Method and system for representing three-dimensional space distribution of inclusions in alloy steel

Abstract

The invention relates to the technical field of ferrous metallurgy, in particular to a three-dimensional space distribution characterization method and system for inclusions in alloy steel. The method comprises the following steps: performing X-ray tomography and layer-cut imaging on an alloy steel sample, performing layer-by-layer scanning imaging, retaining the integrity of the sample, and obtaining three-dimensional distribution information of inclusions; and filling a blank through data interpolation, constructing three-dimensional volume data, and simplifying voxelization to facilitate processing and analysis. Based on voxel data, reconstructing a three-dimensional model, optimizing shape boundaries of inclusions, accurately measuring size parameters, and extracting spatial position coordinates, so as to reveal an influence mechanism on steel performance; and finally, a spatial point process model is selected to fit actual data, a distribution characteristic model is established, data reliability is verified, prediction capability is provided, inclusion distribution can be predicted under different processes, and support is provided for production optimization and the like.

Description

Technical Field

[0001] This invention relates to the field of iron and steel metallurgy technology, and in particular to a method and system for characterizing the three-dimensional spatial distribution of inclusions in alloy steel. Background Technology

[0002] In the steel production process, the presence of inclusions in steel is a problem that cannot be ignored. Inclusions in steel refer to undesirable non-metallic or other metallic substances contained within the steel. The formation of these inclusions is not accidental; they are usually generated during the steel's smelting, casting, or processing. For example, P91 steel, a high-performance alloy steel also known as T91 steel, belongs to the 9% chromium alloy steel series and is widely used in high-temperature, high-pressure industrial environments due to its excellent performance. Inclusions have a crucial impact on the performance of P91 steel. Their type, size, shape, and distribution all affect the steel's mechanical properties, corrosion resistance, and weldability to varying degrees. Therefore, accurate detection of inclusions in P91 steel is particularly necessary.

[0003] CN113030143A discloses a method for detecting the corrosion activity of inclusions in low-alloy steel. The method comprises the following steps: Step 1, cutting a 10mm × 10mm × 10mm sample from a low-alloy steel plate as the original test sample; Step 2, grinding the original test sample with silicon carbide sandpaper and then mechanically polishing it until the surface roughness is less than 0.8. A polished test sample is obtained; then, the polished test sample is cleaned with acetone, deionized water, and alcohol in sequence, and dried to obtain a pre-prepared test sample; Step 3: The pre-prepared test sample is placed in a fully automated inclusion analyzer to measure the number and density ρ of inclusions in the pre-prepared test sample; the inclusion density is the ratio of the number of inclusions in the statistical region to the area of ​​the statistical region; Step 4: The pre-prepared test sample is placed in a field emission scanning electron microscope, and the accelerating voltage of the field emission scanning electron microscope is set to 10-30kV. ; Randomly select 30-50 inclusions from the pre-made test sample, and use X-ray energy dispersive spectroscopy to identify the composition of the 30-50 inclusions selected from the pre-made test sample, and measure the number of inclusion types n in the pre-made test sample, where n is a natural number ≥1 and ≤30-50; Step 5: Select one inclusion from each of the n types of inclusions, and record the radius R of each type of inclusion in sequence, i = 1, ..., n; Step 6: Calculate the residual stress σ at the interface between the i-th type of inclusion and the matrix in each type of inclusion. Step 7: Compare the residual stress σ at the interface between the i-th type of inclusion and the matrix with the compressive yield strength σ of the original test sample. If the obtained residual stress σ is greater than the compressive yield strength σ of the original test sample, then the i-th type of inclusion is a corrosion-active inclusion. If the obtained residual stress σ is less than or equal to the compressive yield strength σ of the original test sample, then the i-th type of inclusion is a non-corrosive inclusion. Step 8: Repeat steps 6 and 7 to obtain the corrosion activity of each type of inclusion in the selected inclusions.

[0004] In existing technologies, imaging techniques are insufficient in resolution and clarity, making it difficult to accurately display the shape, size, and location information of inclusions. This leads to inaccurate parameter measurements due to the inability to accurately identify inclusion boundaries. Furthermore, traditional methods are often limited to the analysis and characterization of existing data, lacking predictive capabilities and making them difficult to apply directly to actual production. Summary of the Invention

[0005] To address the aforementioned problems in the existing technology, the purpose of this invention is to provide a method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel.

[0006] To solve the above problems, the technical solution adopted by the present invention is as follows: A method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel, the method comprising the following steps: Step S1: Randomly select samples from the alloy steel billets or finished alloy steel products to be tested. Cut the selected samples into 10mm × 10mm × 5mm pieces. Remove scratches and deformation layers from the surface of the cut samples using mechanical and chemical polishing methods, controlling the surface flatness of the cut samples to ±0.5 mm. ; Step S2 involves using X-ray tomography to scan the sample that met the flatness standard in step S1 from multiple angles with a step size of 0.5°-5°, and acquiring the obtained tomographic projection data to generate three-dimensional tomographic image data with a first resolution of 10-50. ; The scanned sample was sliced ​​layer by layer using a focused ion beam. After each slice, a second-resolution two-dimensional image was acquired using SEM, resulting in sliced ​​two-dimensional image data. The slice thickness was controlled to be ≤50 mm. The second resolution is 100. -1 ; Step S3: The three-dimensional tomographic image data and the sliced ​​two-dimensional image data are segmented. The segmented three-dimensional tomographic image data is converted into a three-dimensional model by voxelization three-dimensional reconstruction. The inclusions and steel matrix are distinguished in the three-dimensional model based on grayscale threshold and edge detection. The size, shape parameters and position coordinates of the inclusions are calculated. Step S4: Collect data on the number density, size distribution, and shape distribution of the inclusions, calculate the average spacing and aggregation degree of the inclusions in the alloy steel, display the spatial distribution of the inclusions in the form of a three-dimensional graphic, and establish and output a mathematical model of the spatial distribution of the inclusions to describe the spatial distribution characteristics of the inclusions.

[0007] Further, in step S3, the Kriging interpolation algorithm is used to fill in the blank areas between the three-dimensional tomographic image data and the sliced ​​two-dimensional image data, and the interpolated three-dimensional data is discretized into 3D voxels, each 3D voxel having a size ≤1. 3 .

[0008] Furthermore, in step S3, the inclusions are separated from the steel substrate by an adaptive threshold algorithm, and the boundary contour of the inclusions is optimized by edge detection.

[0009] Furthermore, in step S3, noise is eliminated using opening or closing operations, the shape of the inclusions is corrected, and the size and shape parameters are calculated. The shape parameters include at least volume, surface area, and major axis / minor axis ratio.

[0010] Furthermore, a coordinate system based on the three-dimensional model records the spatial position of each inclusion and calculates the relative distance between each inclusion and the steel substrate, wherein the spatial position is a three-dimensional coordinate system.

[0011] Furthermore, in step S4, a Poisson point process or Markov random field model is used to fit the spatial distribution characteristics of the inclusions, quantify the average spacing and the degree of aggregation, wherein the degree of aggregation includes at least the Gini coefficient or the spatial autocorrelation coefficient.

[0012] Further, in step S3, the three-dimensional tomographic image data and the sliced ​​two-dimensional image data are segmented. First, noise in the image is removed, and then the contrast between the inclusions and the steel substrate in the image is enhanced by histogram equalization. The inclusions are then separated from the steel substrate according to the grayscale threshold.

[0013] Further, the feature is that, in step S4, the digital model selects a spatial point process model, fits the spatial point process model with the actual data, and statistically tests the goodness of fit of the mathematical model. After the mathematical model is verified, the inclusion distribution characteristic model is obtained. The process parameters are input through the mathematical model, and the predicted results of the inclusion distribution are output. The process parameters include at least the melting temperature and the type of deoxidizer.

[0014] The system described above for characterizing the three-dimensional spatial distribution of inclusions in alloy steel includes, The sample preparation unit is used to randomly select samples from the alloy steel billet or finished alloy steel product to be tested. The selected samples are cut into 10mm × 10mm × 5mm dimensions, and surface scratches and deformation layers are removed through mechanical and chemical polishing to achieve a sample surface flatness of ±0.5. ; The scanning unit is used for X-ray tomography, wherein the sample is scanned at multiple angles with a step size of 0.5°-5° to acquire tomographic projection data and generate three-dimensional tomographic image data with a first resolution of 10-50. Layer-by-layer slicing is performed using a focused ion beam, and a second-resolution two-dimensional image is acquired using SEM after each slice to form layer-by-layer two-dimensional image data, wherein the slice thickness is ≤50 mm. The second resolution is 100nm-1 ; The data processing unit is used to segment the three-dimensional tomographic image data and the sliced ​​two-dimensional image data, and then convert the three-dimensional tomographic image data into a three-dimensional model through voxelization three-dimensional reconstruction; to distinguish the inclusions from the steel matrix in the three-dimensional model based on grayscale threshold and edge detection; and to calculate the size, shape parameters and position coordinates of the inclusions. The spatial distribution characterization unit is used to statistically analyze the number density, size distribution, and shape distribution data of inclusions, calculate the average spacing and aggregation degree of inclusions in alloy steel, display the spatial distribution of inclusions in a three-dimensional graphical form, and establish and output a mathematical model of the spatial distribution of inclusions to describe the spatial distribution characteristics of inclusions.

[0015] A machine-readable storage medium storing instructions for causing a machine to execute the method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel as described above.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention discloses a method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel. Through steps S1-S4, firstly, the alloy steel sample is scanned and imaged layer by layer using X-ray tomography and FIB-SEM layer-by-layer imaging techniques. This preserves the integrity of the sample while acquiring the three-dimensional spatial distribution information of inclusions within it. Data interpolation fills the blank areas between images, constructing complete three-dimensional volumetric data. This preserves the accuracy of the original data and simplifies the three-dimensional data into a series of small cubic spaces through voxelization, facilitating subsequent processing and analysis. The three-dimensional model reconstruction based on voxel data can more accurately restore the three-dimensional morphology and spatial distribution of inclusions in the steel. After morphological operations optimize the shape and boundaries of inclusions, their volume, surface area, major axis, minor axis, and other dimensional parameters can be measured more accurately. Furthermore, based on the coordinate system of the three-dimensional model, the spatial coordinates of each inclusion can be extracted. Comprehensive analysis of this information can further reveal the mechanism by which inclusions affect the properties of the steel. By selecting a spatial point process model and fitting it with actual data, a model of inclusion distribution characteristics is established. This not only verifies the reliability of the data but also possesses predictive capabilities, enabling the prediction of the spatial distribution characteristics of inclusions under different process conditions. This invention also discloses a system for the above method. This method and system integrate multiple steps, including data statistical analysis, 3D graphical display, and mathematical model establishment. This improves research efficiency and ensures the accuracy and reliability of research results through rigorous statistical analysis and model validation. It can provide strong support for process optimization, quality control, and product performance improvement in actual production. Production parameters can be adjusted based on model prediction results to reduce inclusion formation or improve its distribution. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the steps of this method; Figure 2 This is a schematic diagram of the system structure of this method. Detailed Implementation

[0018] The present invention will be further described below with reference to specific embodiments.

[0019] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0021] It should be noted that the terms "first," "second," "third," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of the invention described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0022] To address the technical problems existing in current technologies, traditional methods for identifying and measuring inclusions, limited by inherent technical limitations, consistently face the challenge of accurately defining inclusion boundaries. Inclusions and the steel matrix exhibit subtle differences in physical properties, and their morphologies are often irregular—for example, some are angular granules, others are diffusely distributed flocculents, and still others form blurred transition zones with the matrix. Traditional detection techniques struggle to capture these subtle features. This ambiguity in boundary identification directly reduces the accuracy of measurements for key parameters. Inclusions with diameters of only a few micrometers may be over- or under-counted by micrometers due to boundary misjudgment; two adjacent inclusions that should exist independently may be misjudged as a single entity due to boundary confusion; and key parameters such as surface area and volume of inclusions will show significant errors due to boundary deviations. This accumulated measurement bias distorts the judgment of the true state of inclusions, making it impossible to accurately determine their actual size, distribution density, and morphological characteristics, thus rendering the analytical conclusions based on this data unreliable. When these biased conclusions are used as a reference in steel production, they fail to provide precise guidance for process adjustments and quality control, making it difficult to address inclusion problems specifically during production. Ultimately, this severely restricts the stability and improvement potential of steel quality. This invention provides a method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel, such as... Figure 1 The diagram illustrates a method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel according to an embodiment of the present invention. The method includes... Step S1: Randomly select samples from the alloy steel billet or finished alloy steel product to be tested. Cut the selected samples into 10mm × 10mm × 5mm dimensions. Remove surface scratches and deformation layers through mechanical polishing and chemical polishing to achieve a sample surface flatness of ±0.5. ; Step S2 involves X-ray tomography, where the sample is scanned from multiple angles with a step size of 0.5°-5° to acquire tomographic projection data and generate three-dimensional tomographic image data with a first resolution of 10-50. Layer-by-layer slicing is performed using a focused ion beam, and a second-resolution two-dimensional image is acquired using SEM after each slice to form layer-by-layer two-dimensional image data, wherein the slice thickness is ≤50 mm. The second resolution is 100nm-1 ; Step S3: Segment the three-dimensional tomographic image data and the sliced ​​two-dimensional image data, and then convert the three-dimensional tomographic image data into a three-dimensional model through voxelization three-dimensional reconstruction; distinguish the inclusions and the steel matrix in the three-dimensional model based on grayscale threshold and edge detection; and calculate the size, shape parameters and position coordinates of the inclusions. Step S4: Collect data on the number density, size distribution, and shape distribution of inclusions, calculate the average spacing and aggregation degree of inclusions in the alloy steel, display the spatial distribution of inclusions in a three-dimensional graphic form, and establish and output a mathematical model of the spatial distribution of inclusions to describe the spatial distribution characteristics of inclusions.

[0023] This invention discloses a method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel. Through steps S1-S4, firstly, the alloy steel sample is scanned and imaged layer by layer using X-ray tomography and FIB-SEM layer-by-layer imaging techniques. This preserves the integrity of the sample while acquiring the three-dimensional spatial distribution information of inclusions within it. Data interpolation fills the blank areas between images, constructing complete three-dimensional volumetric data. This preserves the accuracy of the original data and simplifies the three-dimensional data into a series of small cubic spaces through voxelization, facilitating subsequent processing and analysis. The three-dimensional model reconstruction based on voxel data can more accurately restore the three-dimensional morphology and spatial distribution of inclusions in the steel. After morphological operations optimize the shape and boundaries of inclusions, their volume, surface area, major axis, minor axis, and other dimensional parameters can be measured more accurately. Furthermore, based on the coordinate system of the three-dimensional model, the spatial coordinates of each inclusion can be extracted. Comprehensive analysis of this information can further reveal the mechanism by which inclusions affect the properties of the steel. By selecting a spatial point process model and fitting it with actual data, a model of inclusion distribution characteristics was established. This not only verified the reliability of the data but also demonstrated predictive capabilities, enabling the prediction of the spatial distribution characteristics of inclusions under different process conditions. The entire spatial distribution characterization method integrates multiple steps, including data statistical analysis, 3D graphical visualization, and mathematical model establishment, forming a systematic and scientific research approach. This improves research efficiency while ensuring the accuracy and reliability of the results through rigorous statistical analysis and model validation. It can provide strong support for process optimization, quality control, and product performance improvement in actual production. Production parameters can be adjusted based on model prediction results to reduce inclusion formation or improve its distribution.

[0024] In a more preferred embodiment of the present invention, in step S3, the blank area between the three-dimensional tomographic image data and the sliced ​​two-dimensional image data is filled using the Kriging interpolation algorithm or a deep learning model, and the interpolated three-dimensional data is discretized into voxels, each voxel having a size no greater than 1. 3 Voxels are 3D pixels.

[0025] In a more preferred embodiment of the present invention, in step S3, the inclusions are separated from the steel substrate by an adaptive threshold algorithm, and the boundary contour of the inclusions is optimized by edge detection.

[0026] In a more preferred embodiment of the present invention, in step S3, noise is eliminated using an opening or closing operation method, the shape of the inclusions is corrected, and the size and shape parameters are calculated, wherein the shape parameters include at least volume, surface area, and major axis / minor axis ratio.

[0027] In a more preferred embodiment of the present invention, a coordinate system based on a three-dimensional model records the spatial position of each inclusion and calculates the relative distance between each inclusion and the steel substrate, wherein the spatial position is a three-dimensional coordinate system.

[0028] In a more preferred embodiment of the present invention, in step S4, a Poisson point process or a Markov random field model is used to fit the spatial distribution characteristics of the inclusions, and to quantify their average spacing and degree of aggregation, wherein the degree of aggregation includes at least the Gini coefficient or the spatial autocorrelation index.

[0029] In a more preferred embodiment of the present invention, in step S3, the three-dimensional tomographic image data and the sliced ​​two-dimensional image data are segmented, noise in the image is removed first, the contrast between the inclusions and the steel substrate in the image is enhanced by histogram equalization, and the inclusions are separated from the steel substrate according to the grayscale threshold.

[0030] In a more preferred embodiment of the present invention, in step S4, the mathematical model selects a spatial point process model, fits the spatial point process model with the actual data, and performs a statistical test on the goodness of fit of the mathematical model. After the mathematical model is verified, an inclusion distribution characteristic model is obtained. The process parameters are input through the mathematical model, and the predicted results of inclusion distribution are output. The process parameters include at least melting temperature and deoxidizer type.

[0031] This invention also discloses a system for characterizing the three-dimensional spatial distribution of inclusions in alloy steel as described above, such as... Figure 2 As shown, the system includes, The sample preparation unit is used to randomly select samples from the alloy steel billet or finished alloy steel product to be tested. The selected samples are cut into 10mm × 10mm × 5mm dimensions, and surface scratches and deformation layers are removed through mechanical and chemical polishing to achieve a sample surface flatness of ±0.5. ; The scanning unit is used for X-ray tomography, wherein the sample is scanned at multiple angles with a step size of 0.5°-5° to acquire tomographic projection data and generate three-dimensional tomographic image data with a first resolution of 10-50. Layer-by-layer slicing is performed using a focused ion beam. After each slice, a second-resolution two-dimensional image is acquired using SEM to form layer-by-layer 2D image data. The slice thickness is ≤50 nm, and the second resolution is 100 nm⁻¹. ; The data processing unit is used to segment the three-dimensional tomographic image data and the sliced ​​two-dimensional image data, and then convert the three-dimensional tomographic image data into a three-dimensional model through voxelization three-dimensional reconstruction; to distinguish the inclusions from the steel matrix in the three-dimensional model based on grayscale threshold and edge detection; and to calculate the size, shape parameters and position coordinates of the inclusions. The spatial distribution characterization unit is used to statistically analyze the number density, size distribution, and shape distribution data of inclusions, calculate the average spacing and aggregation degree of inclusions in alloy steel, display the spatial distribution of inclusions in a three-dimensional graphical form, and establish and output a mathematical model of the spatial distribution of inclusions to describe the spatial distribution characteristics of inclusions.

[0032] The present invention provides a machine-readable storage medium storing instructions that cause a machine to execute the construction method described above.

[0033] It should be noted that, for the sake of brevity, the various method embodiments described above are presented as a series of actions. However, those skilled in the art should understand that the present invention is not bound by the described order of actions, and based on the technical concept of the present invention, some steps may be executed in other orders or simultaneously.

[0034] Those skilled in the art should also understand that the embodiments described in the specification are preferred solutions, and the actions and modules involved are not essential components of the present invention. In the above embodiments, the description of each embodiment has its own emphasis; if there are any contents not fully described in a certain embodiment, refer to the relevant descriptions of other embodiments. Furthermore, in the various embodiments of the present invention, each functional unit can be integrated into a processing unit, maintain a separate physical existence, or integrate two or more units into one unit. The integrated unit can be implemented in hardware or as a software functional unit. The above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can be modified and varied in many ways. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel, characterized in that, The method includes the following steps: Step S1: Randomly select samples from the alloy steel billets or finished alloy steel products to be tested. Cut the selected samples into 10mm × 10mm × 5mm pieces. Remove scratches and deformation layers from the surface of the cut samples using mechanical and chemical polishing methods, controlling the surface flatness of the cut samples to ±0.5 mm. ; Step S2 involves using X-ray tomography to scan the sample that met the flatness standard in step S1 from multiple angles with a step size of 0.5°-5°, and acquiring the obtained tomographic projection data to generate three-dimensional tomographic image data with a first resolution of 10-50. ; The scanned sample was sliced ​​layer by layer using a focused ion beam. After each slice, a second-resolution two-dimensional image was acquired using SEM to obtain sliced ​​two-dimensional image data. The slice thickness was controlled to be ≤50 mm. The second resolution is 100. -1 ; Step S3: The three-dimensional tomographic image data and the sliced ​​two-dimensional image data are segmented. The segmented three-dimensional tomographic image data is converted into a three-dimensional model by voxelization three-dimensional reconstruction. The inclusions and steel matrix are distinguished in the three-dimensional model based on grayscale threshold and edge detection. The size, shape parameters and position coordinates of the inclusions are calculated. Step S4: Collect data on the number density, size distribution, and shape distribution of the inclusions, calculate the average spacing and aggregation degree of the inclusions in the alloy steel, display the spatial distribution of the inclusions in the form of a three-dimensional graphic, and establish and output a mathematical model of the spatial distribution of the inclusions to describe the spatial distribution characteristics of the inclusions.

2. The method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel according to claim 1, characterized in that, In step S3, the Kriging interpolation algorithm is used to fill in the blank areas between the three-dimensional tomographic image data and the sliced ​​two-dimensional image data. The interpolated three-dimensional data is discretized into 3D voxels, with each 3D voxel having a size ≤1. 3 .

3. The method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel according to claim 2, characterized in that, In step S3, the inclusions are separated from the steel substrate by an adaptive threshold algorithm, and the boundary contour of the inclusions is optimized by edge detection.

4. The method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel according to claim 1, characterized in that, In step S3, noise is eliminated using opening or closing operations, the shape of inclusions is corrected, and the size and shape parameters are calculated. The shape parameters include at least volume, surface area, and major axis / minor axis ratio.

5. The method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel according to claim 1, characterized in that, A coordinate system based on a three-dimensional model records the spatial position of each inclusion and calculates the relative distance between each inclusion and the steel substrate. The spatial position is a three-dimensional coordinate system.

6. The method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel according to claim 1, characterized in that, In step S4, a Poisson point process or Markov random field model is used to fit the spatial distribution characteristics of the inclusions, quantify the average spacing and the degree of aggregation, wherein the degree of aggregation includes at least the Gini coefficient or the spatial autocorrelation coefficient.

7. The method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel according to claim 1, characterized in that, In step S3, the three-dimensional tomographic image data and the sliced ​​two-dimensional image data are segmented. First, noise in the image is removed. Then, histogram equalization is used to enhance the contrast between the inclusions and the steel substrate in the image. The inclusions are separated from the steel substrate according to the grayscale threshold.

8. The method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel according to any one of claims 1-7, characterized in that, In step S4, the digital model selects the spatial point process model, fits the spatial point process model with the actual data, and performs a statistical test on the goodness of fit of the mathematical model. After the mathematical model is verified, the inclusion distribution characteristic model is obtained. The process parameters are input into a mathematical model, and the predicted results of inclusion distribution are output; the process parameters include at least melting temperature and type of deoxidizer.

9. The system for characterizing the three-dimensional spatial distribution of inclusions in alloy steel as described in any one of claims 1-8, characterized in that, The system includes, The sample preparation unit is used to randomly select samples from the alloy steel billet or finished alloy steel product to be tested. The selected samples are cut into 10mm × 10mm × 5mm dimensions, and surface scratches and deformation layers are removed through mechanical and chemical polishing to achieve a sample surface flatness of ±0.

5. ; The scanning unit is used for X-ray tomography, wherein the sample is scanned at multiple angles with a step size of 0.5°-5° to acquire tomographic projection data and generate three-dimensional tomographic image data with a first resolution of 10-50. Layer-by-layer slicing is performed using a focused ion beam, and a second-resolution two-dimensional image is acquired using SEM after each slice to form layer-by-layer two-dimensional image data, wherein the slice thickness is ≤50 mm. The second resolution is 100nm-1 ; The data processing unit is used to segment the three-dimensional tomographic image data and the sliced ​​two-dimensional image data, and then convert the three-dimensional tomographic image data into a three-dimensional model through voxelization three-dimensional reconstruction; to distinguish the inclusions from the steel matrix in the three-dimensional model based on grayscale threshold and edge detection; and to calculate the size, shape parameters and position coordinates of the inclusions. The spatial distribution characterization unit is used to statistically analyze the number density, size distribution, and shape distribution data of inclusions, calculate the average spacing and aggregation degree of inclusions in alloy steel, display the spatial distribution of inclusions in a three-dimensional graphical form, and establish and output a mathematical model of the spatial distribution of inclusions to describe the spatial distribution characteristics of inclusions.

10. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores instructions for causing the machine to execute the method for characterizing the three-dimensional spatial distribution of inclusions in alloy steel as described in any one of claims 1-8.

Images