Multi-dimensional data-driven-based discrete network modeling method and system for p-aramid paper, and mechanical property prediction method and system

Abstract

This invention provides a method and system for modeling discrete networks of para-aramid paper and predicting its mechanical properties based on multidimensional data-driven approaches, belonging to the field of electronic digital data processing technology. Based on areal density, fiber bulk density, and measured in-plane orientation distribution, this invention analyzes the total number of chopped fibers and reconstructs a random fiber skeleton geometric model. It uses a spatial indexing algorithm to search for fiber intersections, generating connector units in batches to equivalently characterize the "bonding and covering" effect of precipitated fibers. It assigns an anisotropic coupling failure criterion and a quasi-brittle damage evolution model to the connector units, constructing a high-fidelity constitutive model. Then, it applies multiple load conditions for explicit dynamic solutions, outputting macroscopic stress-strain curves and damage evolution cloud maps. This invention realistically reproduces the preferred fiber orientation and layered stacking structure, accurately predicts longitudinal brittle fracture and transverse progressive softening, and is numerically stable with strong engineering guidance.

Description

Technical Field

[0001] This invention belongs to the field of electrical digital data processing technology, and relates to force analysis or optimization, computer-aided design, and the use of finite element methods. Specifically, it relates to a multi-dimensional data-driven discrete network modeling method and system for para-aramid paper, and a mechanical property prediction method and system. Background Technology

[0002] Para-aramid paper, with its excellent specific strength, specific stiffness, high temperature resistance, and flame-retardant insulation properties, is an indispensable core substrate for lightweight, high-strength composite materials in the aerospace field (such as radomes for large aircraft and honeycomb sandwich structures for cabin floors). Para-aramid paper contains chopped fibers as a skeleton and precipitated fibers that coat the chopped fibers.

[0003] In the wet papermaking and high-temperature hot-pressing processes of para-aramid paper, the combined effects of the headbox wire speed ratio and mechanical traction inevitably cause the chopped fibers to undergo preferential orientation along the machine direction (MD), resulting in extremely significant mechanical anisotropy in both the longitudinal and transverse directions (CD) of the para-aramid paper. This macroscopic mechanical anisotropy and its unique failure evolution mode directly determine the final load-bearing limit of aramid paper and subsequent honeycomb core materials under complex stress conditions.

[0004] In terms of microstructure, para-aramid paper differs from ordinary paper-based materials. It is composed of a mixture of highly crystalline, high-strength chopped fibers and thin-film precipitated fibers. During the critical high-temperature hot-pressing process, the precipitated fibers soften and plastically flow, filling the gaps in the chopped fibers that form the skeleton. After cooling, a tight physical wrapping and mechanical interlocking are formed at the nodes of the chopped fibers.

[0005] Currently, the closest approach to predicting the macroscopic properties and microscopic mechanisms of paper-based materials is to use the Discrete Fiber Network (DFN) model. This type of model simulates the overall mechanical response of the network under external loads by randomly generating one-dimensional beam / truss elements to represent the fiber skeleton in the computational space of finite element simulation software and establishing contact or cohesive constraints at the nodes where fibers intersect.

[0006] However, existing discrete fiber network models have the following significant drawbacks when predicting the mechanical properties of para-aramid paper:

[0007] (1) Geometric representation distortion: Existing algorithms mostly use idealized isotropic random generation, lack data-driven based on real fiber orientation distribution (FOD), and cannot accurately restore the preferred orientation of para-aramid paper in the plane.

[0008] (2) Constitutive model mismatch: Existing models are mostly designed for cellulose paper, simplifying nodes to "hydrogen bond-contact friction" or ideal point bonding models. Such models cannot reflect the unique "bonding and coating" characteristics of precipitated fibers on chopped fibers in para-aramid paper. Summary of the Invention

[0009] This invention is made to solve the above-mentioned problems, and aims to provide a method and system for modeling discrete networks of para-aramid paper based on multidimensional data-driven approach, as well as a method and system for predicting mechanical properties.

[0010] This invention provides a multi-dimensional data-driven discrete network modeling method for para-aramid paper, which contains precipitated fibers and chopped fibers. The method includes the following steps: S10, obtaining the areal density of the para-aramid paper. Bulk density of chopped fibers The in-plane orientation distribution characteristics of chopped fibers and the physical encapsulation structure characteristics of precipitated fibers on chopped fibers; S20, after setting the target model size, based on... , The size parameters of the chopped fibers, the mass ratio of chopped fibers to precipitated fibers, and the total number of chopped fibers to be generated in the target model are analyzed. S30, according to In addition to the in-plane orientation distribution characteristics, combined with the spatial joint probability density function, the geometric model of the random fiber skeleton composed of short-cut fibers in the target model is reconstructed; S40, after discretizing the geometric model of the random fiber skeleton with a finite element mesh, the effective contact points between each short-cut fiber are searched based on the spatial indexing algorithm, and connector units are generated in batches at the effective contact points to equivalently characterize the physical wrapping effect of the precipitated fibers on the short-cut fibers, so as to establish a physically equivalent microscopic connection model; S50, the connector units are given anisotropic coupling failure criteria and quasi-brittle damage evolution model to construct the constitutive model of para-aramid paper. The anisotropic coupling failure criteria are parameterized based on the physical wrapping structure characteristics.

[0011] The method for modeling discrete networks of para-aramid paper based on multidimensional data-driven approaches provided by this invention may also have the following feature: wherein, in step S10, the areal density... and volume density The in-plane orientation distribution characteristics of the short-cut fibers were obtained through physical measurement methods, measured by X-ray micro-CT, and the physical encapsulation structure characteristics were observed by scanning electron microscopy.

[0012] The multidimensional data-driven discrete network modeling method for para-aramid paper provided by this invention may also have the following feature: In step S20, the total number of short-cut fibers to be generated in the target model is calculated and rounded according to the following formula: In the above formula, and These represent the length and width of the target model, respectively. This indicates the mass ratio of chopped fibers to precipitated fibers during the initial preparation of para-aramid paper. This refers to the diameter data in the dimensional parameters of chopped fibers. This indicates the length data in the dimensional parameters of chopped fibers.

[0013] The multidimensional data-driven discrete network modeling method for para-aramid paper provided by this invention may also have the following features: In step S10, the in-plane orientation distribution characteristics of the chopped fibers are in-plane orientation frequencies, and step S30 includes the following sub-steps: S31, nonlinear fitting is performed on the in-plane orientation frequencies to measure the orientation standard deviation; S32, in Abaqus finite element software, random distribution is generated within the target model according to a uniform distribution. S33: Set the in-plane azimuth angle for each chopped fiber, making the in-plane azimuth angle follow a Gaussian distribution with the machine direction as the mean and the orientation standard deviation as the variance, so as to truly reflect the mechanical traction effect brought about by the papermaking headbox process used in the preparation of para-aramid paper. Set the polar angle in the thickness direction for each chopped fiber, and constrain the polar angle within a limited uniform distribution range near the horizontal plane, so as to simulate the dense layered stacking characteristics caused by the high temperature hot pressing process used in the preparation of para-aramid paper. S34: After the settings in step S33 are completed, perform a Boolean merging operation on all chopped fibers to obtain the random fiber skeleton geometric model of para-aramid paper.

[0014] The method for modeling discrete networks of para-aramid paper based on multidimensional data-driven approach provided by this invention may also have the following feature: step S32 is executed based on a Python automated script.

[0015] The multidimensional data-driven discrete network modeling method for para-aramid paper provided by this invention may also have the following features: Step S40 includes the following sub-steps: S41, in Abaqus finite element software, Timoshenko beam elements are used to mesh the random fiber skeleton geometric model to capture the large deformation and transverse shear effect of chopped fibers under stress; S42, the kd-tree spatial indexing algorithm is used to traverse all nodes of the beam elements in the global scope, and the Euclidean distance between each node is searched to filter out node pairs whose distance is less than the preset search radius threshold and which do not overlap, so as to determine that the location of the node pair is the effective contact point between the chopped fibers; S43, discrete connector elements are generated in batches at the effective contact points, and a local coordinate system following the local direction of the intersecting chopped fibers is established for each connector element, so as to use the connector element to equivalently represent the physical wrapping effect of the precipitated fiber at the intersection of the chopped fibers.

[0016] The multidimensional data-driven discrete network modeling method for para-aramid paper provided by this invention may also have the following features: Step S50 includes the following sub-steps: S51, defining the coupling mechanical response index of the connector unit under normal tension and tangential shear based on the local coordinate system, so as to construct an anisotropic coupling failure criterion that can reflect the tensile strength and shear weakness of the precipitated fiber to the chopped fiber in the physical wrapping structure; S52, assigning an equivalent elastic stiffness to each connector unit, and setting the coupling tensile limit as the damage initiation criterion based on the physical wrapping structure characteristics, wherein when the force on the connector unit in the normal or tangential direction exceeds the coupling tensile limit, the damage evolution of the connector unit is triggered; S53, setting the displacement difference of the connector unit from the damage initiation displacement to the complete failure displacement to an extremely low value, so as to simulate the physical characteristics of the brittle fracture of the chopped fiber intersection wrapped by the precipitated fiber after reaching the ultimate bearing capacity, thereby introducing a quasi-brittle damage evolution model for the connector unit.

[0017] This invention also provides a multi-dimensional data-driven discrete network modeling system for para-aramid paper, characterized by using any of the aforementioned multi-dimensional data-driven discrete network modeling methods for para-aramid paper, including: a data input module for allowing the user to input the measured areal density of the para-aramid paper. Bulk density of chopped fibers In-plane orientation distribution characteristics of chopped fibers; fiber total count analysis module, used for... , The size parameters of the chopped fibers, the mass ratio of chopped fibers to precipitated fibers, and the total number of chopped fibers to be generated in the target model are analyzed. ; Geometric skeleton reconstruction module, used to reconstruct the geometric skeleton based on In addition to the in-plane orientation distribution characteristics, combined with the spatial joint probability density function, the geometric model of the random fiber skeleton composed of short-cut fibers in the target model is reconstructed; the connection model establishment module is used to discretize the geometric model of the random fiber skeleton into a finite element mesh, search for effective contact points between each short-cut fiber based on the spatial indexing algorithm, and generate connector units in batches at the effective contact points to equivalently characterize the physical wrapping effect of precipitated fibers on short-cut fibers, so as to establish a physically equivalent microscopic connection model; and the constitutive model assignment module is used to assign anisotropic coupling failure criteria and quasi-brittle damage evolution model to the connector units to construct the constitutive model of para-aramid paper.

[0018] This invention also provides a method for predicting the mechanical properties of para-aramid paper based on multidimensional data-driven methods. The mechanical property prediction method is characterized by applying multi-condition boundary conditions to the constitutive model and performing explicit dynamic solutions to output the macroscopic mechanical response and microscopic damage evolution results. The constitutive model is either the multidimensional data-driven discrete network modeling method for para-aramid paper mentioned above or the constitutive model constructed by the multidimensional data-driven discrete network modeling system for para-aramid paper mentioned above.

[0019] This invention also provides a multi-dimensional data-driven system for predicting the mechanical properties of para-aramid paper, characterized by: the aforementioned multi-dimensional data-driven discrete network modeling system for para-aramid paper, used to output the constitutive model of para-aramid paper; and a prediction module used to apply multi-condition boundary conditions to the constitutive model and perform explicit dynamic solutions to output the macroscopic mechanical response and microscopic damage evolution results.

[0020] The beneficial effects of this invention are:

[0021] (1) High microstructure reproduction: This invention introduces orientation statistical features extracted from multidimensional experimental data (such as Micro-CT) to establish a geometric generation algorithm controlled by statistical laws. At the same time, it uses a quota-based analytical algorithm based on the ratio of paper weight to fiber mass to accurately calculate the total number of short fibers, ensuring the mass conservation of the virtual geometric model and the real physical sample. This allows for the true reproduction of the three-dimensional spatial topology and hot-pressed layered stacking characteristics of aramid paper, abandoning the traditional blind random assumptions and solving the problem of lack of physical fidelity in the geometric model.

[0022] (2) Accurate anisotropic failure prediction: The present invention sets up connector units at fiber cross nodes and introduces anisotropic coupling failure criteria. By defining the critical failure force, it physically and equivalently characterizes the real mechanical response of the precipitation fiber to the unique "bonding and covering" structure of the short-cut fiber under complex stress, and truly reproduces the mechanical characteristics of the precipitation fiber of "high tensile strength and weak shear strength". Only one set of micro parameters can simultaneously and accurately predict the longitudinal brittle fracture and transverse progressive softening and disintegration of aramid paper, solving the problem that the node constitutive model cannot characterize the "bonding and covering" characteristics.

[0023] (3) Stable convergence of numerical calculation: The present invention adopts a quasi-brittle numerical regularization strategy in the damage evolution of connector unit, which perfectly balances the physical reality of instantaneous brittle fracture of nodes and the numerical convergence of explicit dynamic solution, ensuring that the calculation process is stable and reliable.

[0024] (4) Strong guidance for industrial applications: This invention establishes a full-process prediction method that includes geometry generation, connection search and automatic application of multi-condition boundary conditions. It can simultaneously predict different failure evolution modes of aramid paper in the longitudinal and transverse directions under a single set of micro parameters. It can directly evaluate the impact of process changes (such as pulp-to-wire ratio, hot pressing pressure, etc.) on macroscopic performance by adjusting micro parameters, which greatly reduces the trial and error cost of aramid paper R&D and has strong engineering guidance value. Attached Figure Description

[0025] Figure 1 This is a flowchart of a method for predicting the mechanical properties of para-aramid paper based on multidimensional data-driven principles, according to an embodiment of the present invention.

[0026] Figure 2 This is an architecture diagram of a multi-dimensional data-driven system for predicting the mechanical properties of para-aramid paper, according to an embodiment of the present invention.

[0027] Figure 3 It is the numerical value of the in-plane orientation frequency measured in step S31 of the test example of the present invention and the result of its nonlinear fitting.

[0028] Figure 4 It is the random fiber skeleton geometric model generated in step S34 of the test example of the present invention (a panoramic view of the fiber mesh generated in the Abaqus window).

[0029] Figure 5 The full-size virtual style rendering model is obtained after setting step S43 in the test example of the present invention.

[0030] Figure 6 These are the boundary conditions for the two sub-models, Model-MD and Model-CD, in step S60 of the test example of this invention.

[0031] Figure 7 It is the macroscopic stress-strain curve extracted in step S60 of the test example of the present invention.

[0032] Figure 8 This is a spatiotemporal evolution cloud map of connector damage (CDMG) and fiber stress (MISES) in step S60 of the test example of the present invention, showing the model tensile failure. Detailed Implementation

[0033] To make the technical means, creative features, objectives and effects of this invention easy to understand, the following embodiments, in conjunction with the accompanying drawings, specifically illustrate a method and system for modeling discrete networks of para-aramid paper based on multidimensional data-driven approaches, and a method and system for predicting mechanical properties.

[0034] Example

[0035] Figure 1This is a flowchart of a method for predicting the mechanical properties of para-aramid paper based on multidimensional data-driven principles, according to an embodiment of the present invention.

[0036] like Figure 1 As shown, this embodiment provides a method for predicting the mechanical properties of para-aramid paper based on multidimensional data-driven methods (steps S10-S50), which uses a discrete network modeling method for para-aramid paper based on multidimensional data-driven methods (steps S10-S40). The para-aramid paper contains chopped fibers as a skeleton and precipitated fibers that coat the chopped fibers.

[0037] The steps of the mechanical property prediction method are as follows:

[0038] S10, Obtain multidimensional experimental data of para-aramid paper:

[0039] (1) Obtain the areal density of para-aramid paper by physical measurement method and bulk density of short-cut fibers .

[0040] (2) The in-plane orientation distribution characteristics of the short-cut fibers were measured by X-ray micro-CT (specifically, the in-plane orientation frequency in this embodiment).

[0041] (3) The physical encapsulation structure of the precipitated fibers to the short-cut fibers was observed by scanning electron microscopy.

[0042] S20, Analyze the total number of short-cut fibers to be generated. The specific method is as follows:

[0043] After setting the target model size, based on the surface density Volume density Dimensional parameters of chopped fibers, mass ratio of chopped fibers to precipitated fibers The total number of short-cut fibers to be generated in the target model is calculated using the following formula and rounded down:

[0044] .

[0045] In the above formula, and These represent the length and width of the target model, respectively. This indicates the mass ratio of chopped fibers to precipitated fibers during the initial preparation of para-aramid paper. This refers to the diameter data in the dimensional parameters of chopped fibers. This indicates the length data in the dimensional parameters of chopped fibers.

[0046] S30, according to In addition to the in-plane orientation frequency, combined with the spatial joint probability density function, the geometric model of the random fiber skeleton composed of chopped fibers in the target model is reconstructed, including the following sub-steps S31~S34:

[0047] S31, the orientation standard deviation is measured by nonlinear fitting of the orientation frequency in the face.

[0048] S32, based on a Python automated script, is randomly generated within the target model using a uniform distribution in the Abaqus finite element software. Each spatial coordinate serves as the center position coordinate of each chopped fiber.

[0049] S33 sets an in-plane azimuth angle for each chopped fiber, making the in-plane azimuth angle follow a Gaussian distribution with the machine direction as the mean and the orientation standard deviation as the variance, so as to truly reflect the mechanical traction effect brought about by the papermaking headbox process used in the preparation of para-aramid paper.

[0050] A polar angle in the thickness direction is set for each chopped fiber, and the polar angle is constrained within a limited uniform distribution range near the horizontal plane to simulate the dense layered stacking characteristics caused by the high-temperature hot pressing process used in the preparation of para-aramid paper.

[0051] S34. After step S33 is completed, a Boolean merging operation is performed on all the short fibers to obtain the random fiber skeleton geometric model of the para-aramid paper.

[0052] S40, perform finite element mesh discretization and generate connector elements, including the following sub-steps S41~S43:

[0053] S41, in Abaqus finite element software, Timoshenko beam elements (B31) were used to mesh the random fiber skeleton geometry model to capture the large deformation and transverse shear effect of short-cut fibers under stress.

[0054] S42 uses the kd-tree spatial indexing algorithm to traverse all nodes of the beam element in the global scope, searches the Euclidean distance between each node to filter out node pairs whose distance is less than the preset search radius threshold and which do not overlap, so as to determine that the location of the node pair is the effective contact point between the short fibers.

[0055] S43, generate discrete connector units (CONN3D2) in batches at the effective contact points, and establish a local coordinate system for each connector unit that follows the local direction of the intersecting short chopped fibers, so as to use the connector unit to equivalently characterize the physical wrapping effect of the precipitated fibers at the intersection of the short chopped fibers.

[0056] S50, Assign anisotropic coupling failure criteria and quasi-brittle damage evolution model to the connector unit to construct the constitutive model of para-aramid paper, including the following sub-steps S51~S53:

[0057] S51 defines the coupling mechanical response index of the connector unit under normal tension and tangential shear based on the local coordinate system, so as to construct an anisotropic coupling failure criterion that can reflect the tensile strength and shear weakness of the precipitated fiber to the chopped fiber in the physical encapsulation structure.

[0058] S52 assigns an equivalent elastic stiffness to each connector unit and sets the coupling tensile limit as a damage initiation criterion based on the physical encapsulation structure characteristics.

[0059] Specifically, when the force on the connector unit in the normal or tangential direction exceeds the coupling tensile limit, damage evolution of the connector unit is triggered.

[0060] S53, the displacement difference of the connector unit from the damage initiation displacement to the complete failure displacement is set to an extremely low value to simulate the physical characteristics of the brittle fracture of the short fiber intersection wrapped by the precipitated fiber after reaching the ultimate bearing capacity, thereby introducing a quasi-brittle damage evolution model for the connector unit, and finally constructing the constitutive model of the para-aramid paper.

[0061] S60, Mechanical Performance Prediction:

[0062] Multiple working condition boundary conditions are applied to the constitutive model and explicit dynamic solutions are performed to output the macroscopic mechanical response and microscopic damage evolution results.

[0063] Figure 2 This is an architecture diagram of a multi-dimensional data-driven system for predicting the mechanical properties of para-aramid paper, according to an embodiment of the present invention.

[0064] like Figure 2 As shown, this embodiment provides a multi-dimensional data-driven system for predicting the mechanical properties of para-aramid paper, which includes a multi-dimensional data-driven discrete network modeling system for para-aramid paper, a system 100, and a prediction module 60.

[0065] The multidimensional data-driven discrete network modeling system 100 for para-aramid paper includes a data input module 10, a fiber count analysis module 20, a geometric skeleton reconstruction module 30, a connection model establishment module 40, and a constitutive model assignment module 50.

[0066] Data input module 10 is used for users to input the measured areal density of para-aramid paper. Bulk density of chopped fibers The in-plane orientation distribution characteristics of short-cut fibers.

[0067] Fiber count analysis module 20 is used for... , The size parameters of the chopped fibers, the mass ratio of chopped fibers to precipitated fibers, and the total number of chopped fibers to be generated in the target model are analyzed. .

[0068] The geometric skeleton reconstruction module 30 is used to reconstruct the geometric skeleton according to In addition to the in-plane orientation distribution characteristics, combined with the spatial joint probability density function, the geometric model of the random fiber skeleton composed of short-cut fibers in the target model is reconstructed.

[0069] The connection model establishment module 40 is used to discretize the random fiber skeleton geometric model into a finite element mesh, search for effective contact points between each chopped fiber based on the spatial indexing algorithm, and generate connector units in batches at the effective contact points to equivalently characterize the physical wrapping effect of the precipitated fiber on the chopped fiber, so as to establish a physically equivalent micro-connection model.

[0070] Constitutive model assignment module 50 is used to assign anisotropic coupling failure criteria and quasi-brittle damage evolution model to the connector unit in order to construct a constitutive model of para-aramid paper.

[0071] The prediction module 60 is used to apply multi-condition boundary conditions to the constitutive model and perform explicit dynamic solutions to output macroscopic mechanical response and microscopic damage evolution results.

[0072] Test case

[0073] This test example uses the multidimensional data-driven para-aramid paper mechanical property prediction system 200 from the embodiments, and conducts actual tests according to the multidimensional data-driven para-aramid paper mechanical property prediction method (steps S10~S60) from the embodiments.

[0074] Specifically:

[0075] In step S10: the areal density of the para-aramid paper is measured. 24g / m 2 Bulk density of chopped fibers It is 1.44 g / cm³ 2 .

[0076] In step S20: the length of the target model Set to 15mm, width Set to 15mm thickness The initial mass ratio of chopped fibers to precipitated fibers during the preparation of para-aramid paper was set to 0.03 mm. Based on 1.5, the dimensional parameters of the chopped fibers are: diameter It is 0.03mm in length. It is 5mm. The total number of short-cut fibers as described in the examples. The formula is used to calculate and round the result, yielding a value of 636.

[0077] In step S30: The result of the nonlinear fitting of the inward orientation frequency in step S31 is as follows Figure 3 As shown; in step S33, the in-plane azimuth angle of each chopped fiber is set to follow a Gaussian distribution. The polar angles in the thickness direction follow a restricted uniform distribution in the range of [85º, 95º]; the random fiber skeleton geometric model generated in step S34 is as follows: Figure 4 As shown.

[0078] In step S40: the global mesh size is set to 0.1mm in step S41; the model after setting in step S43 is as follows. Figure 5 (As shown in the full-size virtual style rendering model). Among them, Figure 5 In this context, CD stands for Cross Direction and MD stands for Machine Direction.

[0079] In step S50: The parameter settings for steps S51 and S52 are shown in Table 1 below; the minimum value in step S53 is set to 0.001 mm.

[0080] Table 1 (Material Constitutive and Connector Unit Equivalent Parameter Table)

[0081]

[0082] In step S60:

[0083] For the constitutive model constructed in step S50, two sets of models are cloned using an automated script.

[0084] (1) For the MD model (Model-MD), couple the left and right edge nodes to the reference point and apply a uniaxial stretch displacement of 5.0 mm along the X-axis.

[0085] (2) For the CD model (Model-CD), couple the top and bottom edge nodes to the reference point and apply a stretching displacement along the Y-axis.

[0086] The boundary conditions for the two sub-models, Model-MD and Model-CD, are as follows: Figure 6 As shown.

[0087] After the explicit dynamic solution is completed, the macroscopic tensile response can be obtained by extracting the reaction force (RF) and displacement (U) at the end reference point.

[0088] In the Visualization module, the support reaction force at the fixed end (RP-Fixed) (RF1 in the MD model, RF2 in the CD model) is extracted as the bearing capacity F. The displacement at the loaded end (RP-Load) (U1 in the MD model, U2 in the CD model) is extracted as the elongation ΔL.

[0089] Engineering stress: .

[0090] Engineering strain: In the above formula, This is the initial length of the model.

[0091] Ultimately, the result is as follows Figure 7 The macroscopic stress-strain curve is shown.

[0092] Simultaneously, connector unit damage variables (CDMG) are extracted to obtain, for example... Figure 8 The spatiotemporal evolution cloud map shown is as follows. Figure 8 Part (a) in the diagram represents the global connector damage (CDMG) cloud map before peak stress, part (b) represents the local fiber Mises stress distribution in the corresponding crack region, part (c) represents the global damage cloud map after peak stress, and part (d) represents the global fiber stress cloud map after peak stress. The results show that the method and system of the embodiment accurately predict the longitudinal brittle localized fracture and transverse diffuse progressive softening disintegration characteristics of para-aramid paper.

[0093] Those skilled in the art should understand that this invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to this invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for modeling discrete networks of para-aramid paper based on multidimensional data-driven approaches, wherein the para-aramid paper contains precipitated fibers and chopped fibers, characterized in that, Includes the following steps: S10, Obtain the areal density of the para-aramid paper. Bulk density of chopped fibers The in-plane orientation distribution characteristics of chopped fibers and the physical encapsulation structure characteristics of precipitated fibers over chopped fibers; S20, after setting the target model size, based on , The size parameters of the chopped fibers, the mass ratio of chopped fibers to precipitated fibers, and the total number of chopped fibers to be generated in the target model are analyzed. ; S30, according to In addition to the in-plane orientation distribution characteristics, combined with the spatial joint probability density function, the geometric model of the random fiber skeleton composed of the chopped fibers in the target model is reconstructed. S40, after discretizing the random fiber skeleton geometric model with a finite element mesh, the effective contact points between each chopped fiber are searched based on the spatial indexing algorithm, and connector units are generated in batches at the effective contact points to equivalently characterize the physical wrapping effect of the precipitated fiber on the chopped fiber, so as to establish a physically equivalent micro-connection model. S50, an anisotropic coupling failure criterion and a quasi-brittle damage evolution model are assigned to the connector unit to construct the constitutive model of the para-aramid paper. The anisotropic coupling failure criterion is parameterized based on the physical encapsulation structure characteristics.

2. The method for modeling discrete networks of para-aramid paper based on multidimensional data-driven approach according to claim 1, characterized in that: in, In step S10, the areal density and the volume density Obtained through physical measurement methods The in-plane orientation distribution characteristics of the chopped fibers were measured using X-ray micro-CT. The physical encapsulation structure features were obtained by scanning electron microscopy.

3. The method for modeling discrete networks of para-aramid paper based on multidimensional data-driven approach according to claim 1, characterized in that: in, In step S20, the total number of short-cut fibers to be generated in the target model is calculated using the following formula and then rounded down: ， In the above formula, and These represent the length and width of the target model, respectively. This indicates the mass ratio of chopped fibers to precipitated fibers during the initial preparation of the para-aramid paper. This refers to the diameter data in the dimensional parameters of the chopped fiber. This refers to the length data in the dimensional parameters of the chopped fiber.

4. The method for modeling discrete networks of para-aramid paper based on multidimensional data-driven approach according to claim 1, Its features are: In step S10, the in-plane orientation distribution characteristic of the chopped fibers is the in-plane orientation frequency. Step S30 includes the following sub-steps: S31, the orientation standard deviation is measured by nonlinear fitting of the in-plane orientation frequency; S32, In the Abaqus finite element software, N spatial coordinates are randomly generated in a uniform distribution within the target model as the center position coordinates of each short fiber; S33 sets an in-plane azimuth angle for each chopped fiber, making the in-plane azimuth angle follow a Gaussian distribution with the machine direction as the mean and the orientation standard deviation as the variance, so as to truly reflect the mechanical traction effect brought about by the headbox process used in the preparation of para-aramid paper. A polar angle in the thickness direction is set for each short chopped fiber, and the polar angle is constrained within a limited uniform distribution range near the horizontal plane to simulate the dense layered stacking characteristics caused by the high-temperature hot pressing process used in the preparation of the para-aramid paper. S34. After step S33 is completed, a Boolean merging operation is performed on all the short fibers to obtain the random fiber skeleton geometric model of the para-aramid paper.

5. The method for modeling discrete networks of para-aramid paper based on multidimensional data-driven approach according to claim 4, characterized in that: in, Step S32 is executed based on a Python automated script.

6. The method for modeling discrete networks of para-aramid paper based on multidimensional data-driven approach according to claim 1, Its features are: Step S40 includes the following sub-steps: S41, In Abaqus finite element software, Timoshenko beam elements are used to mesh the geometric model of the random fiber skeleton in order to capture the large deformation and transverse shear effect of the short-cut fibers under stress. S42, using the kd-tree spatial indexing algorithm, traverse all nodes of the beam element in the global scope, search the Euclidean distance between each node to filter out node pairs whose distance is less than the preset search radius threshold and which do not overlap, so as to determine that the location of the node pair is the effective contact point between the chopped fibers. S43, generate discrete connector units in batches at the effective contact points, and establish a local coordinate system for each connector unit that follows the local direction of the intersecting chopped fibers, thereby using the connector unit to equivalently characterize the physical wrapping effect of the precipitated fibers at the intersection of the chopped fibers.

7. The method for modeling discrete networks of para-aramid paper based on multidimensional data-driven approach according to claim 6, characterized in that: in, Step S50 includes the following sub-steps: S51, Based on the local coordinate system, define the coupling mechanical response index of the connector unit under normal tension and tangential shear respectively, so as to construct an anisotropic coupling failure criterion that can reflect the tensile strength and shear weakness of the precipitated fiber to the chopped fiber in the physical wrapping structure characteristics. S52, assign an equivalent elastic stiffness to each connector unit, and set a coupling tensile limit as a damage initiation criterion based on the physical enclosure structure characteristics. Specifically, when the force on the connector unit in the normal or tangential direction exceeds the coupling tensile limit, damage evolution of the connector unit is triggered. S53, the displacement difference of the connector unit from the damage initiation displacement to the complete failure displacement is set to an extremely low value to simulate the physical characteristics of the brittle fracture of the short fiber intersection wrapped by the precipitated fiber after reaching the ultimate bearing capacity, thereby introducing a quasi-brittle damage evolution model for the connector unit, and finally constructing the constitutive model of the para-aramid paper.

8. A multi-dimensional data-driven discrete network modeling system for para-aramid paper, characterized in that, The method for modeling discrete networks of para-aramid paper based on multidimensional data-driven approaches, as described in any one of claims 1 to 7, includes: The data input module allows users to input the measured areal density of para-aramid paper. Bulk density of chopped fibers In-plane orientation distribution characteristics of chopped fibers; The fiber count analysis module is used to analyze the total number of fibers. , The size parameters of the chopped fibers, the mass ratio of chopped fibers to precipitated fibers, and the total number of chopped fibers to be generated in the target model are analyzed. ; The geometric skeleton reconstruction module is used to reconstruct the geometric skeleton based on... In addition to the in-plane orientation distribution characteristics, combined with the spatial joint probability density function, the geometric model of the random fiber skeleton composed of the chopped fibers in the target model is reconstructed. The connection model establishment module is used to discretize the random fiber skeleton geometric model using a finite element mesh, search for effective contact points between each chopped fiber based on a spatial indexing algorithm, and generate connector units at the effective contact points in batches to equivalently characterize the physical wrapping effect of the precipitated fiber on the chopped fiber, thereby establishing a physically equivalent microscopic connection model; and The constitutive model assignment module is used to assign anisotropic coupling failure criteria and quasi-brittle damage evolution model to the connector unit in order to construct the constitutive model of the para-aramid paper.

9. A method for predicting the mechanical properties of para-aramid paper based on multidimensional data-driven approaches, characterized in that, The mechanical performance prediction method is as follows: apply multi-condition boundary conditions to the constitutive model and perform explicit dynamic solution to output the macroscopic mechanical response and microscopic damage evolution results. The constitutive model is the constitutive model constructed by the multidimensional data-driven modeling method for discrete networks of para-aramid paper as described in any one of claims 1 to 7, or the multidimensional data-driven modeling system for discrete networks of para-aramid paper as described in claim 8.

10. A multi-dimensional data-driven system for predicting the mechanical properties of para-aramid paper, characterized in that, include: The multidimensional data-driven discrete network modeling system for para-aramid paper as described in claim 8 is used to output the constitutive model of the para-aramid paper. and The prediction module is used to apply multi-condition boundary conditions to the constitutive model and perform explicit dynamic solutions to output macroscopic mechanical response and microscopic damage evolution results.

Images