A method, system, electronic device and medium for evaluating grain size of a high-temperature alloy
By combining the PSO-SVM algorithm and Voronoi diagram simulation of polycrystalline structure geometry with COMSOL simulation, the destructive and error problems in the evaluation of grain size of high-temperature alloys were solved, and a non-destructive and high-precision evaluation was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-18
- Publication Date
- 2026-03-17
AI Technical Summary
Existing technologies for evaluating the grain size of high-temperature alloys suffer from problems such as high destructiveness, small sample size, and large evaluation model errors.
An evaluation model was constructed using the PSO-SVM algorithm. The model was combined with Voronoi diagrams to simulate the geometric model of the polycrystalline structure and COMSOL simulation to generate simulation data. The grain size was evaluated using ultrasonic velocity and attenuation coefficient.
Without damaging the metal, the error in grain size evaluation was effectively reduced, and the sample size was expanded and the evaluation accuracy was improved.
Smart Images
Figure CN117153308B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of ultrasonic evaluation of the microstructure of metals, and in particular to a method, system, electronic device and medium for evaluating the grain size of high-temperature alloys. Background Technology
[0002] The material properties of nickel-based superalloys are closely related to their metallographic structure. Inhomogeneous grain size in the alloy structure can severely affect creep rupture life, making metallographic analysis and accurate evaluation of nickel-based superalloys essential. However, while traditional metallographic evaluation methods offer high accuracy, they cause irreversible damage to the metal material during the data acquisition process. Furthermore, these methods suffer from drawbacks such as complex procedures, low coverage, and long processing times, and require a limited sample size. While virtual sample expansion methods, such as multi-distribution total diffusion (MD-MTD), can obtain a larger number of samples, the generated samples deviate significantly from the actual situation, leading to large errors in the generated evaluation model. Summary of the Invention
[0003] The purpose of this invention is to provide a method, system, electronic device, and medium for evaluating the grain size of high-temperature alloys, which can overcome the problems of destructiveness and large errors in the grain size evaluation model under conditions of small sample size in the existing technology.
[0004] To achieve the above objectives, the present invention provides the following solution:
[0005] A method for evaluating the grain size of high-temperature alloys, the evaluation method comprising:
[0006] Obtain accurate data on grain size, ultrasonic velocity, and attenuation coefficient of high-temperature alloy samples;
[0007] Based on the actual data of grain size, ultrasonic velocity, and attenuation coefficient of the high-temperature alloy sample, an evaluation model is constructed using the PSO-SVM algorithm; the input of the evaluation model is the ultrasonic velocity and attenuation coefficient; the output of the evaluation model is the grain size.
[0008] Based on the length and width of the preset region of the high-temperature alloy simulation model and the preset number of grains within the preset model region, a geometric model of the polycrystalline structure of the high-temperature alloy simulated by a Voronoi diagram is generated.
[0009] The geometric model of the polycrystalline structure was simulated to obtain simulation data on grain size, ultrasonic velocity, and attenuation coefficient.
[0010] The evaluation model is trained using the simulated and real data of the attenuation coefficient and ultrasonic velocity, as well as the simulated and real data of the corresponding grain size, to obtain the trained evaluation model.
[0011] The attenuation coefficient and ultrasonic velocity data of the high-temperature alloy to be evaluated are input into the trained evaluation model to obtain the grain size evaluation value.
[0012] Optionally, based on the actual data of grain size, ultrasonic velocity, and attenuation coefficient of the high-temperature alloy sample, an evaluation model is constructed using the PSO-SVM algorithm, specifically including:
[0013] Using the real data of the ultrasonic velocity and attenuation coefficient of the high-temperature alloy sample as input and the grain size of the high-temperature alloy sample as output, the PSO algorithm is applied to determine the parameters of the SVM regression model and construct an evaluation model.
[0014] Optionally, the length and width of the preset region of the high-temperature alloy simulation model and the number of preset grains within the preset model region are determined according to the method of measuring the average grain diameter by metallographic experiments.
[0015] Optionally, the geometric model of the polycrystalline structure is simulated to obtain simulation data on grain size, ultrasonic velocity, and attenuation coefficient, specifically including:
[0016] Using a Gaussian pulse modulated sinusoidal pulse signal as the excitation signal, and the probe frequency for obtaining real ultrasonic velocity data of high-temperature alloy samples as the frequency of the excitation signal, the orthogonal anisotropy parameters of the high-temperature alloy simulation model are set, and the fixed mechanics in elastic waves is applied to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic velocity and attenuation coefficient.
[0017] Optionally, the evaluation method further includes:
[0018] The simulation data of the ultrasonic velocity were corrected.
[0019] A high-temperature alloy grain size evaluation system, applied to the above-mentioned high-temperature alloy grain size evaluation method, the evaluation system comprising:
[0020] The acquisition module is used to acquire real data on grain size, ultrasonic velocity, and attenuation coefficient of high-temperature alloy samples;
[0021] The construction module is used to construct an evaluation model based on the real data of grain size, ultrasonic velocity, and attenuation coefficient of the high-temperature alloy sample, using the PSO-SVM algorithm; the input of the evaluation model is the ultrasonic velocity and attenuation coefficient; the output of the evaluation model is the grain size.
[0022] The generation module is used to generate a polycrystalline geometric model of the high-temperature alloy simulation model based on the length and width of the preset region of the high-temperature alloy simulation model and the preset number of grains in the preset model region.
[0023] The simulation module is used to simulate the geometric model of the polycrystalline structure to obtain simulation data of grain size, ultrasonic velocity and attenuation coefficient;
[0024] The training module is used to train the evaluation model using the simulation data and real data of the attenuation coefficient and ultrasonic velocity, as well as the simulation data and real data of the corresponding grain size, as the training dataset, to obtain the trained evaluation model.
[0025] The evaluation module is used to input the attenuation coefficient and ultrasonic velocity data of the high-temperature alloy to be evaluated into the trained evaluation model to obtain the grain size evaluation value.
[0026] An electronic device includes a memory and a processor, the memory storing a computer program, and the processor running the computer program to enable the electronic device to perform the above-described high-temperature alloy grain size evaluation method.
[0027] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for evaluating the grain size of high-temperature alloys.
[0028] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:
[0029] This invention constructs an evaluation model based on real data of grain size, ultrasonic velocity, and attenuation coefficient of the high-temperature alloy sample using the PSO-SVM algorithm. A polycrystalline structure geometric model of the high-temperature alloy, simulated by a Voronoi diagram, is generated based on the length and width of a preset region within the high-temperature alloy simulation model and the preset number of grains within that region. The polycrystalline structure geometric model is simulated to obtain simulated data for grain size, ultrasonic velocity, and attenuation coefficient. The evaluation model is trained using the simulated data and real data for attenuation coefficient and ultrasonic velocity, as well as the corresponding simulated data and real data for grain size, to obtain a trained evaluation model. Applying the trained evaluation model provided by this invention to evaluate the grain size of the high-temperature alloy can effectively reduce the evaluation error of the grain size of the high-temperature alloy. Attached Figure Description
[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0031] Figure 1 This is a flowchart of the high-temperature alloy grain size evaluation method of the present invention.
[0032] Figure 2 This is a flowchart illustrating the practical application of the high-temperature alloy grain size evaluation method of the present invention.
[0033] Figure 3 This is an evaluation diagram showing the effect of the present invention on the grain size of high-temperature alloy GH4169.
[0034] Figure 4 This is a schematic diagram of the geometric model of the polycrystalline structure with N=200 grains in this invention.
[0035] Figure 5 This is a COMSOL simulation result of the present invention with a grain number N=130.
[0036] Explanation of the markings in the accompanying drawings: 1. Trough area; 2. Crest area. Detailed Implementation
[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0038] The purpose of this invention is to provide a method, system, electronic device, and medium for evaluating the grain size of high-temperature alloys, which can overcome the problems of destructiveness and large errors in the grain size evaluation model under conditions of small sample size in the existing technology.
[0039] The high-temperature alloy grain size evaluation method provided by this invention is essentially a method for evaluating high-temperature alloy grain size based on a virtual sample generation technique using COMSOL with MATLAB co-simulation. This method solves the problems of existing technologies that rely on destructive methods to obtain high-temperature alloy grain size characteristics and suffer from large errors in the evaluation model under small sample conditions. The main steps include: First, establishing an evaluation model through POS-SVM analysis of the original metallographic data; second, generating a polycrystalline Voronoi diagram geometric model using MATLAB; third, importing the generated geometric model into COMSOL for reconstruction, setting corresponding modules to simulate the model, and obtaining simulation data; then, adding the simulation data obtained from the COMSOL simulation to the original data to form a mixed sample, and training the model using the previously established training model to finally obtain the evaluation model. Furthermore, this invention has been experimentally verified using a dataset of high-temperature alloy GH4169. Compared to evaluation models generated using small sample real data and virtual sample expansion methods using the Multi-Distribution Overall Diffusion Technique (MD-MTD), this invention can effectively reduce the evaluation error of high-temperature alloy grain size.
[0040] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0041] Example 1
[0042] like Figure 1 and Figure 2 As shown, the present invention provides a method for evaluating the grain size of high-temperature alloys, the evaluation method comprising:
[0043] Step S1: Obtain real data on grain size, ultrasonic velocity, and attenuation coefficient of the high-temperature alloy sample.
[0044] Step S2: Based on the real data of grain size, ultrasonic velocity, and attenuation coefficient of the high-temperature alloy sample, an evaluation model is constructed using the PSO-SVM algorithm; the input of the evaluation model is the ultrasonic velocity and attenuation coefficient; the output of the evaluation model is the grain size.
[0045] Specifically, using the real data of the ultrasonic velocity and attenuation coefficient of the high-temperature alloy sample as input and the grain size of the high-temperature alloy sample as output, the PSO algorithm is applied to determine the parameters of the SVM regression model and construct an evaluation model.
[0046] Step S3: Based on the length and width of the preset region of the high-temperature alloy simulation model and the preset number of grains in the preset model region, generate a geometric model of the polycrystalline structure of the high-temperature alloy simulated by a Voronoi diagram.
[0047] Specifically, the length and width of the preset region of the high-temperature alloy simulation model and the number of preset grains within the preset model region are determined based on the method of measuring the average grain diameter using metallographic experiments.
[0048] In practical applications, the MPT toolbox in MATLAB software is used to define the length and width of the model region and the number of generated grains. Based on the selected parameters, a polycrystalline structure geometric model simulated by the Voronoi diagram is generated.
[0049] As a specific implementation method, the specific parameters of the polycrystalline structure geometric model include defining the length and width of the model region, the number of generated grains N, and the coordinates of the core points of the Voronoi diagram randomly generated within the model. Specifically, first, the size of the model region is defined, that is, the length, width, and number of generated grains N. Within the defined region, N two-dimensional point coordinates (x, y) are randomly generated; these points are also called the core points of the Voronoi diagram. The geometric model is generated based on the core point coordinates and the boundary.
[0050] Step S4: Simulate the geometric model of the polycrystalline structure to obtain simulation data of grain size, ultrasonic velocity and attenuation coefficient.
[0051] Specifically, a sinusoidal pulse signal modulated by a Gaussian pulse is used as the excitation signal, and the probe frequency for obtaining real data of the ultrasonic velocity of the high-temperature alloy sample is used as the frequency of the excitation signal. The orthogonal anisotropy parameters of the high-temperature alloy simulation model are set, and the fixed mechanics in elastic waves is applied to simulate the geometric model of the polycrystalline structure to obtain simulation data of grain size, ultrasonic velocity and attenuation coefficient.
[0052] In practical applications, the COMSOLWithMATLAB plugin is used to import the polycrystalline structure model generated in MATLAB into COMSOL, realize the geometric reconstruction of the polycrystalline structure in COMSOL, set the simulation model for the reconstructed model, set the condition parameters required for simulation, perform simulation, and obtain simulation data of ultrasonic velocity, attenuation and grain size of the model.
[0053] As a specific implementation method, the module selected is solid mechanics in elastic waves, and a simulation instantaneous study is performed on the reconstructed model. The excitation signal is a Gaussian pulse modulated sinusoidal pulse signal, with the following expression:
[0054] f(t)=(1e-6)*sin(2*pi*f*t)*gp(t) (1)
[0055] Where t is the propagation time in seconds, f is the frequency of the excitation signal in MHz, and gp(t) is the Gaussian function.
[0056] Set the orthogonal anisotropy parameters for the high-temperature alloy material, including its density, Young's modulus, Poisson's ratio, and shear modulus. After completing the excitation signal function and material parameter settings, define the boundary probe, start the simulation, and obtain simulation data. Correct the simulated sound velocity data using the following expression:
[0057]
[0058] Among them, V min V max V' represents the maximum and minimum speed of sound in the real sample, respectively; min V' max V' represents the maximum and minimum sound speeds in the virtual sample, respectively. i V i Let be the speed of sound that needs to be corrected and the corrected speed of sound, respectively; N is the number of virtual samples.
[0059] Furthermore, the simulated samples are evenly divided into 10 equal parts and added to the real samples to form a mixed sample. The mixed sample is then used to train the evaluation model established in step S2. The evaluation performance of the trained model is represented by MAPE, and its expression is as follows:
[0060]
[0061] Where N is the number of samples evaluated by the model, y1 is the true value, and y2 is the evaluation value.
[0062] Once trained, the model can be used as an evaluation model for assessing the grain size of high-temperature metals. The input features of the model are the ultrasonic velocity and attenuation coefficient of the grain, and the output feature is the grain size.
[0063] Step S5: Using the simulated and real data of the attenuation coefficient and ultrasonic velocity, as well as the simulated and real data of the corresponding grain size, as the training dataset, the evaluation model is trained to obtain the trained evaluation model.
[0064] In practical applications, the simulation data obtained by COMSOL is added to the real samples to obtain a mixed sample. The mixed sample is then fed into the evaluation model in step S2 for training to obtain the final evaluation model. The input features of the model are the ultrasonic velocity and attenuation coefficient of the simulation sample set and the original sample set, and the output feature is the average grain size of the high-temperature alloy.
[0065] Step S6: Input the attenuation coefficient and ultrasonic velocity data of the high-temperature alloy to be evaluated into the trained evaluation model to obtain the grain size evaluation value.
[0066] As a specific implementation method, taking GH4169 as an example, the present invention will be specifically described as follows:
[0067] (1) First, through experimental operation, the original ultrasonic and metallographic characteristic parameters of 26 different heat treatment states were obtained, including longitudinal wave velocity, attenuation coefficient and grain size. The longitudinal wave velocity ranged from 5600 to 5900 m / s, the attenuation coefficient ranged from 0.34 to 0.081 dB / mm, and the grain size ranged from 30 to 300 μm.
[0068] (2) Divide the 26 sets of data into 21 sets of training data and 5 sets of test data. Use the longitudinal wave velocity and attenuation coefficient of the training data as the input of the SVM regression model and the grain size as the target output of the model. Then, use PSO to find the optimal parameters of the SVM model. Set the maximum number of iterations of PSO to 200. Then, use the optimal parameters to train the SVM network and obtain the training model.
[0069] (3) The grain size of GH4169 under 26 heat treatment conditions ranged approximately from 30 to 300 μm. Following the method of measuring the average grain diameter using metallographic experiments, the grains were equated to a circular shape. Therefore, the grain size D (in μm) and 1 mm... 2 The conversion relationship of the number of grains N in the region is as follows:
[0070]
[0071] (4) Based on formula (4) and the approximate range of the actual grain size of GH4169, the grain size to be simulated is designed. The minimum designed grain size is 30 μm, the step size is 7.5 μm, and the maximum grain size is 322.5 μm. A total of 40 simulation samples are used. After conversion by formula (4), the corresponding number of grains is 12 to 1415.
[0072] (5) In MATLAB software, using the Voronoi diagram model in the MPT toolkit, the length and width of the region are set to 1×1mm. According to (4), 40 Voronoi diagram geometric models with the number of grains ranging from 12 to 1415 are generated respectively, such as Figure 4 The figure shows the geometric model of the generated polycrystalline structure with N=200 grains.
[0073] (6) Import the 40 generated geometric models into the COMSOL software using the COMSOLwithMATLAB plugin, reconstruct the geometric models in COMSOL, and set the simulation model.
[0074] (7) The module is selected as solid mechanics in elastic waves, and the COMSOL simulation transient study is performed on the reconstructed geometric model.
[0075] (8) Set the size of the geometric model to a rectangular domain of 1x1mm and set the excitation signal to a sinusoidal pulse signal modulated by Gaussian pulse, as shown in formula (1).
[0076] (9) Obtain the probe frequency used in the experiment from the original ultrasound data, set the simulated excitation signal frequency f to 10MHz, set the position parameter of the Gaussian function to 2 / f, and the standard deviation to 0.5 / f.
[0077] (10) Set the orthotropic parameters of GH4169 material.
[0078] (11) Define a boundary probe. The function of the boundary probe is to receive signals. It is equivalent to the receiving probe in actual ultrasonic testing. The position is set at the bottom edge of the geometric model. The ultrasonic signal obtained in the experiment is simulated by monitoring the displacement component in the y direction of the bottom edge.
[0079] (12) Because the material parameters of the sample (such as Young's modulus, density and density, etc.) differ greatly from the material parameters used in the simulation, the ultrasonic velocity in the simulation needs to be corrected. The correction formula is shown in Equation (2).
[0080] (13) Simulation experiments were conducted on 40 geometric models to obtain virtual sound velocity, attenuation, and grain size data for 40 simulations, such as... Figure 5 The figure shows the simulation results in COMSOL when the number of crystal grains N = 130. The figure marks the peak region 2 and the trough region 1, which represent the ultrasonic waves.
[0081] (14) Divide the 40 generated sample data into 10 equal parts, select one value from each part, and obtain a total of ten values. Use the 80 / 20 rule to divide these 10 sets of data into test set and training set, add them to the original metallographic data to form mixed data. The number of test data in the mixed data is 21 original metallographic data plus 8 simulation data, for a total of 29 training data. The test dataset is 5 original metallographic data plus 2 simulation data, for a total of 7 test data.
[0082] (15) Input 29 training data into the established evaluation model, train it, and obtain the evaluation model after mixed sample training. Input 7 test data into the trained evaluation model to obtain the evaluation value. The effect of the model is expressed by equation (3).
[0083] (16) such as Figure 3 As shown, the effect of the present invention can be seen intuitively. Table 1 lists the evaluation index MAPE of the model. From the table, we can see the evaluation error between the original metallographic data and the method of the present invention based on the Multi-Distribution Overall Diffusion Technology (MD-MTD) method.
[0084] Table 1 Evaluation metrics of the model
[0085]
[0086] Compared with existing technologies, the advantages of this invention are:
[0087] 1. To evaluate the grain size of metal by using the speed and attenuation of ultrasound without damaging the metal.
[0088] 2. To address the small sample size problem in the model, the simulated samples effectively expand the real sample space, increasing the number of samples while also improving the accuracy of model evaluation.
[0089] Example 2
[0090] To implement the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, a high-temperature alloy grain size evaluation system is provided below. The evaluation system includes:
[0091] The acquisition module is used to acquire real data on grain size, ultrasonic velocity, and attenuation coefficient of high-temperature alloy samples.
[0092] The construction module is used to construct an evaluation model based on the real data of grain size, ultrasonic velocity, and attenuation coefficient of the high-temperature alloy sample, using the PSO-SVM algorithm; the input of the evaluation model is the ultrasonic velocity and attenuation coefficient; the output of the evaluation model is the grain size.
[0093] The generation module is used to generate a geometric model of the polycrystalline structure of the high-temperature alloy simulated by a Voronoi diagram, based on the length and width of a preset region of the high-temperature alloy simulation model and the preset number of grains within the preset model region.
[0094] The simulation module is used to simulate the geometric model of the polycrystalline structure to obtain simulation data of grain size, ultrasonic velocity and attenuation coefficient.
[0095] The training module is used to train the evaluation model using the simulated and real data of the attenuation coefficient and ultrasonic velocity, as well as the simulated and real data of the corresponding grain size, as training datasets, to obtain the trained evaluation model.
[0096] The evaluation module is used to input the attenuation coefficient and ultrasonic velocity data of the high-temperature alloy to be evaluated into the trained evaluation model to obtain the grain size evaluation value.
[0097] Example 3
[0098] This invention provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor runs the computer program to enable the electronic device to execute the high-temperature alloy grain size evaluation method of Embodiment 1.
[0099] Alternatively, the aforementioned electronic device may be a server.
[0100] In addition, this embodiment of the invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the high-temperature alloy grain size evaluation method of Embodiment 1.
[0101] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.
[0102] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.
Claims
1. A method for evaluating grain size of a high-temperature alloy, characterized by, The evaluation method comprises: obtaining real data of grain size, ultrasonic speed and attenuation coefficient of a high-temperature alloy sample; constructing an evaluation model by using a PSO-SVM algorithm according to the real data of grain size, ultrasonic speed and attenuation coefficient of the high-temperature alloy sample; the input of the evaluation model is the ultrasonic speed and the attenuation coefficient; and the output of the evaluation model is the grain size; generating a polycrystalline structure geometric model of the high-temperature alloy simulation model simulated by a Voronoi diagram according to the length, width and preset number of grains in a preset region of the high-temperature alloy simulation model; simulating the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic speed and attenuation coefficient, specifically comprising: using a Gaussian pulse modulated sine pulse signal as an excitation signal, using a probe frequency of the real data of the ultrasonic speed of the high-temperature alloy sample as the frequency of the excitation signal, setting the orthotropic anisotropy parameters of the high-temperature alloy simulation model, simulating the polycrystalline structure geometric model by using elastic wave neutralization and fixed mechanics to obtain the simulation data of grain size, ultrasonic speed and attenuation coefficient; the excitation signal is a Gaussian pulse modulated sine pulse signal, and the expression is as follows: ； wherein t is the propagation time, the unit is s, f is the frequency of the excitation signal, the unit is MHz, and gp(t) is a Gaussian function; training the evaluation model by using the simulation data and the real data of the attenuation coefficient and the ultrasonic speed and the simulation data and the real data of the grain size as a training data set to obtain a trained evaluation model; inputting the data of the attenuation coefficient and the ultrasonic speed of the high-temperature alloy to be evaluated into the trained evaluation model to obtain a grain size evaluation value.
2. The high temperature alloy grain size evaluation method according to claim 1, characterized by, constructing an evaluation model by using a PSO-SVM algorithm according to the real data of grain size, ultrasonic speed and attenuation coefficient of the high-temperature alloy sample, specifically comprising: using the real data of the ultrasonic speed and the attenuation coefficient of the high-temperature alloy sample as the input and using the grain size of the high-temperature alloy sample as the output, determining the parameters of the SVM regression model by using a PSO algorithm to construct the evaluation model.
3. The high temperature alloy grain size evaluation method according to claim 1, characterized by, determining the length, width and preset number of grains in a preset region of the high-temperature alloy simulation model according to a method of measuring the average grain diameter by metallographic experiment.
4. The high temperature alloy grain size evaluation method according to claim 1, characterized by, The evaluation method further comprises: correcting the simulation data of the ultrasonic speed, and the expression is as follows: where V min , V max are the maximum and minimum values of the sound speed in the real sample, respectively; V min , V max are the maximum and minimum values of the sound speed in the virtual sample, respectively; V i , V i are the i-th sound speed to be corrected and the corrected sound speed, respectively; and N is the number of virtual samples.
5. A high temperature alloy grain size evaluation system, characterized by, The evaluation system comprises: an obtaining module configured to obtain real data of grain size, ultrasonic speed and attenuation coefficient of a high-temperature alloy sample; a constructing module configured to construct an evaluation model by using a PSO-SVM algorithm according to the real data of grain size, ultrasonic speed and attenuation coefficient of the high-temperature alloy sample; the input of the evaluation model is the ultrasonic speed and the attenuation coefficient; and the output of the evaluation model is the grain size; a generating module configured to generate a polycrystalline structure geometric model of the high-temperature alloy simulation model simulated by a Voronoi diagram according to the length, width and preset number of grains in a preset region of the high-temperature alloy simulation model; The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: ； The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes:
6. An electronic device, comprising: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes:
7. A computer-readable storage medium, characterized in that, The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data of grain size, ultrasonic sound speed and attenuation coefficient, and specifically includes: The simulation module is configured to simulate the polycrystalline structure geometric model to obtain simulation data