High-precision modeling method of porous metal material based on random algorithm and application of high-precision modeling method in simulation of electrothermal mechanical properties of material

Abstract

The invention provides a high-precision modeling method of a porous metal material based on a random algorithm and application of the high-precision modeling method in electrothermal mechanical performance simulation of a material, and relates to the technical field of artificial intelligence algorithms, and the method comprises the following steps: reconstructing and optimizing an initial reconstruction model of the porous metal material based on the random algorithm to obtain a three-dimensional reconstruction model; performing slice analysis on the three-dimensional reconstruction model, and determining whether an analysis result meets a precision error standard or not; and if yes, outputting a three-dimensional reconstruction model. By reconstructing and optimizing the initial model, the three-dimensional structure of the porous material can be accurately simulated, and high-precision modeling of the porous metal material is achieved. And pore seeds are generated through a random algorithm and grow, so that the randomness and authenticity of the model are ensured. And the reliability of the model is further improved by setting the precision error standard, and the precision of the porosity and the aperture is ensured. The error problem in the porous metal material modeling process is effectively solved, and a solid foundation is provided for high-precision simulation and material performance prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence algorithms, and particularly relates to a high-precision modeling method for porous metal materials based on a random algorithm and its application in simulating the electro-thermal-mechanical properties of materials. Background Art

[0002] At present, due to their unique structural characteristics, porous materials have broad application prospects in fields such as electronic packaging, energy storage, aerospace, and biomedicine. In particular, for porous structures with non-uniform random pore distributions and homogeneous metal or single-metal matrix materials, their macroscopic properties are closely related to microscopic pore characteristics, especially porosity and pore size distribution. These two key parameters directly affect the core properties of materials such as electrical conductivity, thermal conductivity, mechanical strength, and dielectric properties. Therefore, accurately characterizing and reproducing porosity and pore size distribution is crucial in high-precision structural modeling.

[0003] Currently, commonly used pore structure characterization methods include scanning electron microscopy (SEM), X-ray tomography (X-CT), and mercury intrusion porosimetry (MIP), etc. However, these methods are complex to operate, expensive in equipment, time-consuming in analysis, and difficult to perform in-situ real-time monitoring. In addition, existing numerical modeling methods are mostly based on idealized periodic or regular pore assumptions, making it difficult to truly reproduce the non-uniform random pore characteristics of materials, resulting in a large gap between simulation results and actual material properties. At the same time, due to the lack of an accurate structure-property mapping model, it is difficult to accurately predict material properties under high frequency, high temperature, and complex loads, restricting the performance optimization and engineering applications of porous materials.

[0004] Therefore, there is an urgent need to propose a high-precision modeling method for randomly distributed pores, single or homogeneous systems, which can quickly reconstruct the real pore structure and effectively predict its electrical, thermal, mechanical, and other properties, providing a theoretical basis and computational support for the performance optimization and engineering design of porous materials. Summary of the Invention

[0005] One of the purposes of the present invention is to provide a high-precision modeling method for porous metal materials based on a random algorithm to solve the problems in the background art.

[0006] A high-precision modeling method for porous metal materials based on a random algorithm provided by an embodiment of the present invention includes:

[0007] S1. Based on the random algorithm, reconstruct and optimize the initial reconstruction model of the porous metal material to obtain a three-dimensional reconstruction model;

[0008] S2. Analyze the three-dimensional reconstruction model by slicing and determine whether the analysis result meets the accuracy error standard;

[0009] S3. If it meets the standard, output the three-dimensional reconstruction model.

[0010] Optionally, the initial reconstruction model is generated based on initial reconstruction parameters;

[0011] Among them, the initial reconstruction parameters at least include: defining the model size, the growth probability in each initial direction, the pore seed probability, and the initial volume porosity.

[0012] Optionally, the steps for reconstructing and optimizing the initial reconstruction model include:

[0013] S11. Randomly generate a number of pore seeds in the initial reconstruction model according to the pore seed probability; each pore seed grows randomly towards its neighboring cells;

[0014] S12. Assign a random number between 0 and 1 to each neighboring cell; if the random number assigned to the neighboring cell is less than the growth probability in the corresponding direction, the corresponding neighboring cell is occupied by the pore phase;

[0015] S13. Repeat S11 to S12 until the initial reconstruction model reaches the initially set target volume porosity;

[0016] S14. Use a noise filter to smooth the pore edges of the initial reconstruction model that reaches the target volume porosity.

[0017] Optionally, the steps for slicing and analyzing the three-dimensional reconstruction model include:

[0018] Uniformly cut the three-dimensional reconstruction model into multiple two-dimensional slices along a preset direction to match the cross-section sampling method during actual SEM observation;

[0019] Calculate the simulated average porosity and simulated average pore diameter of the multiple two-dimensional slices respectively, and use them as the analysis results.

[0020] Optionally, the accuracy error criteria include:

[0021] The errors between the simulated average porosity and the simulated average pore diameter in the analysis results and the target average two-dimensional porosity and the target average pore diameter are less than or equal to the threshold.

[0022] Optionally, the steps for obtaining the target average two-dimensional porosity and the target average pore diameter include:

[0023] Determine the target average two-dimensional porosity and the target average pore diameter based on at least a preset number of scanning electron microscope images of the porous metal material.

[0024] Optionally, the noise filter at least includes: Kernel filter, median filter, and Gaussian filter.

[0025] Optionally, if the analysis result does not meet the accuracy error standard, adjust the initial reconstruction parameters of the initial reconstruction model, and re-execute S1 to S3 after adjustment.

[0026] Optionally, the growth probability at least includes: isotropic and anisotropic.

[0027] The application of any of the methods provided by the embodiments of the present invention in the simulation of the electro-thermal mechanical properties of materials includes:

[0028] Perform electro-thermal mechanical property simulation on the output three-dimensional reconstruction model on a finite element simulation software platform;

[0029] Among them, the finite element simulation software platform at least includes: Ansys, COMSOL Multiphysics, and Abaqus.

[0030] The present invention has achieved the following beneficial effects:

[0031] 1. By reconstructing and optimizing the initial model, the present invention can accurately simulate the three-dimensional structure of porous materials, realizing high-precision modeling of porous metal materials. Generating pore seeds through a random algorithm and growing them ensures the randomness and authenticity of the model. The introduction of slice analysis verifies the accuracy of the three-dimensional model, ensuring that the simulation results are consistent with actual observations. The setting of the accuracy error standard further improves the reliability of the model, ensuring the accuracy of porosity and pore size. Overall, it effectively solves the error problem in the modeling process of porous metal materials, providing a solid foundation for high-precision simulation and material property prediction.

[0032] 2. The present invention is applicable to porous homogeneous metal materials with randomly distributed pore structures, breaking through the technical limitations of traditional modeling based on regular periodic structures, having stronger material adaptability and microstructural expression ability, especially suitable for engineering application scenarios with complex morphology and strong structural randomness, effectively expanding the applicable boundary of this type of method in real material modeling.

[0033] 3. The modeling process of the present invention is simple and efficient. Relying on the easily accessible two-dimensional SEM image information, the three-dimensional pore structure can be reconstructed without relying on high-cost experimental means such as X-ray tomography, significantly reducing the technical threshold and time cost of model establishment, and being applicable to batch and automated modeling scenarios. At the same time, the present invention introduces the porosity and pore size distribution parameters extracted from actual images as constraint conditions, improving the reduction degree of the model to the real pore morphology and providing an accurate structural input basis for subsequent performance simulation.

[0034] 4. The three-dimensional pore structure model generated by the present invention has good finite element simulation compatibility and can be directly imported into multiple mainstream multi-physics analysis platforms to carry out performance simulations such as thermology, electricity, and mechanics, supporting the prediction of structural responses and functional evaluations under complex working conditions. After the simulation results are processed by structural correction and error feedback optimization, they have a high degree of consistency with the actual test data, and the error is controlled within an acceptable range (5%), which can be widely applied to the research of material structure-activity relationships, performance prediction, and engineering reliability evaluation, with good accuracy guarantee and application value.

[0035] 5. The present invention realizes the linkage integration among scanning electron microscope (SEM) image recognition, three-dimensional pore structure reconstruction, and finite element simulation, establishing a modeling process integrating image input, stochastic modeling, and performance prediction, overcoming problems such as inconsistent data interfaces and scattered processing steps in traditional methods.

[0036] Other features and advantages of the present invention will be described in the following specification, and part of them will become obvious from the specification or be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in the written specification and the drawings.

[0037] The technical solutions of the present invention will be further described in detail below through the drawings and embodiments. Description of the Drawings

[0038] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0039] Figure 1 It is a schematic diagram of a high-precision modeling method for porous metal materials based on a stochastic algorithm in an embodiment of the present invention;

[0040] Figure 2 It is a schematic diagram of the growth direction of pores in three-dimensional space in an embodiment of the present invention;

[0041] Figure 3 It is a schematic diagram of SEM images of sintered joints under different pore gradients in an embodiment of the present invention;

[0042] Figure 4 It is a schematic diagram of the results of thermoelectric performance simulations under different pore gradients and the simulated thermal / electrical conductivity of joints under different pore gradients in an embodiment of the present invention;

[0043] Figure 5 It is another schematic diagram of the results of thermoelectric performance simulations under different pore gradients and the simulated thermal / electrical conductivity of joints under different pore gradients in an embodiment of the present invention. Detailed Embodiments

[0044] The preferred embodiments of the present invention will be described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only for explaining and understanding the present invention, and are not used to limit the present invention.

[0045] Embodiment 1:

[0046] The embodiment of the present invention provides a high-precision modeling method for porous metal materials based on a random algorithm, as Figure 1 shown, including:

[0047] S1. Based on the random algorithm, reconstruct and optimize the initial reconstruction model of the porous metal material to obtain a three-dimensional reconstruction model;

[0048] S2. Analyze the three-dimensional reconstruction model by slicing and determine whether the analysis result meets the accuracy error standard;

[0049] S3. If it meets the standard, output the three-dimensional reconstruction model.

[0050] The initial reconstruction model is generated based on the initial reconstruction parameters;

[0051] Among them, the initial reconstruction parameters at least include: defining the model size, the growth probability in each initial direction, the pore seed probability, and the initial volume porosity;

[0052] The steps of reconstructing and optimizing the initial reconstruction model include:

[0053] S11. Randomly generate a number of pore seeds in the initial reconstruction model according to the pore seed probability; each pore seed grows randomly towards its neighboring cells;

[0054] S12. Assign a random number between 0 and 1 to each neighboring cell; if the neighboring cell is assigned a random number less than the growth probability in the corresponding direction, the corresponding neighboring cell is occupied by the pore phase;

[0055] S13. Repeat S11 to S12 until the initial reconstruction model reaches the initially set target volume porosity;

[0056] S14. Use a noise filter to smooth the pore edges of the initial reconstruction model that reaches the target volume porosity;

[0057] The steps of analyzing the three-dimensional reconstruction model by slicing include:

[0058] Uniformly cut the three-dimensional reconstruction model into a plurality of two-dimensional slices along a preset direction to match the cross-section sampling method during actual SEM observation;

[0059] Calculate the simulated average porosity and the simulated average pore diameter of the plurality of two-dimensional slices respectively, and use them as the analysis results;

[0060] The precision error criteria include:

[0061] The error between the simulated average porosity and the simulated average pore size in the analysis results and the target average two-dimensional porosity and the target average pore size is less than or equal to the threshold value;

[0062] The growth probability includes at least: isotropy and anisotropy.

[0063] The working principle and beneficial effects of the above technical solution are:

[0064] Input the initial reconstruction parameters, including defining the model size (lx, ly, lz), the growth probability G in each direction initially i , the pore seed probability S p and the initial volume porosity P 3D-init . Generate an initial reconstruction model based on the initial reconstruction parameters.

[0065] According to the pore seed probability S p Randomly generate a number of pore seeds in the initial reconstruction model, and each pore seed grows randomly towards the adjacent cells. Assign a random number between 0 and 1 to each adjacent cell. If this value is less than the growth probability G in the corresponding direction i , then this cell is occupied by the pore phase. This process is repeated until the initially set target volume porosity P is reached 3D-init . Subsequently, use a noise filter to smooth the pore edges and improve the structure quality. The pore growth direction is as Figure 2 shown.

[0066] The pore seed probability is a probability parameter that determines the distribution of pore seeds (initial pore positions), defines the frequency or probability of generating pore seeds in the initial reconstruction model, and determines the possibility of each position becoming a pore seed.

[0067] Pore seeds are the starting points or seeds in the model, which are the starting positions of pores and gradually expand to adjacent cells through random growth in subsequent steps.

[0068] Adjacent cells refer to the cells or regions directly adjacent to the current pore seed position. During the growth process, the pores will expand towards these adjacent cells. The definition of adjacent cells can be geometrically adjacent cells or determined according to the mesh division of the model.

[0069] The random number is a number between 0 and 1, which is used to determine whether to allow the pores to expand to a certain adjacent cell. This random number will be compared with the growth probability to judge whether to occupy this adjacent cell.

[0070] The growth probability is the probability that determines whether a pore seed will expand to a certain neighboring cell. There will be a probability value related to growth for each direction or each neighboring cell. If the generated random number is less than this growth probability, the pore will expand to that neighboring cell.

[0071] The target volume porosity is the design goal of the reconstructed model, that is, the proportion of the volume of pores in the final model to the volume of the entire model. The model will continuously perform pore growth until the set volume porosity is reached.

[0072] The three-dimensional reconstruction model is evenly cut into multiple two-dimensional images along a preset direction to match the cross-section sampling method during actual SEM observation, and the simulated average porosity P of multiple two-dimensional slices is calculated respectively 2D-sim and the simulated average pore size A 2D-sim .

[0073] If the errors between the simulated average porosity P of multiple two-dimensional slices 2D-sim and the simulated average pore size A 2D-sim and the target average two-dimensional porosity P 2D-target and the target average pore size A 2D-target (the error between the simulated average porosity P 2D-sim and the target average two-dimensional porosity P 2D-target , the error between the simulated average pore size A 2D-sim and the target average pore size A 2D-target ) are all less than or equal to the threshold value, then the final three-dimensional reconstruction model is output. The threshold value can be taken as 5%.

[0074] Among them, the growth probability can be set to be isotropic (the same probability in each direction) or anisotropic (different probabilities in different directions) according to actual needs.

[0075] In each two-dimensional slice, calculate the proportion of pores in this slice to obtain the porosity of this slice. Then, take the average of the porosities of all two-dimensional slices to obtain the simulated average porosity, which reflects the porosity of the entire model.

[0076] In each two-dimensional slice, calculate the pore sizes of all pores, and then take the average of the pore sizes of all two-dimensional slices to obtain the simulated average pore size, which reflects the size characteristics of the pores. The number of two-dimensional slices can be flexibly determined according to the model size, structural complexity, and accuracy requirements.

[0077] The present invention is applicable to the scenario of porous materials where the matrix material is a homogeneous metal or a single metal material system and the pore structure is non-uniformly randomly distributed.

[0078] By reconstructing and optimizing the initial model, the present invention can accurately simulate the three-dimensional structure of porous materials and achieve high-precision modeling of porous metal materials. Generating pore seeds through a random algorithm and growing them ensures the randomness and authenticity of the model. The introduction of slice analysis verifies the accuracy of the three-dimensional model and ensures that the simulation results are consistent with actual observations. The setting of the accuracy error standard further improves the reliability of the model and guarantees the accuracy of the porosity and pore size. Overall, it effectively solves the error problem in the modeling process of porous metal materials and provides a solid foundation for high-precision simulation and material property prediction.

[0079] Example 2:

[0080] In the embodiment of the present invention, the steps for obtaining the target average two-dimensional porosity and the target average pore size include:

[0081] Determine the target average two-dimensional porosity and the target average pore size according to at least a preset number of scanning electron microscope images of the porous metal material.

[0082] Extract the two-dimensional porosity and pore size from at least three scanning electron microscope (SEM) images to determine the target average two-dimensional porosity P 2D-target and the target average pore size A 2D-target . The number of SEM images used is greater than or equal to three to improve the accuracy of pore structure feature analysis. The target two-dimensional average porosity obtained from the SEM images should be controlled within the range of 0% to 50%. When the porosity exceeds this range, problems such as high connectivity or unclear boundaries may occur between pores, which can easily lead to distortion of the pore topological structure or a decrease in the convergence of the reconstruction algorithm during the model construction process, thus affecting the structural accuracy and simulation adaptability of the final model.

[0083] Calculate the porosity of each image by extracting two-dimensional pore data from at least three SEM images, and then take the average of the porosities of these images to obtain the target average two-dimensional porosity.

[0084] Analyze the pore sizes in at least three SEM images, extract the size data of each pore, calculate the average pore size of each image, and then take the average of the pore sizes of these images to obtain the target average pore size.

[0085] Example 3:

[0086] In the embodiment of the present invention, the noise filter at least includes: Kernel filter, median filter, and Gaussian filter.

[0087] The noise filter includes but is not limited to Kernel filter, median filter, and Gaussian filter.

[0088] Example 4:

[0089] In the embodiment of the present invention, if the analysis result does not meet the accuracy error standard, adjust the initial reconstruction parameters of the initial reconstruction model, and re - execute S1 to S3 after adjustment.

[0090] If the error is greater than 5%, return to step S1 to re - adjust the modeling parameters.

[0091] Example 5:

[0092] The application of any of the methods provided in the embodiments of the present invention in the simulation of the electro - thermal - mechanical properties of materials includes:

[0093] Performing electro - thermal - mechanical property simulation on the output three - dimensional reconstruction model using a finite - element simulation software platform;

[0094] Among them, the finite - element simulation software platform at least includes: Ansys, COMSOL Multiphysics, and Abaqus.

[0095] Import the reconstructed three - dimensional porous model into the finite - element analysis software for subsequent electro - thermal - mechanical multi - physical - field performance simulation analysis.

[0096] The finite - element simulation analysis software includes but is not limited to Ansys, COMSOL Multiphysics, and Abaqus. In practical applications, relevant process parameters (such as drying temperature and time) can be adjusted according to the characteristics of the organic carrier system during the preparation of the used porous material to obtain the best pore structure and performance.

[0097] Example 6:

[0098] In order to obtain the pore - structure characteristics in actual experiments, experiments with sintering times of 10 minutes, 20 minutes, and 30 minutes were carried out on copper sintered materials under the conditions of a sintering temperature of 250 °C and a sintering pressure of 20 MPa. The obtained pore - structure results are as Figure 3 shown. The average porosities of the three groups of samples measured using ImageJ software are 30.75%, 23.02%, and 16.50% respectively, and they are marked as pore grades A, B, and C. Correspondingly, the average pore diameters of the three groups of samples calculated using MATLAB (2023a) are 0.0074 μm 2 , 0.0090 μm 2 and 0.0105 μm 2 . Subsequently, a high - precision modeling method based on a random algorithm proposed in this paper was used to perform three - dimensional reconstruction of the pore structure. Taking the reconstruction process of pore grade A as an example, the scaling ratio between the actual size and the simulation model is set as That is, the size of the simulation model is 10 times that of the actual SEM observation aperture. The specific steps are as follows:

[0099] Step S1: Input a model size of 40×40×40 voxels. According to the actually measured porosity, set the initial porosity to 35% (slightly higher than the 30.75% measured by actual SEM, and the algorithm will search in a decreasing manner subsequently), and set the pore growth probability to be anisotropic.

[0100] Step S2: Based on three scanning electron microscope (SEM) cross-sectional images, determine that the target average two-dimensional porosity is 30.75% and the target average pore size is 0.0074 μm 2 .

[0101] Step S3: Randomly generate a number of pore seeds inside the model according to the set pore seed probability; the seeds grow randomly towards adjacent units according to the set anisotropic growth probability until the initial volume porosity set in Step S1 is reached; subsequently, use the Kernel noise filter to smooth the pore edges to improve the authenticity and quality of the structure.

[0102] Step S4: Uniformly cut the reconstructed three-dimensional model into ten two-dimensional images along the Y direction to simulate the cross-sectional sampling method during actual SEM observation, and calculate the average porosity and average pore size of each two-dimensional slice.

[0103] Step S5: Conduct a comparative analysis of the simulation results and experimental data. Calculate the simulated average two-dimensional porosity and simulated average pore size of the obtained ten two-dimensional slices, and compare them with the target average two-dimensional porosity and target average pore size determined in Step S2. If the error exceeds 5%, return to Step S1 and readjust the initial modeling parameters until the error meets the requirement within 5% and then enter Step S6.

[0104] Step S6: Output the final high-precision three-dimensional pore reconstruction model and save the corresponding three-dimensional pore structure parameters for subsequent simulation verification.

[0105] Step S7: Import the established three-dimensional pore structure model into the finite element simulation platform Comsol to conduct numerical simulation analysis of the thermal and electrical properties of the material.

[0106] The modeling process for pore grades B and C is the same as described above. The electrical and thermal property results of the material obtained by simulation under different pore grades are as Figure 4 , Figure 5 shown in and Table 1. By introducing appropriate correction factors, including a scaling factor (F) and an offset factor (ΔK), the simulation results are corrected. The finally predicted thermal conductivity and electrical conductivity are shown in Table 2.

[0107] The results show that the present invention achieves an accurate prediction of the electrothermal properties of the material within an error range of 5%. Table 1: Average temperature and electric potential of the upper and lower interfaces of the simulation models with different pore gradients, and temperature difference and potential difference corresponding to the upper and lower interfaces.

[0108]

[0109] Table 2: Comparison between the measured results and simulation results of the joint thermal / electrical conductivity under different pore gradients

[0110]

[0111]

[0112] Example 7:

[0113] In the embodiment of the present invention, when outputting the three-dimensional reconstruction model, emotional intervention is used to obtain the user's in-depth modeling requirement information, and the initial reconstruction parameters of the initial reconstruction model are adjusted accordingly based on the in-depth modeling requirements, and after adjustment, S1 to S3 are re-executed;

[0114] Among them, the steps of obtaining the in-depth modeling requirement information by emotional intervention include:

[0115] When the user views the output three-dimensional reconstruction model, if it is recognized that the number of in-depth modeling requirement items hesitantly selected by the user reaches the upper limit of hesitant selection, for multiple in-depth modeling requirement items, multiple hesitant resolution mechanisms and their resolution times are planned;

[0116] When any resolution time is reached in the future, the corresponding hesitant resolution mechanism is executed for the user;

[0117] Whenever a single hesitant resolution mechanism finishes executing for the user, based on at least one hesitant resolution mechanism for which the resolution time has been reached, the un-reached resolution times are optimized;

[0118] Based on the in-depth modeling requirement items finally selected by the user, the in-depth modeling requirement information is determined.

[0119] The working principle and beneficial effects of the above technical solution are as follows:

[0120] In-depth modeling requirements usually stem from high requirements for aspects such as model accuracy, porosity, pore size distribution, etc., as well as the need to accurately reproduce complex microstructures. These requirements drive the continuous optimization and refinement of the three-dimensional reconstruction model, so as to ensure that the model can accurately reflect the performance and characteristics of the material in practical applications. However, when users select in-depth modeling requirements, due to lack of sufficient understanding of modeling methods, insufficient consideration of data processing and computing resources, and unclear expectations of the final application effect, they may hesitate, thus reducing the selection efficiency and resulting in the inability to quickly promote the further optimization and refinement of the three-dimensional reconstruction model.

[0121] Based on this, embodiments of the present invention plan multiple hesitation resolution mechanisms and their resolution times for multiple in-depth modeling requirement items that the user hesitates to select. When the future resolution time arrives, the corresponding hesitation resolution mechanism is executed on the user to assist in resolving the selection hesitation of the in-depth modeling requirement items. Finally, based on the in-depth modeling requirement items ultimately selected by the user, the in-depth modeling requirement information is determined, realizing the emotion-involved acquisition of the in-depth modeling requirement information, improving the efficiency of the user's selection of in-depth modeling requirements, and quickly promoting the further optimization and refinement of the three-dimensional reconstruction model.

[0122] In addition, whenever a single hesitation resolution mechanism finishes executing on the user, the resolution times that have not been reached are optimized based on at least one hesitation resolution mechanism for which the resolution time has arrived, fully ensuring the suitability of executing the hesitation resolution mechanism on the user and further improving the efficiency of their selection of in-depth modeling requirements.

[0123] Embodiment 8:

[0124] In the embodiments of the present invention, the identification steps of multiple in-depth modeling requirement items that the user hesitates to select include:

[0125] Based on the viewing behavior information of the user when viewing the output three-dimensional reconstruction model, the in-depth modeling requirements to be selected are predicted in real time;

[0126] A requirement waiting area is established. Whenever the predicted in-depth modeling requirement to be selected is confirmed by the user, the corresponding in-depth modeling requirement to be selected is placed in the requirement waiting area;

[0127] Whenever a new in-depth modeling requirement to be selected is predicted and the user has not confirmed it, the requirement waiting area is displayed to the user, and the user is requested to remove the target in-depth modeling requirement to be selected from the requirement waiting area;

[0128] If the user refuses, the target in-depth modeling requirement to be selected is used as the in-depth modeling requirement item that the user hesitates to select;

[0129] Among them, the determination steps of the target in-depth modeling requirement to be selected include:

[0130] The in-depth modeling requirement to be selected in the requirement waiting area whose contradiction degree with the newly predicted in-depth modeling requirement to be selected exceeds the threshold is used as the target in-depth modeling requirement to be selected;

[0131] In addition, the in-depth modeling requirement to be selected that has been placed in the requirement waiting area and the in-depth modeling requirement to be selected indicated by the corresponding removal instruction in combination with the newly predicted in-depth modeling requirement to be selected are used as the target in-depth modeling requirement to be selected.

[0132] The working principle and beneficial effects of the above technical solutions are:

[0133] The viewing behavior information when the user views the output 3D reconstruction model at least includes: perspective adjustment, selection area, interaction operation, viewing content, viewing time, as well as annotations and markings, etc. The viewing behavior information reflects how the user wants to further optimize and refine the 3D reconstruction model. Therefore, the candidate depth modeling requirements to be selected can be predicted in real time based on this. When predicting, an artificial intelligence model obtained by machine learning training using a large amount of historical viewing behavior information marked with candidate depth modeling requirements to be selected can be used for prediction. The degree of contradiction refers to the degree of contradiction between the depth modeling objectives of the candidate depth modeling requirements to be selected in the requirements candidate area and the newly predicted candidate depth modeling requirements. The candidate depth modeling requirements already placed in the requirements candidate area and the newly predicted candidate depth modeling requirements jointly correspond to a removal instruction, indicating which candidate depth modeling requirement needs to be removed due to conflicts or other reasons when the newly predicted candidate depth modeling requirement is placed in the requirements candidate area. For example: Suppose there is already a heat conduction modeling requirement in the requirements candidate area, which simulates the heat conduction performance of porous metal at room temperature. And the newly predicted requirement is to simulate the heat conduction behavior of porous metal under high-temperature conditions. These two requirements conflict in modeling methods, simulation ranges, and parameter settings. Therefore, the system will issue a removal instruction. When the new high-temperature heat conduction modeling requirement is placed in the requirements candidate area, it is required to remove the original room-temperature heat conduction requirement.

[0134] Whenever a new candidate depth modeling requirement is predicted and the user has not confirmed the selection, the user is also requested to remove the target candidate depth modeling requirement from the requirements candidate area. If the user refuses, it means that the target candidate depth modeling requirement is a depth modeling requirement item that the user hesitates to select.

[0135] In the embodiment of the present invention, when identifying multiple depth modeling requirement items that the user hesitates to select, a requirements candidate area is established. Whenever the predicted candidate depth modeling requirement is confirmed by the user, the corresponding candidate depth modeling requirement is placed in the requirements candidate area. The target candidate depth modeling requirement is determined based on the degree of contradiction and the removal instruction, and the user is requested to remove the target candidate depth modeling requirement from the requirements candidate area. If the user refuses, the target candidate depth modeling requirement is used as the depth modeling requirement item that the user hesitates to select, which greatly improves the identification accuracy, comprehensiveness, and efficiency of multiple depth modeling requirement items that the user hesitates to select, improves the applicability of the system, and further improves the accuracy of subsequent hesitation resolution for them.

[0136] Embodiment 9:

[0137] In the embodiment of the present invention, the steps for obtaining the upper limit of hesitation selection include:

[0138] Quantify the total duration of the user viewing the output 3D reconstruction model and the upper limit degree of hesitation allowed represented by the user's historical hesitation resolution times to obtain an allowable value;

[0139] Based on the tolerance value, match the corresponding upper limit of hesitant selection from the upper limit table of hesitant selection.

[0140] The longer the total duration that the user views the output 3D reconstruction model, in order to promote the further optimization and refinement of the 3D reconstruction model as soon as possible, the less time is left for hesitation, which means the smaller the upper limit degree of the user's tolerance for hesitation. The larger the number of historical hesitant resolutions, it means the more times the user has made hesitant selections historically, and the less time is left for hesitation, which means the smaller the upper limit degree of the user's tolerance for hesitation. The final tolerance value can be calculated by the following weighted formula:

[0141] C = w1×T + w2×H; where C is the tolerance value, T is the total duration that the user views the output 3D reconstruction model, H is the number of the user's historical hesitant resolutions, and w1 and w2 are preset weight values.

[0142] The upper limit table of hesitant selection has the upper limits of hesitant selection corresponding to different tolerance values. The upper limit of hesitant selection can be the number of multiple in-depth modeling requirement items of hesitant selection, and can also be the type set of multiple in-depth modeling requirement items of hesitant selection.

[0143] When obtaining the upper limit of hesitant selection in the embodiments of the present invention, the upper limit degree of the user's tolerance for hesitation represented by quantifying the total duration that the user views the output 3D reconstruction model and the number of the user's historical hesitant resolutions is obtained to get the tolerance value. Based on this, the corresponding upper limit of hesitant selection is matched from the upper limit table of hesitant selection, which improves the accuracy and efficiency of obtaining the upper limit of hesitant selection, improves the accuracy of using it as the determination of the timing of multiple hesitant resolution mechanisms and their resolution timing planning, and improves the working efficiency of the system.

[0144] Embodiment 10:

[0145] In the embodiments of the present invention, the planning steps of multiple hesitant resolution mechanisms and their resolution timing include:

[0146] Match the hesitant resolution experience based on multiple in-depth modeling requirement items of hesitant selection;

[0147] Determine the hesitant resolution experience adapted to predict the viewing behavior scenario of the user's future viewing of the output 3D reconstruction model and its adapted viewing behavior scenario;

[0148] Based on the adapted hesitant resolution experience, determine the hesitant resolution mechanism, and its resolution timing is when the user enters the viewing behavior scenario adapted to the adapted hesitant resolution experience.

[0149] The hesitation resolution experience is the experience for resolving hesitation choices of multiple deep modeling requirement items pre-set for hesitation choices. Predict the viewing behavior scenario of the user viewing the output 3D reconstruction model in the future. The viewing behavior scenario can be the content scenario in the output 3D reconstruction model, etc. If the hesitation resolution experience is adapted to the viewing behavior scenario, based on the adapted hesitation resolution experience, determine the hesitation resolution mechanism (direct planning by comparison experience), and its resolution timing is when the user enters the viewing behavior scenario adapted by the adapted hesitation resolution experience. It improves the accuracy and efficiency of planning multiple hesitation resolution mechanisms and their resolution timings, and greatly improves the effect of using them to resolve the user's hesitation choices.

[0150] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A high-precision modeling method for porous metal materials based on a random algorithm, characterized in that, Including: S1. Based on a random algorithm, reconstruct and optimize the initial reconstruction model of the porous metal material to obtain a three-dimensional reconstruction model; S2. Analyze the three-dimensional reconstruction model by slicing and determine whether the analysis result meets the accuracy error standard; S3. If it meets the standard, output the three-dimensional reconstruction model.

2. The high-precision modeling method of porous metal materials based on a random algorithm according to claim 1, characterized in that The initial reconstruction model is generated based on initial reconstruction parameters; Among them, the initial reconstruction parameters at least include: defining the model size, the growth probability in each initial direction, the pore seed probability, and the initial volume porosity.

3. The high-precision modeling method of the porous metal material based on the random algorithm according to claim 2, wherein, The steps for reconstructing and optimizing the initial reconstruction model include: S11. Randomly generate a number of pore seeds in the initial reconstruction model according to the pore seed probability; each pore seed grows randomly towards its neighboring cells; S12. Assign a random number between 0 and 1 to each neighboring cell; if the random number assigned to the neighboring cell is less than the growth probability in the corresponding direction, the corresponding neighboring cell is occupied by the pore phase; S13. Repeat S11 to S12 until the initial reconstruction model reaches the initially set target volume porosity; S14. Use a noise filter to smooth the pore edges of the initial reconstruction model that has reached the target volume porosity.

4. The high-precision modeling method of porous metal materials based on a random algorithm according to claim 1, characterized in that, The steps for analyzing the three-dimensional reconstruction model by slicing include: Uniformly cut the three-dimensional reconstruction model along a preset direction into multiple two-dimensional slices to match the cross-section sampling method during actual SEM observation; Calculate the simulated average porosity and simulated average pore diameter of the multiple two-dimensional slices respectively, and use them as the analysis results.

5. The high-precision modeling method of porous metal materials based on a random algorithm according to claim 1, wherein The accuracy error standard includes: The errors between the simulated average porosity and simulated average pore diameter in the analysis results and the target average two-dimensional porosity and target average pore diameter are less than or equal to the threshold value.

6. The high-precision modeling method of porous metal materials based on a random algorithm according to claim 5, characterized in that, The steps for obtaining the target average two-dimensional porosity and target average pore diameter include: Determine the target average two-dimensional porosity and target average pore diameter based on at least a preset number of scanning electron microscope images of the porous metal material.

7. The high-precision modeling method of porous metal materials based on a random algorithm according to claim 3, characterized in that The noise filter at least includes: Kernel filter, median filter, and Gaussian filter.

8. The high-precision modeling method of porous metal materials based on a random algorithm according to claim 2, characterized in that, If the analysis result does not meet the accuracy error standard, adjust the initial reconstruction parameters of the initial reconstruction model, and re-execute S1 to S3 after adjustment.

9. The high-precision modeling method of the porous metal material based on the random algorithm according to claim 2, characterized in that, The growth probability at least includes: isotropic and anisotropic.

10. Use of the method according to any one of claims 1 to 9 in simulating the electro-thermal mechanical properties of materials, characterized in that, Including: Perform electrothermal mechanical performance simulation on the output three-dimensional reconstruction model using a finite element simulation software platform; Among them, the finite element simulation software platform at least includes: Ansys, COMSOL Multiphysics, and Abaqus.

Images