A method for identifying multiple high-temperature mechanical properties of materials based on a single loading test.

Abstract

This invention discloses a method for identifying multiple high-temperature mechanical properties of materials simultaneously based on a single loading test. First, a sample with a relatively uniform temperature field and a complex strain field is selected as the initial sample using finite element method (FEM). Then, its characteristic dimensions are optimized to obtain an optimized specimen. Next, the optimized specimen is divided into several regions, and the sensitivity of all properties to be identified in each region is calculated to determine the dominant region. Subsequently, the experimental force load, boundary constraints, and the processed temperature field are used as boundary conditions, and the mechanical properties are used as optimization variables to establish a finite element model equivalent to the experiment. Finally, the mechanical properties are identified within the dominant regions, and the identification results are input into the finite element model for verification, achieving the identification of multiple high-temperature material mechanical properties in a single test. This method organically combines material configuration design, high-temperature mechanical testing, high-temperature DIC technology, and optimization algorithms, enabling the simultaneous acquisition of high-temperature mechanical properties of multiple materials based on a single loading test.

Description

Technical Field

[0001] This invention belongs to the field of material mechanical property identification and relates to a high-temperature mechanical experimental method. Specifically, it relates to an identification method that combines material configuration design, high-temperature mechanical testing, high-temperature DIC technology, and optimization algorithms to simultaneously obtain multiple high-temperature mechanical properties. Background Technology

[0002] Hypersonic vehicles experience severe aerodynamic heating during atmospheric flight, with surface temperatures reaching up to 2500°C. The operational safety of their surface thermal protection structures under such extreme high temperatures and complex loads faces significant challenges. The design of practical thermal protection structures requires obtaining material properties through high-temperature mechanical experiments to analyze structural reliability. The accuracy of experimental data significantly impacts structural design, computational model verification, and structural strength analysis. The level of precision determines the accuracy of structural characterization and can further introduce potential failure risks at the material, component, and even overall vehicle levels, potentially threatening the overall safety of the aircraft.

[0003] Current high-temperature mechanical testing methods suffer from limited accuracy due to the inherent complexity of the experiments themselves. Furthermore, most researchers employ uniaxial tension, shear, and compression methods for performance measurement, resulting in low efficiency as each experiment can only obtain a single property, and the inability to measure certain key properties, such as Poisson's ratio, under high-temperature conditions. For expensive composite materials such as C / C and C / SiC composites, there are also problems such as large sample requirements, long preparation cycles, lengthy experimental periods, and high testing costs. The lack of efficient and low-cost methods for identifying the mechanical properties of materials at high temperatures severely restricts the accurate analysis of material structural responses and the assessment of structural load-bearing capacity in high-temperature environments.

[0004] Therefore, establishing a method for identifying multiple high-temperature mechanical properties of materials based on a single loading test is of great significance for reducing the economic and time costs of actual research processes, and is of great value for the performance analysis and safety assessment of materials under high-temperature service conditions. Summary of the Invention

[0005] To address the aforementioned problems in the background technology, this invention provides a method for identifying the high-temperature mechanical properties of multiple materials simultaneously based on a single loading test. This method organically combines material configuration design, high-temperature mechanical testing, high-temperature DIC technology, and optimization algorithms, enabling the simultaneous acquisition of the high-temperature mechanical properties of multiple materials based on a single loading test.

[0006] The objective of this invention is achieved through the following technical solution:

[0007] A method for identifying multiple high-temperature mechanical properties of materials based on a single loading test includes the following steps:

[0008] Step 1: Sample shape and size design:

[0009] Step 1: For each pattern, establish a thermal / electric coupling analysis model using finite element method (FEM) to analyze the uniformity of the temperature field during electric heating. Then, for each pattern model, calculate the strain field distribution of the compressed specimen at high temperature using FEM, selecting a temperature field within the fluctuation range. The most complex strain pattern was used as the initial specimen. The value is selected based on the conductivity and size of different materials;

[0010] Steps 1 and 2: Optimize the characteristic dimensions of the preliminary sample as design parameters, and calculate the variance of the strain components. The uniformity coefficient serves as an optimization design constraint. Based on the sample size and pattern, a suitable optimization algorithm is selected to ensure that the temperature field uniformity coefficient is minimized while maximizing the variance of the strain components. Specifically:

[0011] Strain component variance Defined as:

[0012]

[0013] In the formula, For strain at each node, The average value of the nodal strain. This represents the total number of nodes.

[0014] The temperature field uniformity coefficient is defined as:

[0015]

[0016] In the formula, This represents the maximum temperature at a node in the finite element method. This represents the minimum temperature at a node in the finite element method. The average temperature across all nodes;

[0017] Step 2: Determining the advantageous partitions based on the sensitivity analysis algorithm:

[0018] Step 2: Divide the sample into m regions according to the geometric configuration standard. ;

[0019] Step 22: Based on the identified materials Given a priori mechanical property, set the sampling mean for each mechanical property to be identified, define the sampling standard deviation and upper and lower limits, and then randomly and independently sample all mechanical properties. Next, the finite element model is substituted into the standard tensile calculation, and the input mechanical properties and each partition are calculated. Output the strain field database as a sample library;

[0020] Step Two Three: Using the sample library generated in Step Two Two, and based on the existing FAST sensitivity calculation method, calculate the partitions. The corresponding number Sensitivity to individual mechanical properties In all regions Following the method described above, calculate the sensitivity of all mechanical properties to be identified. , ;

[0021] Step 24: The sum of the sensitivities of all unidentified mechanical properties within each partition is denoted as... , ,Right now Calculate based on the number of partitions Select Maximum sensitivity The corresponding regions, as areas with superior mechanical properties, replace the overall structure in the subsequent identification process. The value is ;

[0022] Step 3: High-Temperature Standard Mechanical Tests and Finite Element Model Establishment

[0023] Step 3: 1. Complete the loading experiment using a high-temperature mechanical property testing platform, and measure the strain field of the experimental advantageous zone using a DIC device. The experimental temperature field data were obtained by measuring with a thermal imager. ;

[0024] Step 3.2: Realizing the experimental temperature field based on linear interpolation By interpolating the finite element model, extracting the coordinates of the finite element model elements to establish a reference coordinate system, and performing coordinate transformation and interpolation matching on the temperature field pixels of the thermal imager, the spatial correspondence between the experimental temperature field and the numerical model is achieved, thus obtaining the temperature field data of the finite element model. ;

[0025] Step 3: Set the force load, boundary constraints, and temperature field in the experiment as boundary conditions, and use the mechanical properties corresponding to each dominant region as optimization variables to establish a finite element model equivalent to the experiment, and calculate the strain field of the dominant region. ;

[0026] Step 4: Identification of multiple mechanical properties based on genetic algorithm:

[0027] Step 41: Identify the mechanical properties within the advantageous partition using a performance inversion method based on a genetic algorithm, and set the objective function... As the objective function of the genetic algorithm:

[0028]

[0029] In the formula, , , Finite element analysis of strain field in advantageous regions With DIC measurement of strain field The three strain component fields The normalized difference; using each mechanical property as the optimization variable, perform parallel iterative optimization of multiple properties; set thresholds according to actual conditions. The performance combination is continuously updated through genetic operators during the iteration process, when the objective function... Converging to a set threshold At that time, the optimal solution for each material property is output as the identification result;

[0030] Step 42: Verify the identified high-temperature mechanical properties. Input the identification results into the finite element model for back-calculation verification. Quantitatively compare the calculated strain cloud map with the experimental results. Take the simulated strain and experimental strain at all node positions of the finite element model. When the average relative error (MRE) is less than a certain threshold, the identification results of this set of performance parameters are considered reliable, and the single identification is completed.

[0031] Compared with the prior art, the present invention has the following advantages:

[0032] 1. This invention increases the complexity of the material strain field under unidirectional load through material configuration design.

[0033] 2. This invention divides the material into sections and performs sensitivity analysis on the different mechanical properties of different regions. It only uses the regions with high sensitivity for performance identification, resulting in higher accuracy.

[0034] 3. This invention utilizes a thermal imager to measure the temperature field, and the temperature field data is interpolated and reconstructed into a finite element numerical model, taking into account the influence of the temperature field on the mechanical properties of materials.

[0035] 4. This invention utilizes a genetic algorithm for multiple iterations to obtain the properties of multiple materials in a single experiment, thereby reducing costs and shortening the experimental cycle.

[0036] 5. This invention provides a complete method from designing the shape and size of the sample to determining the advantageous partitions based on the sensitivity analysis algorithm, to establishing high-temperature standard mechanical tests and finite element models, and then to identifying multiple mechanical properties based on genetic algorithms, providing support for the simultaneous identification of multiple mechanical properties of materials at high temperatures. Attached Figure Description

[0037] Figure 1The flowchart shows a method for identifying multiple high-temperature mechanical properties of materials based on a single loading test.

[0038] Figure 2 The following is a detailed flowchart of an example. Detailed Implementation

[0039] The technical solution of the present invention will be further described below with reference to the accompanying drawings, but it is not limited thereto. Any modifications or equivalent substitutions to the technical solution of the present invention that do not depart from the spirit and scope of the technical solution of the present invention should be covered within the protection scope of the present invention.

[0040] This invention provides a method for identifying multiple high-temperature mechanical properties of materials simultaneously based on a single loading test. First, using the finite element method, a preliminary specimen with a relatively uniform temperature field and complex strain field is selected from various specimen styles capable of generating complex strains. Then, its characteristic dimensions are optimized to obtain an optimized specimen. Next, the optimized specimen is divided into several regions according to its geometric configuration. A sensitivity algorithm is used to calculate the sensitivity of all properties to be identified in each region to determine the dominant region. Subsequently, a high-temperature mechanical test is conducted to obtain the test temperature field and test strain field. The test force load, boundary constraints, and the processed temperature field are used as boundary conditions, and the mechanical properties are used as optimization variables to establish a finite element model equivalent to the experiment. Finally, a performance inversion method based on a genetic algorithm is used to identify the mechanical properties within the dominant region, and the identification results are input into the finite element model for verification, achieving the identification of multiple high-temperature material mechanical properties in a single test. Figure 1 As shown, the specific steps are as follows:

[0041] Step 1: Sample Shape and Size Design: This step aims to maximize the complexity of the strain field distribution in the inversion region while ensuring the uniformity of the temperature field at high temperatures, thus providing a feasible sample configuration for subsequent high-temperature performance inversion experiments. The specific steps are as follows:

[0042] Step 1: Using various shape design methods such as Meuwissen specimens, Haddadi specimens, and perforated specimens, a thermal / electrical coupling analysis model is established for each shape to analyze the uniformity of the temperature field under electric heating. Then, for each shape model, the strain field distribution of the compressed specimen at high temperature is calculated using finite element method, selecting a temperature field within the fluctuation range. The most complex strain pattern was used as the initial specimen.

[0043] Steps 1 and 2: Define the variance of strain components at each node in the preliminary finite element model of the specimen. :

[0044]

[0045] In the formula, For strain at each node, The average value of the nodal strain. This represents the total number of nodes.

[0046] The temperature field uniformity coefficient is defined as:

[0047]

[0048] In the formula, This represents the maximum temperature at a node in the finite element method. This represents the minimum temperature at a node in the finite element method. The average temperature across all nodes.

[0049] Step 13: Optimize the characteristic dimensions of the preliminary specimen (such as opening radius, cutting radius, and cutting center position) as design parameters, and calculate the variance of the strain components. The uniformity coefficient serves as an optimization design constraint. Based on the specimen size and style, a suitable optimization algorithm is selected to ensure the uniformity coefficient is minimized while maximizing the variance of the strain components. In this case, the specimen size and style are used as optimized specimens in the subsequent identification process.

[0050] Step 2: Determining the advantageous regions based on the sensitivity analysis algorithm: This step avoids the problem of unsatisfactory identification results caused by the low sensitivity of some regions to mechanical properties in traditional methods, effectively improving the accuracy of the results. The specific steps are as follows:

[0051] Step 2: Divide the sample into m regions according to the geometric configuration standard. .

[0052] Step 22: Based on the identified materials Given a priori mechanical property, set the sampling mean for each mechanical property to be identified, define the sampling standard deviation and upper and lower limits, and then randomly and independently sample all mechanical properties. Next, the finite element model is substituted into the standard tensile calculation, and the input mechanical properties and each partition are calculated. The output strain field database serves as a sample library.

[0053] Step Two Three: Using the sample library generated in Step Two Two, and based on the existing FAST sensitivity calculation method, calculate the partitions. The corresponding number Sensitivity to individual mechanical properties In all regions Following the method described above, calculate the sensitivity of all mechanical properties to be identified. , .

[0054] Step 24: The sum of the sensitivities of all unidentified mechanical properties within each partition is denoted as... ,Right now Calculate based on the number of partitions Select Maximum sensitivity The corresponding regions, as areas with superior mechanical properties, replace the overall structure in the subsequent identification process. The value is .

[0055] Step 3: High-Temperature Standard Mechanical Test and Finite Element Model Establishment: This step ensures the consistency between the model input and the experimental boundary conditions, guaranteeing the reliability of the inversion identification results. The specific steps are as follows:

[0056] Step 3: 1. Complete the loading experiment using the existing high-temperature mechanical property testing platform (T / CSCM 03-2023), and use DIC equipment to measure the strain field of the experimental advantageous zones. The experimental temperature field data were obtained by measuring with a thermal imager. .

[0057] Step 3.2: Realizing the experimental temperature field based on linear interpolation By interpolating the finite element model, extracting the coordinates of the finite element model elements to establish a reference coordinate system, and performing coordinate transformation and interpolation matching on the temperature field pixels of the thermal imager, the spatial correspondence between the experimental temperature field and the numerical model is achieved, thus obtaining the temperature field data of the finite element model. .

[0058] Step 3: Set the force load, boundary constraints, and temperature field in the experiment as boundary conditions, and use the mechanical properties corresponding to each dominant region as optimization variables to establish a finite element model equivalent to the experiment. The strain field of the dominant region can then be calculated. .

[0059] Step 4: Identification of Multiple Mechanical Properties Based on Genetic Algorithm: This step overcomes the shortcomings of traditional methods that require multiple independent experiments to obtain various properties, and achieves parallel identification of multiple high-temperature mechanical properties. The specific steps are as follows:

[0060] Step 41: Identify mechanical properties within the advantageous partition using a performance inversion method based on a genetic algorithm; set the objective function... As the objective function of the genetic algorithm:

[0061]

[0062] In the formula, , , Finite element analysis of strain field in advantageous regions With DIC measurement of strain field The three strain component fields The normalized difference is calculated. Multiple mechanical properties are optimized using parallel iterative optimization. Thresholds are set based on the actual conditions such as specimen size and shape. In the iterative process, the performance combination is continuously updated through genetic operators such as selection, crossover, and mutation, when the objective function... Converging to a set threshold At that time, the optimal solution for each material property is output as the identification result.

[0063] Step 42: Verify the identified high-temperature mechanical properties. Input the identification results into the finite element model for back-calculation verification. Quantitatively compare the calculated strain cloud map with the experimental results. Take the simulated strain and experimental strain at all node positions of the finite element model. When the average relative error (MRE) is less than a certain threshold, such as 10%, the identification results of this set of performance parameters are considered reliable, and the single identification is completed.

[0064] Example:

[0065] This embodiment provides a complete method from designing the shape and size of the specimen to determining the advantageous partitions based on the sensitivity analysis algorithm, to establishing a high-temperature standard mechanical test and finite element model, and then to identifying multiple mechanical properties based on a genetic algorithm. The method includes the following steps:

[0066] Step 1: Taking C / C composite materials as an example, the overall process framework of the example is as follows: Figure 2 As shown, a thermoelectric coupling model and a high-temperature compression model were established for each sample in the finite element method. The circular notch sample with the most uniform temperature field and the most complex strain field was selected as the initial sample.

[0067] Step 2: Cutting radius Width at the narrowest point of the sample Strain component variance The uniformity coefficient is used as an optimization design constraint, and the optimized sample size is obtained using the simple downhill method as the optimization algorithm, such as... Figure 2 As shown, the optimized sample participates in the subsequent identification process.

[0068] Step 3: Divide the entire recognition area into 12 sub-regions based on symmetry. .

[0069] Step 4: In the area There are three properties to be identified, namely elastic modulus shear modulus Poisson's ratio This embodiment utilizes the prior conditions of C / C composite materials, assuming that the true value of the material's elastic modulus is... The true value of Poisson's ratio is shear modulus A database of input parameters was established by randomly sampling 1000 times using ±20% of the elastic modulus, ±0.15 of the Poisson's ratio, and ±20% of the shear modulus. These parameters were then substituted into the finite element software for calculation to obtain the strain field of each region. The input mechanical properties and their corresponding strain fields for each region were used as the sample database.

[0070] Step 5: Using the sample library generated in Step 4, and based on existing FAST sensitivity calculation methods, calculate the first... The corresponding partition is the first Sensitivity to individual mechanical properties In all regions Following the method described above, the sensitivity of the three mechanical properties to be identified is calculated. .

[0071] Step 6: Obtain the sensitivity result based on the calculation in Step 5. ,calculate Select the largest The six corresponding advantageous regions are used in the subsequent identification process to replace the overall structure.

[0072] Step 7: Conduct a tensile loading test using a high-temperature mechanical property testing platform. The temperature environment of the testing platform is set to... The strain field was measured using a high-temperature DIC device. The thermal imager measured the temperature field. .

[0073] Step 8: In the temperature field image obtained by the thermal imager, each pixel has a specific coordinate in the reference coordinate system; by searching outward from the coordinates of each element in the finite element model and using the interpolation function built into the mathematical software, the temperature field data corresponding to that element coordinate is obtained, i.e., the temperature field of the finite element model. .

[0074] Step 9: The force load is a displacement load c mm / min, which is applied until d mm is reached and then stopped. The boundary constraint is that the lower end is fixed. The temperature field is measured in Step 8. A finite element model is established, and the strain field can be calculated. .

[0075] Step 10: Set the threshold The elastic modulus was obtained through iterative optimization using a genetic algorithm. Poisson's ratio shear modulus .

[0076] Step 11: Input the results measured in Step 10 into the finite element model for calculation to obtain the strain field simulation results. Calculate the error between the strain at the nodes and the experimental values. If the average relative error is within 10%, the identification results are considered reliable.

Claims

1. A method for identifying multiple high-temperature mechanical properties of materials based on a single loading test, characterized in that... The method includes the following steps: Step 1: Sample shape and size design: Step 1: For each pattern, establish a thermal / electric coupling analysis model using finite element method (FEM) to analyze the uniformity of the temperature field during electric heating. Then, for each pattern model, calculate the strain field distribution of the compressed specimen at high temperature using FEM, selecting a temperature field within the fluctuation range. The most complex strain pattern was used as the initial specimen. Steps 1 and 2: Optimize the characteristic dimensions of the preliminary sample as design parameters, and calculate the variance of the strain components. The uniformity coefficient is used as an optimization design constraint. An appropriate optimization algorithm is selected according to the sample size and pattern to ensure that the temperature field uniformity coefficient is minimized while maximizing the variance of the strain components. Step 2: Determining the advantageous partitions based on the sensitivity analysis algorithm: Step 2: Divide the sample into m regions according to the geometric configuration standard. ; Step 22: Based on the identified materials Given a priori mechanical property, set the sampling mean for each mechanical property to be identified, define the sampling standard deviation and upper and lower limits, and then randomly and independently sample all mechanical properties. Next, the finite element model is substituted into the standard tensile calculation, and the input mechanical properties and each partition are calculated. Output the strain field database as a sample library; Step Two Three: Using the sample library generated in Step Two Two, and based on the existing FAST sensitivity calculation method, calculate the partitions. The corresponding number Sensitivity to individual mechanical properties In all regions Following the method described above, calculate the sensitivity of all mechanical properties to be identified. , ; Step 24: The sum of the sensitivities of all unidentified mechanical properties within each partition is denoted as... , ,Right now Calculate based on the number of partitions Select Maximum sensitivity The corresponding regions, as areas with superior mechanical properties, replace the overall structure in the subsequent identification process. The value is ; Step 3: High-Temperature Standard Mechanical Tests and Finite Element Model Establishment Step 3:

1. Complete the loading experiment using a high-temperature mechanical property testing platform, and measure the strain field of the experimental advantageous zone using a DIC device. The experimental temperature field data were obtained by measuring with a thermal imager. ; Step 3.2: Realizing the experimental temperature field based on linear interpolation By interpolating the finite element model, extracting the coordinates of the finite element model elements to establish a reference coordinate system, and performing coordinate transformation and interpolation matching on the temperature field pixels of the thermal imager, the spatial correspondence between the experimental temperature field and the numerical model is achieved, thus obtaining the temperature field data of the finite element model. ; Step 3: Set the force load, boundary constraints, and temperature field in the experiment as boundary conditions, and use the mechanical properties corresponding to each dominant region as optimization variables to establish a finite element model equivalent to the experiment, and calculate the strain field of the dominant region. ; Step 4: Identification of multiple mechanical properties based on genetic algorithm: Step 41: Identify the mechanical properties within the advantageous partition using a performance inversion method based on a genetic algorithm, and set the objective function... Using various mechanical properties as the objective function of the genetic algorithm, multi-performance parallel iterative optimization is performed; thresholds are set according to the actual situation. The performance combination is continuously updated through genetic operators during the iteration process, when the objective function... Converging to a set threshold At that time, the optimal solution for each material property is output as the identification result; Step 42: Verify the identified high-temperature mechanical properties. Input the identification results into the finite element model for back-calculation verification. Quantitatively compare the calculated strain cloud map with the experimental results. Take the simulated strain and experimental strain at all node positions of the finite element model. When the average relative error (MRE) is less than a certain threshold, the identification results of this set of performance parameters are considered reliable, and the single identification is completed.

2. The identification method for simultaneously obtaining multiple high-temperature mechanical properties of materials based on a single loading test, as described in claim 1, is characterized in that... The variance of the strain components Defined as: In the formula, For strain at each node, The average value of the nodal strain. This represents the total number of nodes.

3. The identification method for simultaneously obtaining multiple high-temperature mechanical properties of materials based on a single loading test, as described in claim 1, is characterized in that... The temperature field uniformity coefficient is defined as follows: In the formula, This represents the maximum temperature at a node in the finite element method. This represents the minimum temperature at a node in the finite element method. The average temperature across all nodes.

4. The identification method for simultaneously obtaining multiple high-temperature mechanical properties of materials based on a single loading test, as described in claim 1, is characterized in that... The objective function The calculation formula is: In the formula, , , Finite element analysis of strain field in advantageous regions With DIC measurement of strain field The three strain component fields The normalized difference.

Images