A method and system for predicting the flexural performance of steel tube concrete components reinforced with carbon fiber reinforced composite materials

Abstract

The present invention provides a method and system for predicting the flexural performance of carbon fiber reinforced composite material reinforced concrete-filled steel tube components, belonging to the field of high-performance composite material applications. The method comprises: establishing a finite element model; wherein the finite element model is used to simulate the debonding failure and fracture failure of the carbon fiber reinforced composite material in the carbon fiber reinforced composite material reinforced concrete-filled steel tube components; determining key parameters based on the finite element model; generating multiple data sets including debonding strain and ultimate bending moment based on the key parameters; and establishing and training a neural network model based on the multiple data sets so that the neural network model outputs target debonding strain and target ultimate bending moment in the flexural performance of the carbon fiber reinforced composite material reinforced concrete-filled steel tube components. The method provided by the present invention can accurately predict the debonding strain and ultimate bending moment of the carbon fiber reinforced composite material reinforced concrete-filled steel tube components under bending.

Description

Technical Field

[0001] The present invention belongs to the field of application of high-performance composite materials, and in particular relates to a method and system for predicting the bending performance of a steel tube concrete component reinforced with carbon fiber reinforced composite materials. Background Art

[0002] Steel tube concrete components are widely used in construction and bridge engineering due to their high bearing capacity and good ductility. However, with the increase of service life or changes in load conditions, such components are prone to insufficient bearing capacity due to material performance degradation or design defects, and are in urgent need of efficient reinforcement. Carbon fiber reinforced composite materials have become an important technical means to reinforce steel tube concrete structural components due to their advantages such as high strength, light weight and convenient construction. The reinforcement process usually adopts the wet pasting method, that is, the carbon fiber reinforced composite material is pasted on the surface of the steel tube concrete component to be reinforced with an adhesive. However, during the bending process, the steel tube concrete component reinforced with carbon fiber reinforced composite material may experience two complex failure modes: debonding failure between the carbon fiber reinforced composite material and the steel tube, or fracture failure of the carbon fiber reinforced composite material itself. These two complex failure modes are affected by the coupling of multiple parameters such as the yield strength of the steel, the number of carbon fiber reinforced composite material layers, and the bonding interface performance, making it extremely difficult to predict their mechanical behavior.

[0003] Current prediction methods mainly rely on simplified empirical formulas or traditional finite element simulations. Empirical formulas usually ignore the interaction between the nonlinear behavior of the interface and other dimensions and material parameters, resulting in conservative and inaccurate prediction results. Although traditional finite element analysis can partially simulate the failure process, it requires a large number of trial calculations and is computationally expensive, making it difficult to quickly respond to engineering needs. In addition, related schemes are mostly limited to single failure mode analysis and lack a dynamic characterization of the competition mechanism between debonding and fracture. Therefore, establishing an efficient and high-precision prediction method for the flexural performance of carbon fiber reinforced composite reinforced steel tube concrete components, especially accurately quantifying the quantitative relationship between each key parameter and the debonding strain and ultimate bending moment, has become the key to solving the bottleneck of carbon fiber reinforced composite reinforcement technology. Summary of the Invention

[0004] In view of the above problems, embodiments of the present application provide a method and system for predicting the flexural properties of carbon fiber reinforced composite material reinforced concrete-filled steel tube components, so as to overcome the above problems or at least partially solve the above problems.

[0005] In a first aspect, an embodiment of the present application provides a method for predicting the bending performance of a carbon fiber reinforced composite reinforced concrete-filled steel tube member, comprising:

[0006] Establishing a finite element model; wherein the finite element model is used to simulate the debonding failure and fracture failure of the carbon fiber reinforced composite material in the carbon fiber reinforced composite material reinforced steel tube concrete member;

[0007] Determine key parameters based on the finite element model; wherein the key parameters include steel yield strength, carbon fiber reinforced composite material bonding length, number of carbon fiber reinforced composite material layers, carbon fiber reinforced composite material elastic modulus, carbon fiber reinforced composite material tensile strength, bonding interface tangential fracture energy, bonding interface tangential bonding strength, steel pipe diameter-to-thickness ratio, and steel pipe diameter;

[0008] Based on the key parameters, generating a plurality of data sets including debonding strain and ultimate bending moment;

[0009] Based on the multiple data sets, a neural network model is established and trained so that the neural network model outputs the target debonding strain and target ultimate bending moment in the bending performance of the carbon fiber reinforced composite material reinforced steel tube concrete member.

[0010] Furthermore, the establishing of the finite element model includes:

[0011] Obtaining a three-dimensional model and material parameters of the carbon fiber reinforced composite material reinforced steel tube concrete component; wherein the material parameters include at least: steel parameters, carbon fiber reinforced composite material parameters, concrete parameters and bonding interface parameters;

[0012] Based on the three-dimensional model and the material parameters, boundary conditions and loads are applied to establish the finite element model;

[0013] The boundary conditions represent the constraint state of the finite element model, and the load represents the external force acting on the finite element model.

[0014] Furthermore, determining key parameters based on the finite element model includes:

[0015] Based on the finite element model, changing the size of the three-dimensional model and the value of the material parameter, and respectively obtaining the degree of influence of the three-dimensional model and the material parameter on the performance of the carbon fiber reinforced composite material reinforced steel tube concrete component;

[0016] Based on the degree of influence, the key parameters are determined.

[0017] Furthermore, the generating of multiple data sets including debonding strain and ultimate bending moment based on the key parameters includes:

[0018] Performing parameterization processing on the key parameters so that the key parameters vary within a preset range;

[0019] changing the value of the key parameter, running the finite element model, and obtaining the debonding strain and the ultimate bending moment corresponding to each running of the finite element model;

[0020] The debonding strain and the ultimate bending moment obtained each time are respectively associated with the values of the corresponding key parameters to generate a plurality of the data sets.

[0021] Furthermore, the establishing and training of a neural network model based on the plurality of data sets so that the neural network model outputs the target debonding strain and target ultimate bending moment in the bending performance of the carbon fiber reinforced composite reinforced steel tube concrete member includes:

[0022] Normalizing the plurality of data sets, and dividing the normalized data sets into a training set and a test set;

[0023] Constructing an initial neural network model; wherein the initial neural network model includes: an input layer, a hidden layer, and an output layer, the input layer receives the key parameters, the output layer outputs the debonding strain or the ultimate bending moment, and the hidden layer is used to connect the input layer and the output layer;

[0024] Based on the training set and the test set, minimizing the mean square error and maximizing the coefficient of determination as the solution goal, performing multiple iterative solutions to obtain the trained neural network model;

[0025] Based on the neural network model, the target debonding strain and the target ultimate bending moment are output.

[0026] Furthermore, the method of performing multiple iterations based on the training set and the test set with the goal of minimizing the mean square error and maximizing the coefficient of determination to obtain the trained neural network model includes the following steps:

[0027] Step 1: Initialize the weights and biases in the initial neural network model;

[0028] Step 2: Inputting the training set into the initial neural network model, obtaining the predicted output and performing denormalization processing, and determining the mean square error and the coefficient of determination based on the predicted output and the actual output after the denormalization processing;

[0029] Step 3: Based on the mean square error, update the weight and the bias through a back propagation algorithm to minimize the mean square error;

[0030] Step 4: During the training process, the neural network model is validated using the test set to maximize the determination coefficient;

[0031] Step 5: Repeat the iterative training process from step 2 to step 4 until the neural network model reaches a preset maximum number of iterations or the model converges.

[0032] Furthermore, outputting the target debonding strain and the target ultimate bending moment based on the neural network model includes:

[0033] Based on the neural network model, obtaining updated target weights and target biases;

[0034] Based on the target weight and the target bias, the target debonding strain and the target ultimate bending moment are output in combination with a double S-type transfer function and a linear transfer function.

[0035] Furthermore, the hidden layer adopts the double S-shaped transfer function, and the output layer adopts the linear transfer function.

[0036] Furthermore, after establishing the finite element model, the method further includes:

[0037] The finite element model is verified, wherein the verification of the finite element model comprises the following steps:

[0038] Comparing the calculation results of the finite element model with the test results, wherein the calculation results and the test results include a bending moment-mid-span deflection curve and a strain distribution curve;

[0039] When the calculation result is consistent with the test result, the finite element model is determined to have passed the verification.

[0040] In a second aspect of the present application, a system for predicting the bending performance of a carbon fiber reinforced composite reinforced concrete-filled steel tube member is provided, comprising:

[0041] Establishing a unit for establishing a finite element model; wherein the finite element model is used to simulate the debonding failure and fracture failure of the carbon fiber reinforced composite material in the carbon fiber reinforced composite material reinforced steel tube concrete member;

[0042] a determination unit, configured to determine key parameters based on the finite element model; wherein the key parameters include steel yield strength, carbon fiber reinforced composite material bonding length, number of carbon fiber reinforced composite material layers, carbon fiber reinforced composite material elastic modulus, carbon fiber reinforced composite material tensile strength, bonding interface tangential fracture energy, bonding interface tangential bonding strength, steel pipe diameter-to-thickness ratio, and steel pipe diameter;

[0043] a generating unit, configured to generate a plurality of data sets including debonding strain and ultimate bending moment based on the key parameters;

[0044] An output unit is used to establish and train a neural network model based on the multiple data sets, so that the neural network model outputs the target debonding strain and target ultimate bending moment in the bending performance of the carbon fiber reinforced composite material reinforced steel tube concrete member.

[0045] Furthermore, the establishing unit includes:

[0046] a determining unit, configured to change the size of the three-dimensional model and the value of the material parameter based on the finite element model, and respectively obtain the degree of influence of the three-dimensional model and the material parameter on the performance of the carbon fiber reinforced composite material reinforced steel tube concrete component;

[0047] Based on the degree of influence, the key parameters are determined.

[0048] Furthermore, the generating unit includes:

[0049] a first processing unit, configured to perform parameterization processing on the key parameter so that the key parameter changes within a preset range;

[0050] changing the value of the key parameter, running the finite element model, and obtaining the debonding strain and the ultimate bending moment corresponding to each running of the finite element model;

[0051] The debonding strain and the ultimate bending moment obtained each time are respectively associated with the values of the corresponding key parameters to generate a plurality of the data sets.

[0052] Furthermore, the output unit includes:

[0053] a second processing unit, configured to normalize the plurality of data sets and divide the normalized data sets into a training set and a test set;

[0054] Constructing an initial neural network model; wherein the initial neural network model includes: an input layer, a hidden layer, and an output layer, the input layer receives the key parameters, the output layer outputs the debonding strain or the ultimate bending moment, and the hidden layer is used to connect the input layer and the output layer;

[0055] Based on the training set and the test set, minimizing the mean square error and maximizing the coefficient of determination as the solution goal, performing multiple iterative solutions to obtain the trained neural network model;

[0056] Based on the neural network model, the target debonding strain and the target ultimate bending moment are output.

[0057] Furthermore, the second processing unit includes the following steps:

[0058] Step 1: Initialize the weights and biases in the initial neural network model;

[0059] Step 2: Inputting the training set into the initial neural network model, obtaining the predicted output and performing denormalization processing, and determining the mean square error and the coefficient of determination based on the predicted output and the actual output after the denormalization processing;

[0060] Step 3: Based on the mean square error, update the weight and the bias through a back propagation algorithm to minimize the mean square error;

[0061] Step 4: During the training process, the neural network model is validated using the test set to maximize the determination coefficient;

[0062] Step 5: Repeat the iterative training process from step 2 to step 4 until the neural network model reaches a preset maximum number of iterations or the model converges.

[0063] Furthermore, the second processing unit includes:

[0064] A first acquisition unit, configured to acquire an updated target weight and target bias based on the neural network model;

[0065] Based on the target weight and the target bias, the target debonding strain and the target ultimate bending moment are output in combination with a double S-type transfer function and a linear transfer function.

[0066] Furthermore, the hidden layer adopts the double S-shaped transfer function, and the output layer adopts the linear transfer function.

[0067] Furthermore, after establishing the finite element model, the system further includes:

[0068] A verification unit is used to verify the finite element model, wherein the verification of the finite element model includes the following steps:

[0069] Comparing the calculation results of the finite element model with the test results, wherein the calculation results and the test results include a bending moment-mid-span deflection curve and a strain distribution curve;

[0070] When the calculation result is consistent with the test result, the finite element model is determined to have passed the verification.

[0071] Therefore, the present embodiment provides a method for predicting the bending performance of carbon fiber reinforced composite reinforced concrete-filled steel tube components. A finite element model can be established to simulate the complex failure mode of carbon fiber reinforced composite reinforced concrete-filled steel tube components during bending, including debonding failure and fracture failure of carbon fiber reinforced composite materials. Secondly, the key parameters are determined based on the finite element model, and the yield strength of steel, the bonding length of carbon fiber reinforced composite materials, the number of carbon fiber reinforced composite materials, the elastic modulus of carbon fiber reinforced composite materials, the tensile strength of carbon fiber reinforced composite materials, the tangential fracture energy of the bonding interface, the tangential bonding strength of the bonding interface, the diameter of the steel pipe, and the bending performance of the steel pipe can be predicted. Thickness ratio and steel tube diameter can generate multiple data sets including debonding strain and ultimate bending moment, and use multiple data sets to establish and train a neural network model. The neural network model can learn the complex nonlinear relationship between input parameters and output results, thereby outputting the target debonding strain and target ultimate bending moment of the bending performance of carbon fiber reinforced composite material reinforced steel tube concrete components. This method overcomes the shortcomings of traditional methods that require a large amount of experimental data and complex calculation processes, and can accurately predict the debonding strain and ultimate bending moment of carbon fiber reinforced composite material reinforced steel tube concrete components during bending, solving the problem of difficulty in predicting the mechanical properties of components under complex failure mode conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0073] Figure 1 This is a flowchart of the steps of a method for predicting the bending performance of a carbon fiber reinforced composite material reinforced steel tube concrete member provided in an embodiment of the present application;

[0074] Figure 2 is a schematic diagram of a finite element model provided in an embodiment of the present application;

[0075] Figure 3 Schematic diagram of the quantity distribution of a debonding strain finite element model provided in an embodiment of the present application;

[0076] Figure 4 Schematic diagram of the quantity distribution of a limit bending moment finite element model provided in an embodiment of the present application;

[0077] Figure 5 (a) is a schematic diagram of a sensitivity analysis of debonding strain to parameter changes provided in an embodiment of the present application;

[0078] Figure 5(b) is a schematic diagram of a sensitivity analysis of the ultimate bending moment to parameter changes provided in an embodiment of the present application;

[0079] Figure 6 This is a schematic diagram of an optimal number of hidden layer neurons provided by an embodiment of the present application;

[0080] Figure 7 This is a schematic diagram of the structure of an initial neural network model provided in an embodiment of the present application;

[0081] Figure 8 This is a schematic diagram of a training process of a neural network model provided in an embodiment of the present application;

[0082] Figure 9 (a) is for Figure 8 A schematic diagram of the performance of a debonding strain neural network model is provided;

[0083] Figure 9 (b) is for Figure 8 A schematic diagram of the performance of a neural network model of ultimate bending moment is provided;

[0084] Figure 10 (a) is a schematic diagram showing the comparison results of the debonding strain predicted by a neural network model provided in an embodiment of the present application and the target value in a test set;

[0085] Figure 10 (b) is a schematic diagram showing the comparison results of the debonding strain predicted by a neural network model provided in an embodiment of the present application and the target value in the validation set;

[0086] Figure 10 (c) is a schematic diagram showing the comparison results of the debonding strain predicted by a neural network model provided in an embodiment of the present application and the target value in the total set;

[0087] Figure 11 (a) is a schematic diagram showing the comparison results of the ultimate bending moment predicted by a neural network model and the target value in a test set provided in an embodiment of the present application;

[0088] Figure 11 (b) is a schematic diagram showing the comparison results of the ultimate bending moment predicted by a neural network model and the target value in a validation set provided in an embodiment of the present application;

[0089] Figure 11 (c) is a schematic diagram showing the comparison results of the ultimate bending moment predicted by a neural network model and the target value in the total set provided in an embodiment of the present application;

[0090] Figure 12 (a) is a schematic diagram showing the comparison results of the bending moment-mid-span deflection curves of a carbon fiber reinforced composite material debonding failure provided in an embodiment of the present application;

[0091] Figure 12 (b) is a schematic diagram showing the comparison results of the bending moment-mid-span deflection curves of a carbon fiber reinforced composite material fracture failure provided in an embodiment of the present application;

[0092] Figure 13 (a) is a schematic diagram showing the comparison results of strain distribution curves of debonding failure of a carbon fiber reinforced composite material provided in an embodiment of the present application;

[0093] Figure 13 (b) is a schematic diagram showing the comparison results of strain distribution curves of fracture failure of a carbon fiber reinforced composite material provided in an embodiment of the present application;

[0094] Figure 14 This is a system diagram for predicting the bending performance of steel tube concrete components reinforced with carbon fiber reinforced composite materials, provided in an embodiment of the present application. DETAILED DESCRIPTION

[0095] The exemplary embodiments of the present application will be described in more detail below in conjunction with the accompanying drawings in the embodiments of the present application. Although the accompanying drawings show exemplary embodiments of the present application, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0096] In the related art, carbon fiber reinforced composite reinforced concrete-filled steel tube components may experience complex failure modes during bending, including debonding and fracture failure of the carbon fiber reinforced composite. These failure modes are influenced by multiple factors, such as material properties, geometric dimensions, and bonding interface properties, making the prediction of their bending performance complex and challenging. Traditional prediction methods may not fully consider all influencing factors and their interactions, resulting in inaccurate or unreliable prediction results, and may require a large amount of experimental data and complex calculation processes, which is inefficient.

[0097] In view of this, this embodiment provides a method for predicting the flexural performance of carbon fiber reinforced composite (CFRP) reinforced concrete-filled steel tube (CFTU) components. By establishing a finite element model, the mechanical behavior of CFRP reinforced concrete-filled steel tube (CFTU) components during bending is simulated, including debonding and fracture failure of the CFRP components. The finite element model can account for the influence of multiple factors, such as material nonlinearity and geometric nonlinearity, thereby improving prediction accuracy. Based on the finite element model, key parameters influencing the flexural performance of CFRP reinforced concrete-filled steel tube (CFTU) components are determined. These parameters include steel yield strength, CFRP bond length, number of CFRP layers, CFRP elastic modulus, CFRP tensile strength, tangential fracture energy at the bonding interface, tangential bond strength at the bonding interface, steel tube diameter-to-thickness ratio, and steel tube diameter. By identifying these key parameters, a more comprehensive understanding of the component's flexural performance is achieved. Based on these key parameters, multiple data sets, including debonding strain and ultimate bending moment, are generated. These data sets can be obtained through finite element analysis or experiments, providing rich data support for subsequent neural network training. Finally, a neural network model is established and trained using the generated data sets. The neural network model can learn the complex nonlinear relationship between input parameters (key parameters) and output results (debonding strain and ultimate bending moment), thereby realizing rapid and accurate prediction of the flexural performance of carbon fiber reinforced composite reinforced concrete-filled steel tube members to solve the above problems.

[0098] Reference Figure 1 , Figure 1 This is a flowchart of a method for predicting the bending performance of a carbon fiber reinforced composite material reinforced steel tube concrete member provided in an embodiment of the present application. Figure 1 It can be seen that including:

[0099] Step S101: establishing a finite element model; wherein the finite element model is used to simulate the debonding failure and fracture failure of the carbon fiber reinforced composite material in the carbon fiber reinforced composite material reinforced concrete-filled steel tube member.

[0100] In this embodiment, debonding failure refers to the destruction of the bonding interface between the carbon fiber reinforced composite material and the steel pipe, resulting in a loss of adhesion between the carbon fiber reinforced composite material and the steel pipe, and the carbon fiber reinforced composite material falling off the surface of the steel pipe. Debonding failure usually occurs when the bonding strength between the carbon fiber reinforced composite material and the steel pipe is insufficient. Debonding failure affects the overall bearing capacity and deformation capacity of the carbon fiber reinforced composite material reinforced steel tube concrete component. Specifically, during bending, the bonding interface between the carbon fiber reinforced composite material and the steel pipe slips and debonds, ultimately causing the carbon fiber reinforced composite material to fall off the surface of the steel pipe. Fracture failure refers to the fracture of the carbon fiber reinforced composite material itself, that is, the carbon fiber reinforced composite material reaches its ultimate tensile strength during tension and breaks. Fracture failure usually occurs when the strength of the carbon fiber reinforced composite material is insufficient or there are local defects. Fracture failure will cause the bearing capacity of the component to drop sharply. During bending, the carbon fiber reinforced composite material reaches its ultimate tensile strength, resulting in obvious cracks and eventual fracture.

[0101] Debonding failure and fracture failure are both failure modes for carbon fiber reinforced composite reinforced concrete-filled steel tube components. To better predict the performance of carbon fiber reinforced composite reinforced concrete-filled steel tube components in actual engineering applications, it is necessary to analyze these two failure modes. Therefore, a finite element model can be established to simulate the occurrence of these two failure modes and evaluate their impact on the performance of carbon fiber reinforced composite reinforced concrete-filled steel tube components.

[0102] In this embodiment, in order to ensure that the finite element model can simulate in detail the mechanical behavior (debonding failure and fracture failure) of the CFRP-reinforced CFRP steel tube concrete-filled member during bending, the established finite element model needs to consider multiple factors such as the geometric shape, material properties, bonding interface characteristics, and load conditions of the CFRP-reinforced CFRP steel tube concrete-filled member, so as to make the debonding failure and fracture failure simulated by the finite element model more accurate.

[0103] Step S102: Determine key parameters based on the finite element model; wherein the key parameters include steel yield strength, carbon fiber reinforced composite material bonding length, number of carbon fiber reinforced composite material layers, carbon fiber reinforced composite material elastic modulus, carbon fiber reinforced composite material tensile strength, bonding interface tangential fracture energy, bonding interface tangential bonding strength, steel pipe diameter-to-thickness ratio, and steel pipe diameter.

[0104] In this example, a finite element model was used to individually change the specific value of each parameter while keeping other parameters constant. The impact of each parameter change on the performance of CFRP-reinforced concrete-filled steel tube components was analyzed, particularly on debonding and fracture failure. For example, increasing the yield strength of the steel can enhance the component's bending resistance but may increase the risk of debonding in the CFRP. Increasing the bond length of the CFRP can improve adhesion and reduce the risk of debonding. Each parameter that may affect the performance of CFRP-reinforced concrete-filled steel tube components was analyzed, and the parameters with the most significant performance impact were identified as key parameters, namely, the yield strength of the steel, the bond length of the CFRP, the number of CFRP layers, the elastic modulus of the CFRP, the tensile strength of the CFRP, the tangential fracture energy of the bonding interface, the tangential bond strength of the bonding interface, the steel tube diameter-to-thickness ratio, and the steel tube diameter. In addition, in addition to the single parameter analysis, the simultaneous variation of multiple parameters can also be considered. Specifically, by combining different parameter values, the performance of the component under conditions of multiple parameter variations can be simulated.

[0105] Step S103: Based on the key parameters, generate multiple data sets including debonding strain and ultimate bending moment.

[0106] In this embodiment, the key parameters that significantly affect the performance of carbon fiber reinforced composite reinforced concrete-filled steel tube components are selected, namely, steel yield strength, carbon fiber reinforced composite bonding length, carbon fiber reinforced composite layer number, carbon fiber reinforced composite elastic modulus, carbon fiber reinforced composite tensile strength, bonding interface tangential fracture energy, bonding interface tangential bonding strength, steel tube diameter-to-thickness ratio, and steel tube diameter. Then, according to the actual engineering requirements and the range of values that can be explained by the constitutive model, a reasonable value range is set for each parameter. Using the established finite element model, the specific values of each key parameter are changed one by one, and multiple finite element simulation analyses are performed. In each simulation, key performance indicators such as the debonding strain and ultimate bending moment of the component during bending are recorded, and the debonding strain and ultimate bending moment data obtained from each finite element simulation are collected and sorted to form a data set. Each data set contains the debonding strain and ultimate bending moment values under different parameter combinations, as well as the corresponding key parameter values.

[0107] Step S104: establishing and training a neural network model based on the plurality of data sets, so that the neural network model outputs a target debonding strain and a target ultimate bending moment in the bending performance of the carbon fiber reinforced composite material reinforced steel tube concrete member.

[0108] In this embodiment, multiple data sets containing debonding strains and ultimate bending moments under different key parameter combinations have been obtained in step S103, and each data set contains the values of the key parameters and the corresponding debonding strain and ultimate bending moment results. Then, normalization processing is performed on multiple data sets, and data of different units and scales are converted to the range of [-1, 1] to improve the training efficiency and accuracy of the neural network. Secondly, an initial neural network model is constructed. The data set is divided into a training set and a test set, and the key parameters are used as input, and the debonding strain and ultimate bending moment corresponding to the key parameters are used as output to train the initial neural network model. During the training process, the weights and bias values are iteratively optimized until the maximum number of iterations is reached or the algorithm converges, and finally the trained neural network model is obtained, and the neural network model is used to output the target debonding strain and target ultimate bending moment in the bending performance of carbon fiber reinforced composite reinforced steel tube concrete components.

[0109] In summary, by establishing a finite element model, the complex failure modes of CFRP reinforced concrete-filled steel tube components during bending can be simulated, including debonding failure and fracture failure of CFRP. Secondly, the key parameters are determined based on the finite element model, and by changing the steel yield strength, CFRP bonding length, number of CFRP layers, CFRP elastic modulus, CFRP tensile strength, tangential fracture energy of the bonding interface, tangential bonding strength of the bonding interface, steel tube diameter-to-thickness ratio, and steel tube diameter among the key parameters, multiple data sets including debonding strain and ultimate bending moment can be generated. The neural network model is established and trained using multiple data sets. The neural network model can learn the complex nonlinear relationship between the input parameters and the output results, thereby outputting the target debonding strain and target ultimate bending moment of the bending performance of CFRP reinforced concrete-filled steel tube components. This method overcomes the shortcomings of traditional methods that require a large amount of experimental data and complex calculation processes, and can accurately predict the debonding strain and ultimate bending moment of CFRP reinforced concrete-filled steel tube components during bending, solving the problem of difficulty in predicting complex failure modes.

[0110] In a specific embodiment, the establishing of the finite element model includes: obtaining a three-dimensional model and material parameters of the carbon fiber reinforced composite material reinforced steel tube concrete component; wherein the material parameters include at least: steel parameters, carbon fiber reinforced composite material parameters, concrete parameters and bonding interface parameters; based on the three-dimensional model and the material parameters, applying boundary conditions and loads to establish the finite element model; wherein the boundary conditions represent the constraint state of the finite element model, and the loads represent the external forces acting on the finite element model.

[0111] In this embodiment, the three-dimensional model represents the dimensions of an actual carbon fiber reinforced composite reinforced concrete-filled steel tube component, specifically geometric parameters such as the steel tube's diameter, thickness, and length, as well as the number of carbon fiber reinforced composite layers, thickness, and bonding length. Steel parameters may include elastic modulus, ultimate strength, and yield strength. Carbon fiber reinforced composite parameters may include elastic modulus, yield strength, ultimate strength, fracture energy, tensile strength, and compressive strength. Concrete parameters may include ultimate tensile strength and fracture energy. Bonding interface parameters may include elastic modulus and fracture energy of the interface. Based on the three-dimensional model and material parameters, finite element software is used to create a geometric model of the carbon fiber reinforced composite reinforced concrete-filled steel tube component. This geometric model includes the steel tube, concrete, and carbon fiber reinforced composite layers. Boundary conditions represent the constraints of the finite element model, which may include fixed constraints, displacement constraints, and symmetry constraints. The choice of boundary conditions can be tailored to the actual project and is not limited in this embodiment. Loads represent the external forces acting on the finite element model. External forces can be concentrated forces, distributed forces, bending moments, and pressures. When applying loads, they are determined based on the actual conditions of the CFRP-reinforced concrete-filled steel tubular member. For example, for bending members, bending moments or transverse loads can be applied to simulate their stress conditions. Therefore, by applying fixed constraints to specific portions of the geometric model (such as one end), and applying concentrated forces, distributed forces, and bending moments to specific locations or surfaces, boundary conditions and loads are applied, and a finite element model is established.

[0112] Secondly, there may be symmetry in the geometric shape, material properties and boundary conditions of the CFRP reinforced steel tube concrete components. Therefore, if the components are axially symmetrical and cross-sectionally symmetrical, it is possible to cut them on the symmetry plane and retain 1 / 4 of the part for modeling and analysis. Therefore, the construction of the finite element model can refer to Figure 2 To achieve, Figure 2 This is a schematic diagram of a finite element model provided in an embodiment of the present application, Figure 2 As can be seen, a 1 / 4 component model was established, taking into account the symmetry of geometry, materials, and boundary conditions. Symmetry planes 1 and 2 were set for this purpose. Rigid support lines and loading lines were used to simulate the specimen support and loading boundaries. The former employed fixed constraints, while the latter released displacement along the y-direction. The cohesive layer mesh in the end region of the CFRP was refined to simulate the debonding between the CFRP and the steel. To reduce the hourglass effect caused by this element, four layers of mesh were laid across the thickness of the steel pipe.

[0113] In a specific embodiment, determining the key parameters based on the finite element model includes: changing the size of the three-dimensional model and the value of the material parameter based on the finite element model, and respectively obtaining the degree of influence of the three-dimensional model and the material parameters on the performance of the carbon fiber reinforced composite material reinforced steel tube concrete component; and determining the key parameters based on the degree of influence.

[0114] In this embodiment, according to the finite element model, the dimensions of the three-dimensional model, such as the steel pipe diameter Ds, the steel pipe wall thickness ts, the carbon fiber reinforced composite material bonding length Lcf, etc., and material parameters, such as the steel yield strength fy, the concrete cubic compressive strength fcu, the carbon fiber reinforced composite material elastic modulus Ecf, and the carbon fiber reinforced composite material tensile strength fcf, are defined as variable parameters. Changes in the dimensions and material parameters of the three-dimensional model may affect the performance of the carbon fiber reinforced composite material reinforced steel tube concrete component. Therefore, the variation range of each parameter can be set according to the actual engineering application scenario and the applicable range of material properties. For example, the steel pipe diameter Ds ranges from 100 to 1200 mm; the steel pipe diameter-to-thickness ratio Ds / ts ranges from 20 to 100; the steel pipe yield strength fy ranges from 200 to 800 MPa; the concrete cubic compressive strength fcu ranges from 30 to 70 MPa; the carbon fiber reinforced composite material elastic modulus Ecf ranges from 100 to 500 GPa; the tensile strength fcf of carbon fiber reinforced composite materials ranges from 1000 to 5000 MPa; the number of carbon fiber reinforced composite layers ncf ranges from 2 to 21; the maximum normal traction stress t1 ranges from 5 to 45 MPa; the pure mode I interface fracture energy GⅠ ranges from 0.01 to 0.2 mJ / mm2; the maximum tangential traction stress ts0 ranges from 10 to 30 MPa; the pure mode II interface fracture energy GⅡ ranges from 1 to 5 mJ / mm2; the ratio of the shear section length Lcf,s of the carbon fiber reinforced composite material to the shear section length Ls of the component Lcf,s / Ls ranges from 1 / 6 to 2 / 3. Based on the changes in geometric and material parameters, a large number of finite element models can be constructed, referring to Figure 3 and Figure 4 , Figure 3 : is a schematic diagram of the quantity distribution of a debonding strain finite element model provided in an embodiment of the present application, Figure 4 is a schematic diagram of the quantity distribution of a limit bending moment finite element model provided in an embodiment of the present application, from Figure 3 and Figure 4 As can be seen from the figure, the finite element model has a wide range of geometric and material properties. A parametric analysis was then performed based on the debonding strain finite element model and the ultimate bending moment finite element model to determine the degree to which the 3D model and material parameters affect the performance of CFRP-reinforced concrete-filled steel tubular components. Finally, based on this degree of influence, the most influential parameters were selected as key parameters.

[0115] Assuming that the basic parameters are Ds=100 mm, Ds / ts=40, fy=400 MPa, fcu=40 MPa, Ecf=200 GPa, fcf=3000 MPa, ncf=3, t1=30 MPa, GⅠ=0.15 mJ / mm2, ts0=20 MPa, GⅡ=3 mJ / mm2 and Lcf,s / Ls=1 / 3, refer to Figure 5 (a) with Figure 5 (b), Figure 5 (a) is a schematic diagram of a sensitivity analysis of debonding strain to parameter changes provided in an embodiment of the present application. Figure 5 (b) is a schematic diagram of a sensitivity analysis of the ultimate bending moment to parameter changes provided in an embodiment of the present application; Figure 5 (a) with Figure 5 The horizontal axis in (b) represents the parameter level, which increases from left to right. Figure 5 (a) with Figure 5 As shown in (b), the minimum values of t1 and GⅠ within the current parameter range are 5 MPa and 0.01 mJ / mm2, respectively, far below their normal values of 33.4 MPa and 0.184 mJ / mm2. However, these values still have no effect on the ultimate bending moment. This indicates that the parameters t1 and GⅠ, which define the normal traction-separation response, have little effect on the debonding strain and ultimate bending moment. Furthermore, the debonding strain and ultimate bending moment are also insensitive to changes in the concrete cube compressive strength fcu. Therefore, the maximum normal traction stress t1, the pure mode I interfacial fracture energy GⅠ, and the concrete cube compressive strength fcu are not considered key parameters for further analysis.

[0116] In summary, the yield strength of steel, the bonding length of CFRP, the number of CFRP layers, the elastic modulus of CFRP, the tensile strength of CFRP, the tangential fracture energy of the bonding interface, the tangential bonding strength of the bonding interface, the diameter-to-thickness ratio of the steel pipe and the diameter of the steel pipe all have a significant impact on the performance of CFRP-reinforced concrete-filled steel tube components and can be used as key parameters.

[0117] In a specific embodiment, generating multiple data sets including debonding strain and ultimate bending moment based on the key parameters includes: parameterizing the key parameters so that the key parameters vary within a preset range; changing the values of the key parameters, running the finite element model, and obtaining the debonding strain and ultimate bending moment corresponding to each run of the finite element model; and associating the debonding strain and ultimate bending moment obtained each time with the corresponding values of the key parameters to generate the multiple data sets.

[0118] In this embodiment, to ensure that the parameter ranges of key parameters are consistent with actual engineering applications and to avoid setting unrealistic extreme values, the key parameters are parameterized. The purpose of parameterization is to systematically vary these key parameters within a preset range to generate different combinations of parameter values for subsequent finite element analysis. The key parameter values are then varied within the preset range, and the finite element model is run. The debonding strain and ultimate bending moment corresponding to each finite element model run are then obtained. The debonding strain and ultimate bending moment obtained for each run are then associated with the corresponding key parameter values to generate multiple data sets. Each data set includes the key parameters and the debonding strain and ultimate bending moment associated with their values.

[0119] In a specific embodiment, the neural network model is established and trained based on the multiple data sets so that the neural network model outputs the target debonding strain and target ultimate bending moment in the bending performance of the carbon fiber reinforced composite material reinforced steel tube concrete member, including: normalizing the multiple data sets and dividing the multiple data sets after normalization into a training set and a test set; constructing an initial neural network model; wherein the initial neural network model includes: an input layer, a hidden layer and an output layer, the input layer receives the key parameters, the output layer outputs the debonding strain or the ultimate bending moment, and the hidden layer is used to connect the input layer and the output layer; based on the training set and the test set, with minimizing the mean square error and maximizing the coefficient of determination as the solution goal, multiple iterative solutions are performed to obtain the trained neural network model; based on the neural network model, the target debonding strain and the target ultimate bending moment are output.

[0120] In this embodiment, because different data have different units and scales, multiple datasets are first normalized to ensure that the data in the datasets are within the same scale range. For example, variables are restricted to the range [−1, 1]. The normalized datasets are then divided into training and test sets. For example, the training set accounts for 80% of the total dataset, and the test set accounts for 20%. This embodiment does not specify the ratio of the training set to the test set and can be set based on actual project conditions.

[0121] The ultimate bending moment and debonding strain of steel tube concrete components reinforced with carbon fiber reinforced composite materials are determined by key parameters. Due to the complexity of the determination mechanism, a neural network model can be used to predict the ultimate bending moment and debonding strain of steel tube concrete components reinforced with carbon fiber reinforced composite materials. The neural network model in this embodiment can be an ANN (Artificial Neural Network) model. Information propagation in the neural network model is performed by a link that receives information from a processing unit (neuron) and passes it to the next neuron. Each piece of information is affected by a weight, reflecting the importance of the input variable to the output. Once a neuron receives information, it merges it with other information from different neurons through a combination function. The combined information is then transmitted to the next unit. This iterative process is repeated until the algorithm accurately fits the data, or stops when the maximum number of iterations is reached. Reference Figure 7 , Figure 7 This is a schematic diagram of the structure of an initial neural network model provided in an embodiment of the present application. Figure 7 As can be seen from the figure, the structure of the initial neural network model is usually composed of three types of layers: input layer, hidden layer and output layer. In this embodiment, based on the training set and the test set, with the goal of minimizing the mean square error and maximizing the coefficient of determination, multiple iterations are performed to obtain the trained neural network model. The input layer is the input parameter for model training and testing. The hidden layer is responsible for connecting the input layer and the output layer that transmits the model results. i The formula (1) for the output of a neuron is:

[0122] Formula (1);

[0123] in, y i It is i The output of the hidden layer neurons, f is the activation function, is the weighted sum of neurons, n is the number of neurons in the input layer, x j Indicates that it comes from j The input of a neuron, For the j Input layer neurons to the i The weights of the neurons in the hidden layer, b i It is i The bias value of each hidden layer neuron.

[0124] from Figure 7It can be seen that the input layer can input key parameters, the output layer outputs the debonding strain or the ultimate bending moment, and the hidden layer is used to connect the input layer and the output layer. Then, according to the training set and the test set, multiple iterative solutions are performed with the goal of minimizing the mean square error and maximizing the deterministic coefficient to obtain the trained neural network model. In this embodiment, for the debonding strain and the ultimate bending moment, it cannot be guaranteed that the debonding strain and the ultimate bending moment obtained by performing multiple iterative solutions with the goal of minimizing the mean square error and maximizing the deterministic coefficient are both optimal solutions. Therefore, the neural models corresponding to the debonding strain and the ultimate bending moment can be trained respectively with the goal of minimizing the mean square error and maximizing the deterministic coefficient to obtain the target debonding strain and the target ultimate bending moment.

[0125] Since different parameters have different units and scales, the variables (including independent variables and dependent variables) are normalized using the following formula before training begins. Formula (2) is used to limit the variables to the range of [−1, 1].

[0126] Formula (2)

[0127] in, x is the actual value of the variable, x n For variables x The normalized value, x min and x max Variables x The minimum and maximum values in the database.

[0128] Neural network models are trained using the classic back-propagation (BP) method. Two common optimization algorithms, the Levenberg-Marquardt (LM) and Bayesian regularization (BR), are commonly used to train BP neural networks. The LM algorithm is a smooth, coordinated combination of the Newton method and gradient descent, mitigating the BP algorithm's shortcomings of slow iteration speed and proneness to local optimality. Furthermore, the BR algorithm is a network training function that optimizes and updates weights and biases based on the LM algorithm. This function minimizes the combination of squared error and weights, then determines the correct combination to produce a network with reasonable generalization performance.

[0129] The LM algorithm and the BR algorithm can be used to pre-train the database. Since the BR algorithm does not require a validation dataset, the neural network model can be trained and tested using 80% and 20% of the data, respectively. To generate neuron outputs and ensure data is transmitted through the hidden layer and output layer, the two transfer functions are Tansig (double S-shaped) and Purelin (linear), respectively, as shown below:

[0130] Formula (3);

[0131] Formula (4);

[0132] Using mean square error ( MSE ) and the coefficient of determination ( R 2 ) evaluate the performance of the neural network model, which are expressed as the following formulas:

[0133] Formula (5);

[0134] Formula (6);

[0135] in, n ' is the total number of samples, y i is the predicted value, t i is the target value, is the mean of the target values.

[0136] It is worth noting that when formula (5) is used to evaluate the error, the problem is that the gap is large. t i , which may cause the error of the neural network model to be larger t i The values are significantly correlated, which reduces the t i The value of the adjustment effect on the neural network model parameters. For example, in the current database, due to the consideration of different diameter sizes, this leads to the large diameter ( D s =1200 mm) component load compared to small diameter ( D s =100 mm) components have a large load difference (the maximum difference between the two is more than 100 times). This means that in order to minimize the mean square error during the training process of the neural network model, its model training results will mainly rely on the parameters of the large diameter components, while the results of the small diameter components are ignored. In other words, the current MSE Methods for evaluating model error may result in smaller t i Value is calculated MSE The value is ignored, which significantly reduces the generalization performance of the model.

[0137] To solve the above problem, the load and debonding strain parameters can also be logarithmized to reduce the influence of large target values on the overall error. Before normalizing the debonding strain or ultimate bending moment in the database, it can be rewritten using formula (7):

[0138] Formula (7);

[0139] in, t i Represents the target debonding strain or target ultimate bending moment.

[0140] Considering the number of hidden layers and hidden layer neurons in the neural network model and the complex nonlinearity of the model, most neural network models currently only provide the prediction results of the model but do not provide an explicit calculation method for the prediction results. In addition, due to the use of too many neurons, the explicit calculation method provided is too complicated. Therefore, under the condition of ensuring the accuracy of the model calculation, the structure of the neural network model can be defined by using as few hidden layer neurons as possible. D s 、 t s 、 f y 、 E cf 、 f cf 、 n cf 、 t s0 、 G Ⅱ ,as well as L cf,s / L s Parameters are used as input to predict the ultimate bending moment , and the debonding strain prediction results As output. The input parameters ignore the parameters that have little impact on the results, such as t 1. G Ⅰ and f cu In the prediction model of debonding strain, the fracture failure of carbon fiber reinforced composite materials, namely the tensile strength f cf , to ensure the debonding failure of carbon fiber reinforced composite materials, refer to Figure 6 , Figure 6 This is a schematic diagram of the optimal number of hidden layer neurons provided by an embodiment of the present application. Figure 6It can be seen that, considering the performance of the neural network model and the complexity of the model, the optimal number of hidden layer neurons in the neural network model for predicting debonding strain and ultimate bending moment is determined to be 6 and 5 respectively.

[0141] In a specific embodiment, the method of performing multiple iterative solutions based on the training set and the test set with the goal of minimizing the mean square error and maximizing the coefficient of determination to obtain the trained neural network model includes the following steps:

[0142] Step 1: Initialize the weights and biases in the initial neural network model;

[0143] Step 2: Inputting the training set into the initial neural network model, obtaining the predicted output and performing denormalization processing, and determining the mean square error and the coefficient of determination based on the predicted output and the actual output after the denormalization processing;

[0144] Step 3: Based on the mean square error, update the weight and the bias through a back propagation algorithm to minimize the mean square error;

[0145] Step 4: During the training process, the neural network model is validated using the test set to maximize the determination coefficient;

[0146] Step 5: Repeat the iterative training process from step 2 to step 4 until the neural network model reaches a preset maximum number of iterations or the model converges.

[0147] In this embodiment, weights are parameters connecting different neurons in a neural network, representing the importance or influence of input signals in the transmission process. Biases are additional parameters for each neuron in the neural network, used to adjust the neuron's activation threshold. Before training the initial neural network model, the weights and biases in the initial neural network model are initialized. Specifically, the weights can be randomly initialized based on the number of neurons in the input layer and hidden layer, and the biases can be initialized to small random values or zero. The key parameters of the training set are then input into the initial neural network model. The model's predicted output is calculated through forward propagation. The predicted output is converted from a normalized range back to the original data range. The denormalized predicted output is then compared with the actual output to calculate the mean squared error (MSE). This can be used to evaluate the neural network model's prediction accuracy. An optimization algorithm is then used to update the weights and biases to minimize the MSE. During training, the model can also be validated using a test set. The MSE and coefficient of determination on the validation set are calculated to ensure the generalization ability of the neural network model and prevent overfitting. The above steps are repeated until the preset maximum number of iterations is reached or the model converges (i.e., the network output error is less than a certain threshold), completing the training and obtaining the neural network model. The neural network model obtained after training is used to predict the target debonding strain and target ultimate bending moment corresponding to the set parameter range.

[0148] For example, the following will refer to Figures 8 to 11 , the target debonding strain and target ultimate bending moment are explained, Figure 8 This is a training flow diagram of a neural network model provided in an embodiment of the present application, starting from Figure 8 As we can see, first let , and normalize all variables → build the initial neural network → start iteration counting → update weights and biases → randomly divide the data set into training set and test set → use the training set and test set to train and test the constructed initial neural network model, and save the neural network model after each training until the number of iterations is greater than 1000 or the iteration converges → obtain the outputs corresponding to all neural networks and perform denormalization processing → finally select the neural network model corresponding to the smallest mean square error and the highest coefficient of determination.

[0149] Figure 9 (a) is for Figure 8 A schematic diagram of the performance of a debonding strain neural network model is provided; Figure 9 (b) is for Figure 8A schematic diagram illustrating the performance of a neural network model for ultimate bending moment is provided. At the beginning of the training process, the mean squared error (MSE) values for both the training and test sets are high. As training progresses, the MSE values for both the training and test sets decrease, indicating that the neural network model has learned the relationship between input and output. The neural network models for predicting debonding strain and ultimate bending moment achieved their best training results at iterations 130 and 389, respectively, with MSE values of 0.00035 and 0.0000158.

[0150] Figure 10 (a) is a schematic diagram showing the comparison results of the debonding strain predicted by a neural network model provided in an embodiment of the present application and the target value in a test set; Figure 10 (b) is a schematic diagram showing the comparison results of the debonding strain predicted by a neural network model provided in an embodiment of the present application and the target value in the validation set; Figure 10 (c) is a schematic diagram showing the comparison results of the debonding strain predicted by a neural network model provided in an embodiment of the present application and the target value in the total set; Figure 11 (a) is a schematic diagram showing the comparison results of the ultimate bending moment predicted by a neural network model and the target value in a test set provided in an embodiment of the present application; Figure 11 (b) is a schematic diagram showing the comparison results of the ultimate bending moment predicted by a neural network model and the target value in a validation set provided in an embodiment of the present application; Figure 11 (c) is a schematic diagram of the comparison results of the ultimate bending moment predicted by a neural network model provided in an embodiment of the present application and the target value in the total set; the total set includes a test set and a validation set. Figure 10 (a) Figure 10 (b) Figure 10 (c) Figure 11 (a) Figure 11 (b) and Figure 11 The comparison results in (c) show that the predicted values (Y) of all data are basically consistent with the target values (T), that is, the data points are distributed near the straight line Y = T. For the neural network models that predict debonding strain and ultimate bending moment, their test sets R 2 The (coefficient of determination) values are 0.97225 and 0.99968, respectively, indicating that the trained neural network model has excellent generalization performance.

[0151] In a specific embodiment, outputting the target debonding strain and the target ultimate bending moment based on the neural network model includes: obtaining updated target weights and target biases based on the neural network model; and outputting the target debonding strain and the target ultimate bending moment based on the target weights and the target biases in combination with a double S-type transfer function and a linear transfer function.

[0152] In this embodiment, according to the neural network model, the updated target weight is obtained. Sum bias and target bias .

[0153] Then, according to the target weight and target bias, combined with the double S-type transfer function and the linear transfer function, the target debonding strain for predicting the carbon fiber reinforced composite reinforced circular steel tube concrete flexural member can be obtained. and target ultimate bending moment The explicit calculation formulas of , namely formula (8) and formula (9), are as follows:

[0154] Formula (8);

[0155] Formula (9);

[0156] in, The calculated result is micro strain, The unit is MPa. In formula (8) , i =[1,6]; in formula (9) , i =[1,5]. and The calculation formula is as follows:

[0157] Formula (10);

[0158] Formula (11);

[0159] in, and The unit is mm; 、 、 、 The unit is MPa; G Ⅱ The unit is N / mm. 、 、 、 、 、 、 、 、 as well as 、 、 、 、 、 、 、 、 、 The values of are shown in Table 1 and Table 2 respectively.

[0160] Table 1- The constant value in the calculation equation of

[0161] ;

[0162] Table 2- The constant value in the calculation equation of

[0163] ;

[0164] In a specific embodiment, after establishing the finite element model, the method further includes:

[0165] The finite element model is verified, wherein the verification of the finite element model includes the following steps: comparing the calculation results of the finite element model with the test results, wherein the calculation results and the test results include a bending moment-mid-span deflection curve and a strain distribution curve; when the calculation results are basically consistent with the test results, the finite element model is determined to have passed the verification.

[0166] In this embodiment, a moment-deflection curve and a strain distribution curve are obtained through experiments. Parameters consistent with the experimental conditions are then input into a finite element model, and the finite element model is run to obtain calculation results. The calculated results from the finite element model are then compared with the experimental results. The shape and key points of the moment-deflection curve (such as the ultimate moment and deflection) as well as the regularity and key points of the strain distribution (such as the maximum strain and strain gradient) are examined, and the percentage error between the two is calculated. If the error is within an acceptable range (typically less than 20%), the model is deemed to have passed verification, and subsequent steps are then performed.

[0167] For example, refer to the comparison chart of finite element model calculation results and test results. Figure 12 (a) is a schematic diagram showing the comparison results of the bending moment-mid-span deflection curves of a carbon fiber reinforced composite material debonding failure provided in an embodiment of the present application; Figure 12 (b) is a schematic diagram showing the comparison results of the bending moment-mid-span deflection curves of a carbon fiber reinforced composite material fracture failure provided in an embodiment of the present application; Figure 13 (a) is a schematic diagram showing the comparison results of strain distribution curves of debonding failure of a carbon fiber reinforced composite material provided in an embodiment of the present application; Figure 13 (b) is a schematic diagram showing the comparison of strain distribution curves of fracture failure of a carbon fiber reinforced composite material provided in an embodiment of the present application; wherein the FE result is the calculation result of the finite element model. Figure 12 (a) and Figure 12 As can be seen from (b), the moment-mid-span deflection curve obtained by the finite element model is basically consistent with the test results. Figure 13 (a) and Figure 13 As shown in (b), the finite element model effectively simulates the strain distribution of CFRP in CFRP-reinforced concrete-filled steel tubular members subjected to flexural loading. Comparative results show that the finite element model provides excellent simulation results for the moment-midspan deflection curves and strain distribution for specimens experiencing both debonding and fracture failure. This validates the finite element model's accuracy in simulating the flexural performance of CFRP-reinforced concrete-filled steel tubular members.

[0168] The present application also provides a system for predicting the bending performance of carbon fiber reinforced composite reinforced concrete-filled steel tube components. Figure 14 , Figure 14 This is a system diagram for predicting the bending performance of carbon fiber reinforced composite reinforced concrete-filled steel tube components provided by the embodiment of the present application. Figure 14 As can be seen from the figure, the system includes: an establishing unit 141, a determining unit 142, a generating unit 143 and an output unit 144.

[0169] Establishing unit 141, for establishing a finite element model; wherein the finite element model is used to simulate the debonding failure and fracture failure of the carbon fiber reinforced composite material in the carbon fiber reinforced composite material reinforced steel tube concrete member;

[0170] a determining unit 142 for determining key parameters based on the finite element model; wherein the key parameters include steel yield strength, carbon fiber reinforced composite material bonding length, number of carbon fiber reinforced composite material layers, carbon fiber reinforced composite material elastic modulus, carbon fiber reinforced composite material tensile strength, bonding interface tangential fracture energy, bonding interface tangential bonding strength, steel pipe diameter-to-thickness ratio, and steel pipe diameter;

[0171] A generating unit 143 is configured to generate a plurality of data sets including debonding strain and ultimate bending moment based on the key parameters;

[0172] The output unit 144 is used to establish and train a neural network model based on the multiple data sets, so that the neural network model outputs the target debonding strain and target ultimate bending moment in the bending performance of the carbon fiber reinforced composite material reinforced steel tube concrete member.

[0173] In this embodiment, the establishing unit 141 includes:

[0174] a determining unit, configured to change the size of the three-dimensional model and the value of the material parameter based on the finite element model, and respectively obtain the degree of influence of the three-dimensional model and the material parameter on the performance of the carbon fiber reinforced composite material reinforced steel tube concrete component;

[0175] Based on the degree of influence, the key parameters are determined.

[0176] In this embodiment, the generating unit 143 includes:

[0177] a first processing unit, configured to perform parameterization processing on the key parameter so that the key parameter changes within a preset range;

[0178] changing the value of the key parameter, running the finite element model, and obtaining the debonding strain and the ultimate bending moment corresponding to each running of the finite element model;

[0179] The debonding strain and the ultimate bending moment obtained each time are respectively associated with the values of the corresponding key parameters to generate a plurality of the data sets.

[0180] In this embodiment, the output unit 144 includes:

[0181] The second processing unit is used to normalize the multiple data sets and divide the normalized data sets into a training set and a test set; construct an initial neural network model; wherein the initial neural network model includes: an input layer, a hidden layer and an output layer, the input layer receives the key parameters, the output layer outputs the debonding strain or the ultimate bending moment, and the hidden layer is used to connect the input layer and the output layer; based on the training set and the test set, with minimizing the mean square error and maximizing the coefficient of determination as the solution goal, multiple iterative solutions are performed to obtain the trained neural network model; based on the neural network model, the target debonding strain and the target ultimate bending moment are output.

[0182] In this embodiment, the second processing unit includes the following steps:

[0183] Step 1: Initialize the weights and biases in the initial neural network model;

[0184] Step 2: Inputting the training set into the initial neural network model, obtaining the predicted output and performing denormalization processing, and determining the mean square error and the coefficient of determination based on the predicted output and the actual output after the denormalization processing;

[0185] Step 3: Based on the mean square error, update the weight and the bias through a back propagation algorithm to minimize the mean square error;

[0186] Step 4: During the training process, the neural network model is validated using the test set to maximize the determination coefficient;

[0187] Step 5: Repeat the iterative training process from step 2 to step 4 until the neural network model reaches a preset maximum number of iterations or the model converges.

[0188] In this embodiment, the second processing unit includes:

[0189] The first acquisition unit is used to obtain updated target weights and target biases based on the neural network model; based on the target weights and the target biases, combined with a double S-type transfer function and a linear transfer function, output the target debonding strain and the target ultimate bending moment.

[0190] In this embodiment, the hidden layer adopts the double S-shaped transfer function, and the output layer adopts the linear transfer function.

[0191] In this embodiment, after establishing the finite element model, the system further includes:

[0192] A verification unit is used to verify the finite element model, wherein the verification of the finite element model includes the following steps: comparing the calculation results of the finite element model with the test results, wherein the calculation results and the test results include a bending moment-mid-span deflection curve and a strain distribution curve; when the calculation results are consistent with the test results, determining that the finite element model has passed the verification.

[0193] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0194] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods and apparatuses according to the embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0195] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0196] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0197] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they are aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0198] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

[0199] The above is a detailed introduction to a method for predicting the bending performance of steel tube concrete components reinforced with carbon fiber reinforced composite materials provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for predicting the flexural properties of carbon fiber reinforced composite reinforced concrete-filled steel tube components, characterized in that: include: Establishing a finite element model; wherein the finite element model is used to simulate the debonding failure and fracture failure of the carbon fiber reinforced composite material in the carbon fiber reinforced composite material reinforced steel tube concrete member; Determine key parameters based on the finite element model; wherein the key parameters include steel yield strength, carbon fiber reinforced composite material bonding length, number of carbon fiber reinforced composite material layers, carbon fiber reinforced composite material elastic modulus, carbon fiber reinforced composite material tensile strength, bonding interface tangential fracture energy, bonding interface tangential bonding strength, steel pipe diameter-to-thickness ratio, and steel pipe diameter; Based on the key parameters, multiple data sets including debonding strain and ultimate bending moment are generated; wherein, based on the key parameters, multiple data sets including debonding strain and ultimate bending moment are generated, including: Performing parameterization processing on the key parameters so that the key parameters vary within a preset range; changing the value of the key parameter, running the finite element model, and obtaining the debonding strain and the ultimate bending moment corresponding to each running of the finite element model; Associating the debonding strain and the ultimate bending moment obtained each time with the corresponding values of the key parameters to generate a plurality of data sets; Based on the multiple data sets, a neural network model is established and trained so that the neural network model outputs the target debonding strain and target ultimate bending moment in the bending performance of the carbon fiber reinforced composite material reinforced steel tube concrete member.

2. The method according to claim 1, characterized in that The establishing of the finite element model comprises: Obtaining a three-dimensional model and material parameters of the carbon fiber reinforced composite material reinforced steel tube concrete component; wherein the material parameters include at least: steel parameters, carbon fiber reinforced composite material parameters, concrete parameters and bonding interface parameters; Based on the three-dimensional model and the material parameters, boundary conditions and loads are applied to establish the finite element model; wherein the boundary conditions represent the constraint state of the finite element model, and the loads represent the external forces acting on the finite element model.

3. The method according to claim 2, characterized in that Determining key parameters based on the finite element model includes: Based on the finite element model, changing the size of the three-dimensional model and the value of the material parameter, and respectively obtaining the degree of influence of the three-dimensional model and the material parameter on the performance of the carbon fiber reinforced composite material reinforced steel tube concrete component; Based on the degree of influence, the key parameters are determined.

4. The method according to claim 1, wherein The establishing and training of a neural network model based on the plurality of data sets so that the neural network model outputs a target debonding strain and a target ultimate bending moment in the bending performance of the carbon fiber reinforced composite material reinforced steel tube concrete member includes: Normalizing the plurality of data sets, and dividing the normalized data sets into a training set and a test set; Constructing an initial neural network model; wherein the initial neural network model includes: an input layer, a hidden layer, and an output layer, the input layer receives the key parameters, the output layer outputs the debonding strain or the ultimate bending moment, and the hidden layer is used to connect the input layer and the output layer; Based on the training set and the test set, minimizing the mean square error and maximizing the coefficient of determination as the solution goal, performing multiple iterative solutions to obtain the trained neural network model; Based on the neural network model, the target debonding strain and the target ultimate bending moment are output.

5. The method according to claim 4, characterized in that The method of performing multiple iterations based on the training set and the test set with the goal of minimizing the mean square error and maximizing the coefficient of determination to obtain the trained neural network model includes the following steps: Step 1: Initialize the weights and biases in the initial neural network model; Step 2: Inputting the training set into the initial neural network model, obtaining the predicted output and performing denormalization processing, and determining the mean square error and the coefficient of determination based on the predicted output and the actual output after the denormalization processing; Step 3: Based on the mean square error, update the weight and the bias through a back propagation algorithm to minimize the mean square error; Step 4: During the training process, the neural network model is validated using the test set to maximize the determination coefficient; Step 5: Repeat the iterative training process from step 2 to step 4 until the neural network model reaches a preset maximum number of iterations or the model converges.

6. The method according to claim 4, characterized in that Outputting the target debonding strain and the target ultimate bending moment based on the neural network model includes: Based on the neural network model, obtaining updated target weights and target biases; Based on the target weight and the target bias, the target debonding strain and the target ultimate bending moment are output in combination with a double S-type transfer function and a linear transfer function.

7. The method according to claim 6, characterized in that The hidden layer adopts the double S-shaped transfer function, and the output layer adopts the linear transfer function.

8. The method according to claim 1, characterized in that After establishing the finite element model, the method further includes: The finite element model is verified, wherein the verification of the finite element model comprises the following steps: Comparing the calculation results of the finite element model with the test results, wherein the calculation results and the test results include a bending moment-mid-span deflection curve and a strain distribution curve; When the calculation result is consistent with the test result, the finite element model is determined to have passed the verification.

9. A system for predicting the flexural performance of carbon fiber reinforced composite reinforced concrete-filled steel tube members, characterized in that: include: Establishing a unit for establishing a finite element model; wherein the finite element model is used to simulate the debonding failure and fracture failure of the carbon fiber reinforced composite material in the carbon fiber reinforced composite material reinforced steel tube concrete member; a determination unit, configured to determine key parameters based on the finite element model; wherein the key parameters include steel yield strength, carbon fiber reinforced composite material bonding length, number of carbon fiber reinforced composite material layers, carbon fiber reinforced composite material elastic modulus, carbon fiber reinforced composite material tensile strength, bonding interface tangential fracture energy, bonding interface tangential bonding strength, steel pipe diameter-to-thickness ratio, and steel pipe diameter; A generating unit is configured to generate a plurality of data sets including debonding strain and ultimate bending moment based on the key parameters; wherein generating a plurality of data sets including debonding strain and ultimate bending moment based on the key parameters comprises: Performing parameterization processing on the key parameters so that the key parameters vary within a preset range; changing the value of the key parameter, running the finite element model, and obtaining the debonding strain and the ultimate bending moment corresponding to each running of the finite element model; Associating the debonding strain and the ultimate bending moment obtained each time with the corresponding values of the key parameters to generate a plurality of data sets; An output unit is used to establish and train a neural network model based on the multiple data sets, so that the neural network model outputs the target debonding strain and target ultimate bending moment in the bending performance of the carbon fiber reinforced composite material reinforced steel tube concrete member.

Images