A method for calibrating parameters of child car seat foam materials based on viscoelastic constitutive model
By directly calibrating the parameters of the foam material for child car seats through energy absorption tests and optimizing the model using ISIGHT software, the problem of inaccurate simulation of the energy absorption characteristics of foam materials in existing technologies has been solved, achieving faster and more accurate simulation results.
Patent Information
- Application Number
- CN202410192411.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-02-21
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2044-02-21
AI Technical Summary
Existing technologies cannot accurately represent the energy absorption characteristics of foam materials used in child car seats, resulting in inaccurate simulation results and extending the design cycle.
A viscoelastic constitutive method was adopted to directly calibrate the parameters of foam materials through energy absorption experiments, establish a finite element simulation model, optimize the parameters using ISIGHT software, and perform automated verification using software such as ANSA, LS-DYNA, and PYTHON.
This study achieved accurate simulation of the energy absorption characteristics of foam materials, shortened the design cycle, and improved the accuracy and reliability of simulation results.
Smart Images

Figure CN117995327B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of simulation technology for child car seat foam materials, and more specifically, to a method for parameter calibration of car seat foam materials. Background Technology
[0002] With the continuous increase in car ownership in my country, people are using cars more frequently, leading to a more frequent traffic accident rate. As a result, the safety of child occupants in collisions is receiving increasing attention. Children, due to their smaller size and incomplete physical development, present significant challenges compared to adult occupants. To reduce injuries to child occupants in accidents, child safety seats are being provided to improve their protection.
[0003] Foam materials in child car seats play a particularly important role in protecting children, absorbing collision energy and mitigating impact. Accurately simulating the energy absorption effect of foam materials during a collision is a crucial aspect of child car seat design simulation. For child car seat manufacturers, accurately representing the energy absorption characteristics of foam materials is a key focus in child car seat design simulation. The selection of appropriate foam materials can shorten the design cycle and improve the energy absorption effect of the foam, making it significant for child car seat design.
[0004] The stress-strain curve of a foam material consists of two parts: the loading stage and the unloading stage. The FU-CHANG-FOAM model is used to simulate low-density and medium-density foams and can be used to simulate foam materials for child car seats. The stress-strain curve in the loading stage can be defined by the user, while the stress-strain curve in the unloading stage is adjusted using Young's modulus E, hysteresis unloading factor HU, damping DAMP, and the shape factor SHAPE during unloading. The unloading stage determines whether the energy absorption characteristics of the foam material meet practical requirements; therefore, the parameter calibration of foam materials focuses on the simulation of the unloading stage.
[0005] The energy absorption test involves dropping a metal head mold from a certain height to the center of a foam board placed on a flat surface, and measuring the acceleration time history curve of the metal head mold from the moment of contact with the foam board. Currently, the parameter calibration of foam materials mainly relies on simulation calibration using the stress-strain curve during the unloading phase of compression. After calibrating the stress-strain curve during the unloading phase, an energy absorption test is needed to verify the energy absorption characteristics of the foam. The energy absorption characteristics of foam play a crucial role in the use of child car seats and need to be reflected more accurately. Summary of the Invention
[0006] The purpose of this invention is to overcome the problem that existing child seat foam material parameter calibration methods cannot accurately express the energy absorption characteristics of foam materials. It provides a child seat foam material parameter calibration method based on viscoelastic constitutive model. Instead of calibrating material parameters through stress-strain curves during the unloading stage, it directly calibrates material parameters through energy absorption tests, thereby achieving accurate simulation of the energy absorption characteristics of child seat foam materials, ensuring the accuracy of simulation results, and shortening the development cycle of child seats.
[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:
[0008] A method for calibrating parameters of child car seat foam materials based on viscoelastic constitutive modeling includes the following steps:
[0009] S1: Conduct energy absorption tests on foam materials, collect acceleration data of the head mold during the energy absorption test, and obtain the acceleration time history curve of the head mold;
[0010] S2: Establish a finite element simulation model, extract parameters for material calibration, modify the parameters, and perform calculations to solve the finite element simulation model;
[0011] S3: Process the acceleration data of the head model in the energy absorption test, build a parameter optimization model, perform parameter optimization iterative calculations, and calibrate the parameters.
[0012] Furthermore, energy absorption tests were conducted on the foam material, and acceleration data of the head mold was collected during the energy absorption test to obtain the acceleration time history curve of the head mold, specifically:
[0013] The head mold is installed on the test device, and a foam board is placed directly below the head mold. An acceleration sensor is installed on the head mold, allowing it to fall freely and collide with the foam board. At the same time, a high-speed camera is used to film the process of the head mold falling and contacting the foam board.
[0014] Data from the accelerometer on the head mold is read, and the acceleration data is filtered to obtain the initial acceleration time history curve C0 of the head mold in the vertical direction.
[0015] Furthermore, a finite element simulation model is established, including:
[0016] Mesh generation: The 3D model of the head mold used in the energy absorption test of foam material is imported into HYPERMESH software, and a 3D model is established according to the size of the foam used in the test; in the software, hexahedral solid mesh is generated for the foam structure, tetrahedral solid mesh is generated for the head mold, and shell element mesh is generated on the surface of the foam and the head mold to establish the contact relationship between the head mold and the foam;
[0017] Mesh model attribute assignment: Set solid elements for hexahedral solid mesh and tetrahedral solid mesh, and specify the head model as a rigid body; set two-dimensional elements for shell element mesh.
[0018] Material assignment for parts: A FU-CHANG-FOAM constitutive model is established based on the stress-strain curve of the foam material. The loading stage of the stress-strain curve is set, and the initial values of the material parameters Young's modulus E, damping DAMP, hysteresis unloading factor HU, and unloading shape factor SHAPE are assigned.
[0019] Model constraints and contact settings: Contact settings are implemented using shell element meshes of the head mold and foam surface, and the vertical degree of freedom of the bottom of the foam is constrained;
[0020] Loading settings: Set the velocity load on the entire head mold according to the velocity at which the head mold impacts the foam during the energy absorption test;
[0021] Output results: Set the software to output the overall acceleration history data of the head model; once the finite element simulation model is established, output the original k file test.k.
[0022] Furthermore, the parameters used for material calibration are extracted and modified, specifically as follows:
[0023] Open the original k file test.k in ANSA software and create a new optimization task OPTIMIZATION_TASK;
[0024] Add the design variable DESIGN_VARIABLE to OPTIMIZATION_TASK, adding the foam material parameters E, DAMP, HU, and SHAPE as design variables and specifying their ranges;
[0025] Set the output file to the modified k file test1.k, set the txt file used for parameter modification to DVFile.txt, and run the optimization task;
[0026] Save the parameter adjustment program to the file opt.ansa.
[0027] Furthermore, the finite element simulation model is calculated and solved, specifically as follows:
[0028] The modified k-file test1.k of the finite element simulation model was solved using LS-DYNA software, and the result file nodout was obtained.
[0029] Furthermore, the acceleration data of the head model in the energy absorption test were processed, specifically as follows:
[0030] The initial acceleration time history curve C0 of the head mold when the foam is impacted vertically by the head mold in the energy absorption test is processed to obtain the target acceleration time history curve C1, so that the sampling frequency of the test data is the same as the frequency of the finite element simulation result data.
[0031] Furthermore, a parameter optimization model is built, specifically as follows:
[0032] A parameter optimization model was built using one Exploration module and four Simcode modules in the ISIGHT software. The Exploration module is the program module, the Simcode module is the application module, and the four Simcode modules are connected in series. The four Simcode modules are named Simcode module 1, Simcode module 2, Simcode module 3 and Simcode module 4.
[0033] Simcode module 1 runs the ANSA program to set parameters; Simcode module 2 runs the LS-DYNA program to perform energy absorption test simulation; Simcode module 3 runs the LS-PREPOST program to output nodal acceleration data; and Simcode module 4 runs the PYTHON program to output curve correlation coefficients and mean square errors.
[0034] Furthermore, after the parameter optimization model framework is built, the application module and program module are configured with file input, parameterization, data transfer settings, optimization algorithm selection, and target setting. Each module is specifically configured as follows:
[0035] The input file for Simcode module 1 is DVFile.txt, which is the parameter modification file called by ANSA. The parameters to be modified include E, DAMP, HU, and SHAPE. The command file is opt.ansa, an ANSA file used to adjust material parameters when calling the parameter modification file. The output file does not need to be set.
[0036] The input file for Simcode module 2 does not need to be set; the command file is a batch file that calls LS-DYNA for simulation and solution, and the simulation and solution are performed by calling this file; the output file does not need to be set.
[0037] The input file for Simcode module 3 does not need to be set; the command file is the acceleration data file of the head model obtained by post-processing the simulation results using LS-PREPOST. Its contents include reading the vertical acceleration data of nodout and outputting the acceleration time history curve C2 to the data file acc.csv.
[0038] The input file for Simcode module 4 does not need to be set; the command file calls PYTHON to compare the acceleration time history curve C2 in the obtained data file acc.csv with the head model acceleration time history curve C1 obtained after energy absorption test post-processing, and outputs the correlation coefficient R and mean square error MSE of the two curves to the result file bata.txt; the output file is the correlation coefficient R and mean square error MSE in the result file bata.txt.
[0039] The Exploration module offers multiple combined strategies. The selected strategy is the Approximation Loop, a global optimization algorithm based on approximation model updates. This algorithm is suitable for problems in nonlinear design spaces. The Exploration module includes four settings: option settings, variable settings, constraint settings, and objective settings.
[0040] Furthermore, the option settings, variable settings, constraint settings, and target settings in the Exploration module are as follows:
[0041] Option settings: Maximum number of sub-optimization runs selected as 20; Optimization method selected as NLPQLP as the preferred algorithm;
[0042] Variable settings: Select E, DAMP, HU, and SHAPE as optimization variables, and set -75% of the initial value of the variables as the lower limit and +75% as the upper limit.
[0043] Constraint settings: No constraints;
[0044] Objective settings: The correlation coefficient R and the mean squared error MSE are set as objectives, with the goal of maximizing the correlation coefficient R and minimizing the MSE.
[0045] Furthermore, the global optimization algorithm based on approximate model update first establishes an approximate model of the local design space, performs optimization calculations on the approximate model to obtain an estimate of the local optimal solution, and continuously updates the position of the approximate model through verification calculations to continue optimization, finally obtaining a verified global solution;
[0046] The optimization iterative calculation is performed in the ISIGHT software. The entire iteration process is recorded in the History interface of the running portal. After the iteration is completed, the optimal solution of the optimization model after actual simulation is given. The desired foam material parameters can be obtained without solving and verifying in LS-DYNA again.
[0047] Compared with the prior art, the present invention has the following advantages and beneficial effects:
[0048] This invention considers the energy absorption characteristics of foam materials for material parameter calibration. Instead of fitting the stress-strain curve during the unloading stage of the foam material, it directly fits the results from the energy absorption test, reducing unnecessary time and enabling a more accurate simulation of the properties of foam materials.
[0049] This invention modifies the foam parameters involved in parameter calibration by calling ANSA software, which can avoid the problem of k-file errors caused by excessive data character length in the optimization results obtained during the optimization design process, and can realize the automatic verification process.
[0050] This invention employs a global optimization algorithm based on approximate model updates, which can continuously verify the effectiveness of the approximate model during the global optimization process. Furthermore, the optimization results obtained are verified by simulation, eliminating the need for additional simulation to verify the material parameter calibration results. This simplifies the process and reduces unnecessary time. Attached Figure Description
[0051] Figure 1 This is a flowchart of the method for calibrating parameters of child seat foam materials based on viscoelastic constitutive model according to the present invention.
[0052] Figure 2 This is a schematic diagram of the parameter optimization model established in the ISIGHT software for the foam material parameter calibration method of this invention. Detailed Implementation
[0053] The method for calibrating parameters of child seat foam materials based on viscoelastic constitutive model according to the present invention will be further described below with reference to the accompanying drawings and specific embodiments.
[0054] Currently, the parameters of child car seat foam materials are calibrated by simulating the characteristics of the foam material during the unloading phase of compression tests. However, this method cannot accurately represent the energy absorption characteristics of foam materials. In collision simulations using foam materials, the energy absorption characteristics demonstrate the cushioning and energy-absorbing effect of the foam, playing a crucial role in protecting child occupants. Therefore, accurately representing the energy absorption characteristics of foam materials is an important issue.
[0055] This invention provides a parameter calibration method for foam materials based on the FU-CHANG-FOAM model, which accurately expresses the energy absorption characteristics of foam materials and completes the parameter calibration of child car seat foam materials. Because this invention uses a global optimization algorithm based on approximate model updates, it can automatically verify the parameters during the parameter optimization process, ensuring the accuracy and reliability of the calibrated parameter results and eliminating the need for manual verification of the optimal parameters obtained after calibration.
[0056] Please see Figure 1 This invention discloses a method for calibrating parameters of child seat foam materials based on viscoelastic constitutive modeling, comprising the following steps:
[0057] S1: Conduct energy absorption tests on foam materials, collect acceleration data of the head mold during the energy absorption test, and obtain the acceleration time history curve of the head mold;
[0058] S2: Establish a finite element simulation model, extract parameters for material calibration, modify the parameters, and perform calculations to solve the finite element simulation model;
[0059] S3: Process the acceleration data of the head model in the energy absorption test, build a parameter optimization model, perform parameter optimization iterative calculations, and calibrate the parameters.
[0060] In step S1, an energy absorption test of the foam material is conducted. Acceleration data of the head mold is collected during the test, and its acceleration time-history curve is obtained. Specifically, the head mold is installed on the test device, and a foam board is placed directly below it. An acceleration sensor is installed on the head mold. The head mold is allowed to fall freely and collide with the foam board. Simultaneously, a high-speed camera captures the process of the head mold falling and contacting the foam board. The data from the acceleration sensor on the head mold is read, and after filtering, the initial acceleration time-history curve C0 of the head mold in the vertical direction is obtained.
[0061] In step S2, a finite element simulation model is established, including mesh generation, assigning mesh model properties, assigning part materials, setting model constraints and contact, setting loading, and outputting results.
[0062] Grid generation
[0063] The 3D model of the 6kg head mold used in the energy absorption experiment of foam materials was imported into HYPERMESH software. A 3D model of the foam board (486.2mm*486.2mm*50mm) was then created directly below the head mold to ensure contact between the two models, thus starting the simulation from the moment the head mold collides with the foam. In the software, a hexahedral solid mesh was created for the foam structure, and a tetrahedral solid mesh was created for the head mold. This ensured that the mesh shapes of the head mold and the foam were consistent in the vertical projection, increasing the accuracy of the simulation. The surface meshes of the foam and the head mold were extracted, and shell element meshes sharing nodes were created to establish the contact relationship between the head mold and the foam.
[0064] Mesh model attribute assignment: Set solid elements for hexahedral solid meshes and tetrahedral solid meshes, and specify the head model as a rigid body to shorten the simulation time. Set shell element meshes to two-dimensional elements.
[0065] Material assignment for parts: Establish a FU-CHANG-FOAM constitutive model for the foam, set the loading stage of the stress-strain curve, and assign initial values to the material parameters Young's modulus E, damping DAMP, hysteresis unloading factor HU, and unloading shape factor SHAPE based on the stress-strain curve of the foam material.
[0066] Model constraints and contact settings: Contact settings are achieved using shell element meshes of the head mold and foam surface. Since it is a contact between foam and metal, soft contact with soft=1 is used, the friction coefficient is set to 0.3, and the vertical degree of freedom of the bottom of the foam is constrained.
[0067] Loading settings: Based on the height difference of 0.1m between the head mold and the foam during the energy absorption test, the overall velocity load of the head mold is set to 1.4m / s using the keyword *INITIAL_VELOCITY.
[0068] Output results: The software is configured to output the overall historical acceleration data of the head model, with the output frequency matching the frequency of the actual energy absorption test. Once the finite element simulation model is established, the original k-file test.k is output.
[0069] Since the optimal parameter results of the optimized design may occupy more than 10 characters, and k-files have a limit on the number of characters that can be used for data, exceeding this limit will result in an error. Therefore, using ANSA software to modify the parameters allows the parameter optimization results to be automatically and smoothly imported into the finite element simulation model for verification calculations.
[0070] Specifically, in step S2, the parameters used for material calibration are extracted and modified, including:
[0071] Open the original k file test.k in ANSA software and create a new optimization task OPTIMIZATION_TASK;
[0072] Add the design variable DESIGN_VARIABLE to OPTIMIZATION_TASK, adding the foam material parameters E, DAMP, HU, and SHAPE as design variables, and giving them a range of 0-500 so that these design variables can be adjusted within the set range;
[0073] Set the output file to the modified k file test1.k, set the txt file used for parameter modification to DVFile.txt, and run the optimization task;
[0074] Save the parameter adjustment program to the file opt.ansa for future parameter modifications.
[0075] In step S2, the finite element simulation model is calculated and solved. Specifically, the modified k file test1.k of the finite element simulation model is calculated and solved using LS-DYNA software to obtain the result file nodout.
[0076] In step S3, the acceleration data of the head mold in the energy absorption test is processed. Specifically, the initial acceleration time history curve C0 of the head mold when the foam is impacted vertically by the head mold in the energy absorption test is processed to obtain the target acceleration time history curve C1, so that the sampling frequency of the test data is the same as the frequency of the finite element simulation result data.
[0077] Please see Figure 2 This invention uses ISIGHT software to build a parameter optimization model, which is built using one Exploration module and four Simcode modules in ISIGHT software. The Exploration module is a program module, the Simcode module is an application module, and the four Simcode modules are connected in series. The four Simcode modules are named Simcode module 1, Simcode module 2, Simcode module 3 and Simcode module 4.
[0078] Simcode module 1 runs the ANSA program to set parameters; Simcode module 2 runs the LS-DYNA program to perform energy absorption test simulation; Simcode module 3 runs the LS-PREPOST program to output nodal acceleration data; and Simcode module 4 runs the PYTHON program to output curve correlation coefficients and mean square errors.
[0079] After the parameter optimization model framework is built, the application module and program module are configured with file input, parameterization, data transfer settings, optimization algorithm selection, and target setting. Each module is specifically configured as follows:
[0080] The input file for Simcode module 1 is DVFile.txt, which is the parameter modification file called by ANSA. The parameters to be modified include E, DAMP, HU, and SHAPE. The command file is a batch file (.bat) that uses the ANSA file opt.ansa to adjust material parameters when calling the parameter modification file. The output file does not need to be set.
[0081] The input file for Simcode module 2 does not need to be set; the command file is a batch file that calls LS-DYNA for simulation and solution, and the simulation and solution are performed by calling this file; the output file does not need to be set.
[0082] The input file for Simcode module 3 does not need to be set; the command file is the acceleration data file of the head model obtained by post-processing the simulation results using LS-PREPOST. Its contents include reading the vertical acceleration data of nodout and outputting the acceleration time history curve C2 to the data file acc.csv.
[0083] The input file for Simcode module 4 does not need to be set; the command file calls PYTHON to compare the acceleration time history curve C2 in the obtained data file acc.csv with the head model acceleration time history curve C1 obtained after energy absorption test post-processing, and outputs the correlation coefficient R and mean square error MSE of the two curves to the result file bada.txt; the output file is the correlation coefficient R and mean square error MSE in the result file bada.txt.
[0084] The Exploration module offers multiple combined strategies. We selected the Approximation Loop strategy, a global optimization algorithm based on approximation model updates. This algorithm is suitable for problems in nonlinear design spaces. The Exploration module includes four settings: option settings, variable settings, constraint settings, and objective settings. Each setting is detailed below:
[0085] Option settings: Select 20 for the maximum number of sub-optimization runs (i.e., the maximum number of sub-optimizations), and select NLPQLP as the preferred algorithm for the optimization method.
[0086] Variable settings: Select E, DAMP, HU, and SHAPE as optimization variables, and set -75% of the initial value of the variables as the lower limit and +75% as the upper limit.
[0087] Constraint settings: No constraints.
[0088] Objective settings: The correlation coefficient R and the mean squared error MSE are set as objectives, with the goal of maximizing the correlation coefficient R and minimizing the MSE.
[0089] This invention employs a global optimization algorithm based on an approximate model update. First, an approximate model of the local design space is established, and optimization calculations are performed on this approximate model to obtain an estimate of the local optimum. Then, through verification calculations, the position of the approximate model is continuously updated for further optimization, ultimately yielding a verified global solution.
[0090] Optimization iterative calculations were performed in the ISIGHT software, and the entire iteration process was recorded in the History interface of the launch portal. After 15 iterations, the optimal solution of the optimized model, verified by actual simulation, was obtained, with a correlation coefficient R of 0.9878 and a mean square error MSE of 0.4302. The validated and calibrated foam material parameters were obtained as follows: DAMP = 0.482, E = 67.63 MPa, HU = 0.001, and SHAPE = 21.167.
[0091] In summary, this invention first implemented a finite element simulation process: model building and k-file output were performed using HYPERMESH software, and material parameter adjustments were made using ANSA software. Then, a parameter optimization process was implemented: the energy absorption characteristic curve of the foam material, i.e., the acceleration history curve of the head mold, was obtained through actual energy absorption experiments. A parameter optimization process was then established in ISIGHT software, using batch files to call software such as ANSA, LS-DYNA, LS-PREPOST, and PYTHON to perform parameter modifications, simulation runs, post-simulation processing, and data processing. This automated the parameter optimization process and enabled automatic verification, reflecting the energy absorption characteristics of the foam and ensuring the reliability of the optimization results.
[0092] The above description is a detailed description of the preferred embodiments of the present invention. However, the embodiments are not intended to limit the scope of the patent application of the present invention. All equivalent changes or modifications made under the technical spirit disclosed in the present invention should fall within the patent scope covered by the present invention.
Claims
1. A method for calibrating parameters of child car seat foam materials based on viscoelastic constitutive model, characterized in that, Includes the following steps: S1: Conduct energy absorption tests on foam materials, collect acceleration data of the head mold during the energy absorption test, and obtain the acceleration time history curve of the head mold, specifically: The head mold is installed on the test device, and a foam board is placed directly below the head mold. An acceleration sensor is installed on the head mold, allowing it to fall freely and collide with the foam board. At the same time, a high-speed camera is used to film the process of the head mold falling and contacting the foam board. Read the data from the accelerometer on the head mold, and after filtering, obtain the initial acceleration time history curve C0 of the head mold in the vertical direction; S2: Establish a finite element simulation model, extract parameters for material calibration, modify the parameters, and perform calculations to solve the finite element simulation model; Establishing a finite element simulation model includes: Mesh generation: The 3D model of the head mold used in the energy absorption test of foam material is imported into HYPERMESH software, and a 3D model is established according to the size of the foam used in the test; in the software, hexahedral solid mesh is generated for the foam structure, tetrahedral solid mesh is generated for the head mold, and shell element mesh is generated on the surface of the foam and the head mold to establish the contact relationship between the head mold and the foam; Mesh model attribute assignment: Set solid elements for hexahedral solid mesh and tetrahedral solid mesh, and specify the head model as a rigid body; set two-dimensional elements for shell element mesh. Material assignment for parts: A FU-CHANG-FOAM constitutive model is established based on the stress-strain curve of the foam material. The loading stage of the stress-strain curve is set, and the initial values of the material parameters Young's modulus E, damping DAMP, hysteresis unloading factor HU, and unloading shape factor SHAPE are assigned. Model constraints and contact settings: Contact settings are implemented using shell element meshes of the head mold and foam surface, and the vertical degree of freedom of the bottom of the foam is constrained; Loading settings: Set the velocity load on the entire head mold according to the velocity at which the head mold impacts the foam during the energy absorption test; Results output: Set the software to output the overall historical acceleration data of the head model; after the finite element simulation model is established, output the original k file test.k; S3: Process the acceleration data of the head model in the energy absorption test, build a parameter optimization model, perform parameter optimization iterative calculation, and calibrate the parameters; The parameter optimization model is built as follows: A parameter optimization model was built using one Exploration module and four Simcode modules in the ISIGHT software. The Exploration module is the program module, the Simcode module is the application module, and the four Simcode modules are connected in series. The four Simcode modules are named Simcode module 1, Simcode module 2, Simcode module 3 and Simcode module 4. Simcode module 1 runs the ANSA program to set parameters; Simcode module 2 runs the LS-DYNA program to perform energy absorption test simulation; Simcode module 3 runs the LS-PREPOST program to output nodal acceleration data; and Simcode module 4 runs the PYTHON program to output curve correlation coefficients and mean square errors.
2. The method for calibrating parameters of child car seat foam materials based on viscoelastic constitutive model according to claim 1, characterized in that, Extract the parameters used for material calibration and modify them, specifically: Open the original k file test.k in ANSA software and create a new optimization task OPTIMIZATION_TASK; Add the design variable DESIGN_VARIABLE to OPTIMIZATION_TASK, adding the foam material parameters E, DAMP, HU, and SHAPE as design variables and specifying their ranges; Set the output file to the modified k file test1.k, set the txt file used for parameter modification to DVFile.txt, and run the optimization task; Save the parameter adjustment program to the file opt.ansa.
3. The method for calibrating parameters of child car seat foam materials based on viscoelastic constitutive model according to claim 2, characterized in that, The finite element simulation model is solved by calculation, specifically as follows: The modified k-file test1.k of the finite element simulation model was solved using LS-DYNA software, and the result file nodout was obtained.
4. The method for calibrating parameters of child car seat foam materials based on viscoelastic constitutive model according to claim 3, characterized in that, The acceleration data of the head model in the energy absorption test were processed as follows: The initial acceleration time history curve C0 of the head mold when the foam is impacted vertically by the head mold in the energy absorption test is processed to obtain the target acceleration time history curve C1, so that the sampling frequency of the test data is the same as the frequency of the finite element simulation result data.
5. The method for calibrating parameters of child car seat foam materials based on viscoelastic constitutive model according to claim 1, characterized in that, After the parameter optimization model framework is built, the application module and program module are configured with file input, parameterization, data transfer settings, optimization algorithm selection, and target setting. Each module is specifically configured as follows: The input file for Simcode module 1 is DVFile.txt, which is the parameter modification file called by ANSA. The parameters to be modified include E, DAMP, HU, and SHAPE. The command file is opt.ansa, an ANSA file used to adjust material parameters when calling the parameter modification file. The output file does not need to be set. The input file for Simcode module 2 does not need to be set; the command file is a batch file that calls LS-DYNA for simulation and solution. The simulation and solution are performed by calling this file. No settings are required for the output file; The input file for Simcode module 3 does not need to be set; the command file is the acceleration data file of the head model obtained by post-processing the simulation results using LS-PREPOST. Its contents include reading the vertical acceleration data of nodout and outputting the acceleration time history curve C2 to the data file acc.csv. The input file for Simcode module 4 does not need to be set; the command file calls PYTHON to compare the acceleration time history curve C2 in the obtained data file acc.csv with the head model acceleration time history curve C1 obtained after energy absorption test post-processing, and outputs the correlation coefficient R and mean square error MSE of the two curves to the result file bata.txt; the output file is the correlation coefficient R and mean square error MSE in the result file bata.txt. The Exploration module offers multiple combined strategies. The selected strategy is the ApproximationLoop global optimization strategy based on approximation model updates. This global optimization algorithm based on approximation model updates is suitable for problems in nonlinear design spaces. The Exploration module includes four settings: option settings, variable settings, constraint settings, and objective settings.
6. The method for calibrating parameters of child car seat foam materials based on viscoelastic constitutive model according to claim 5, characterized in that, The options, variables, constraints, and objectives in the Exploration module are as follows: Option settings: Maximum number of sub-optimization runs selected as 20; Optimization method selected as NLPQLP as the preferred algorithm; Variable settings: Select E, DAMP, HU, and SHAPE as optimization variables, and set -75% of the initial value of the variables as the lower limit and +75% as the upper limit. Constraint settings: No constraints; Objective settings: The correlation coefficient R and the mean squared error MSE are set as objectives, with the goal of maximizing the correlation coefficient R and minimizing the MSE.
7. The method for calibrating parameters of child seat foam materials based on viscoelastic constitutive model according to claim 5, characterized in that, The global optimization algorithm based on approximate model update first establishes an approximate model of the local design space, performs optimization calculations on the approximate model to obtain an estimate of the local optimal solution, and continuously updates the position of the approximate model through verification calculations to continue optimization, and finally obtains the verified global solution. The optimization iterative calculation is performed in the ISIGHT software. The entire iteration process is recorded in the History interface of the running portal. After the iteration is completed, the optimal solution of the optimization model after actual simulation is given. The desired foam material parameters can be obtained without solving and verifying in LS-DYNA again.
