data is normalized and the prediction results of the normalized data are generated through the back-propagation neural network.

[0112] The prediction results are denormalized to restore the prediction results of the normalized data to the scale of the training data.

[0113] The neuron parameters of the back-propagation network are iteratively updated based on the LM algorithm to obtain the proxy model.

[0114] In one example, the solution module 400 includes:

[0115] A preset fiber geometric parameter range is obtained, and an initial population satisfying the fiber geometric parameter range is randomly generated, wherein each individual in the population represents the genetic geometric parameter of the fiber.

[0116] The fitness of each individual is calculated through the agent model, and the fitness represents the evaluation index of explosion response.

[0117] Based on fitness, individuals are iteratively updated through selection operators, crossover operators, and mutation operators to form an updated population.

[0118] In response to meeting the preset iteration condition, the iteration is stopped and the individual with the highest fitness is taken as the target geometric parameter.

[0119] In one example, the geometric parameter module 100 includes: establishing a three-dimensional mesoscopic model based on geometric parameters through modeling software, and meshing the three-dimensional mesoscopic model through meshing software to obtain a mesh file, and the geometric parameters include: fiber shape, length, diameter, and volume content.

[0120] In one example, the explosion response evaluation index in the response evaluation module 200 includes: the deformation displacement, the force and the absorbed energy of the center of the rear panel of the fiber hybrid composite material.

[0121] In one example, simulation parameters in the response evaluation module 200 include: material properties, boundary conditions, mutual contact conditions, and blast loads.

[0122] In summary, the modeling software is used to perform modeling and automatically divide the mesh to obtain the mesh file to ensure that each area is properly divided, so as to accurately capture the effect of fiber on the mechanical properties of the hybrid material. The proxy model established based on the back propagation neural network has a strong nonlinear solving ability, which can accurately predict the explosion resistance of fiber hybrid composites and solve the problem that the finite element calculation of the explosion resistance of composite materials takes a lot of time. The genetic algorithm is used to simulate the biological natural selection process and select the target geometric parameters with the highest fitness from the preset fiber geometric parameter range. In this way, by introducing the proxy model to calculate the fitness value of the genetic individual, the optimal target geometric parameters are obtained after cyclic iteration, which improves the explosion resistance of the optimized fiber hybrid composite.

[0123] In a third aspect, an embodiment of the present application provides an electronic device, Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for optimizing the explosion-proof structure of a fiber hybrid composite material provided in the first aspect is implemented. Figure 7The electronic device 60 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0124] The electronic device 60 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 60 may include but are not limited to: at least one processor 61, at least one memory 62, and a bus 63 connecting different system components (including the memory 62 and the processor 61).

[0125] The bus 63 includes a data bus, an address bus, and a control bus.

[0126] The memory 62 may include a volatile memory, such as a random access memory (RAM) 621 and / or a cache memory 622 , and may further include a read-only memory (ROM) 623 .

[0127] The memory 62 may also include a program / utility 625 having a set (at least one) of program modules 624, such program modules 624 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0128] The processor 61 executes various functional applications and data processing by running the computer program stored in the memory 62, such as the method for optimizing the explosion-proof structure of fiber hybrid composite materials provided in the first aspect of the present application.

[0129] The electronic device 60 may also communicate with one or more external devices 64 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 65. Furthermore, the model-generated device 60 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 66. As shown, the network adapter 66 communicates with other modules of the model-generated device 60 via a bus 63. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the model-generated device 60, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

[0130] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0131] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a program stored thereon, and when the program is executed by a processor, the method for optimizing the explosion-proof structure of a fiber hybrid composite material provided in the first aspect is implemented.

[0132] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0133] In a possible embodiment, the present invention can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps of the method for optimizing the explosion-proof structure of fiber hybrid composite materials provided in the first aspect.

[0134] The program code for executing the present invention may be written in any combination of one or more programming languages, and may be executed entirely on a user device, partially on a user device, as an independent software package, partially on a user device and partially on a remote device, or entirely on a remote device.

[0135] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for optimizing explosion-proof structure of fiber hybrid composite material, characterized in that: include: The geometric parameters of the fiber are obtained by a Latin hypercube sampling method, a three-dimensional mesoscopic model is established based on the geometric parameters, and the three-dimensional mesoscopic model is meshed to obtain a mesh file; Based on the grid file and the pre-configured simulation parameters, a finite element calculation is performed by a finite element analysis method, and an explosion response evaluation index is obtained according to the result of the finite element calculation; Determine the number of neurons of the back propagation neural network based on the geometric parameters and the number of the explosion response evaluation indicators, use the geometric parameters and the corresponding explosion response evaluation indicators as training data, and train the back propagation neural network based on the training data to obtain a proxy model; Searching genetic geometric parameters from a preset fiber geometric parameter range by using a genetic algorithm, predicting the fitness of the genetic geometric parameters by using the proxy model, and iteratively solving the target geometric parameters based on the fitness, including: Obtaining a preset fiber geometric parameter range, randomly generating an initial population that meets the fiber geometric parameter range, each individual in the population characterizing the genetic geometric parameters of the fiber, The fitness of each individual is calculated by the agent model, and the fitness represents the explosion response evaluation index. Based on the fitness, the individuals are iteratively updated by selecting an operator, a crossover operator, and a mutation operator to form an updated population. In response to meeting a preset iteration condition, the iteration is stopped, and the individual with the highest fitness is used as the target geometric parameter.

2. The method for optimizing explosion-proof structure of fiber hybrid composite material according to claim 1, characterized in that: The back propagation neural network includes: an input layer, a hidden layer and an output layer. The method of determining the number of neurons of the back propagation neural network based on the number of the geometric parameters and the explosion response evaluation index includes: Determine the number of neurons in the input layer according to the number of the geometric parameters; Determining the number of neurons in the output layer according to the number of explosion response evaluation indicators corresponding to the geometric parameters; The number of neurons in the hidden layer is determined based on the number of neurons in the input layer and the number of neurons in the output layer. The specific formula is as follows: Among them, h represents the number of neurons in the hidden layer, m represents the number of neurons in the input layer, and n represents the number of neurons in the output layer. Represents a constant between [1,10].

3. The method for optimizing explosion-proof structure of fiber hybrid composite material according to claim 2, characterized in that: The step of training the back propagation neural network based on the training data to obtain a proxy model includes: Normalizing the training data, and generating prediction results of the normalized data through the back propagation neural network; Denormalizing the prediction results to restore the prediction results of the normalized data to the scale of the training data; The neuron parameters of the back-propagation neural network are iteratively updated based on the LM algorithm to obtain a proxy model.

4. The method for optimizing explosion-proof structure of fiber hybrid composite material according to claim 1, characterized in that: The step of establishing a three-dimensional mesoscopic model based on the geometric parameters and meshing the three-dimensional mesoscopic model to obtain a mesh file includes: The three-dimensional mesoscopic model is established based on the geometric parameters by modeling software, and the three-dimensional mesoscopic model is meshed by the meshing software to obtain the mesh file. The geometric parameters include: fiber shape, length, diameter, and volume content.

5. The method for optimizing explosion-proof structure of fiber hybrid composite material according to claim 1, characterized in that: The explosion response evaluation indexes include: deformation displacement, applied force and absorbed energy at the center of the rear panel of the fiber hybrid composite material.

6. The method for optimizing explosion-proof structure of fiber hybrid composite material according to claim 1, characterized in that: The simulation parameters include: material properties, boundary conditions, mutual contact conditions and explosion loads.

7. A fiber hybrid composite material explosion-proof structure optimization system, characterized in that: include: Geometric parameter module: used to obtain geometric parameters of the fiber by Latin hypercube sampling method, establish a three-dimensional mesoscopic model based on the geometric parameters, and mesh the three-dimensional mesoscopic model to obtain a mesh file; Response evaluation module: used for performing finite element calculation by finite element analysis method based on the grid file and pre-configured simulation parameters, and obtaining explosion response evaluation index according to the result of the finite element calculation; Training module: used to determine the number of neurons of the back propagation neural network based on the number of the geometric parameters and the explosion response evaluation index, use the geometric parameters and the corresponding explosion response evaluation index as training data, and train the back propagation neural network based on the training data to obtain a proxy model; Solving module: used to search genetic geometric parameters from a preset fiber geometric parameter range through a genetic algorithm, predict the fitness of the genetic geometric parameters through the proxy model, and iteratively solve the target geometric parameters based on the fitness, including: Obtaining a preset fiber geometric parameter range, randomly generating an initial population that meets the fiber geometric parameter range, each individual in the population characterizing the genetic geometric parameters of the fiber, The fitness of each individual is calculated by the agent model, and the fitness represents the explosion response evaluation index. Based on the fitness, the individuals are iteratively updated by selecting an operator, a crossover operator, and a mutation operator to form an updated population. In response to meeting a preset iteration condition, the iteration is stopped, and the individual with the highest fitness is used as the target geometric parameter.

8. An electronic device, characterized in that: include Memory, processor, and A computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for optimizing the explosion-proof structure of a fiber hybrid composite material as claimed in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for optimizing the explosion-proof structure of a fiber hybrid composite material according to any one of claims 1 to 6 is implemented.

Images