Method for fully automatically detecting inclusions in rare earth-containing stainless steel

Through the fully automatic detection method, scanning electron microscope and OTS software combined with machine learning algorithms, the problem of inclusion modification in the existing technology cannot be detected, and high-precision inclusion detection and detailed description are achieved.

CN120490541APending Publication Date: 2025-08-15INNER MONGOLIA UNIV OF SCI & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510910198.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-02
Publication Date
2025-08-15

AI Technical Summary

Technical Problem

The prior art cannot effectively detect the state of inclusions after modification in rare earth stainless steel.

Method used

The fully automatic detection method is adopted, and the metal samples are pre-grinded and polished, and the scanning electron microscope and OTS inclusion automatic analysis software are used, combined with machine learning algorithms, to accurately identify the type, size, distribution and chemical composition of inclusions, and set reasonable scanning analysis modes and magnification to achieve detailed detection of inclusions in rare earth stainless steel.

Benefits of technology

It improves the accuracy and accuracy of detection, avoids detection omissions or misjudgments, can effectively detect the state of inclusions after the rare earth element is modified, and describes the composite inclusions in detail.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120490541A_ABST
    Figure CN120490541A_ABST
Patent Text Reader

Abstract

The invention discloses a method for fully automatically detecting inclusions in rare earth-containing stainless steel, and relates to the technical field of analysis and detection of steel and iron materials. The method comprises the following steps: obtaining a rare earth-containing stainless steel metal sample; counting the size range of different inclusions in the rare earth-containing stainless steel metal sample, determining the detection mode, the detection area and the measurement map precision of the inclusions, and setting the lower limit of the minimum detection size, the maximum detection size being not limited; according to the equivalent circle diameter of the inclusions, setting an analysis threshold value, and determining a scanning analysis mode of the inclusions; determining the amplification factor range of the scanning electron microscope according to the detection area of the inclusions; and carrying out detection statistics on the inclusions in the sample through OTS inclusion automatic analysis software to obtain the types, sizes, distribution and chemical components of the inclusions. The method provided by the invention can be used for accurately detecting the state of the inclusion in the rare earth-containing stainless steel after modification.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of steel material analysis and detection, and in particular to a method for fully automatic detection of inclusions in rare earth-containing stainless steel. Background Art

[0002] With the advancement of steelmaking technology, higher requirements are being placed on inclusions in steel. Furthermore, inclusions in steel have a significant impact on the mechanical properties of experimental steels, particularly in the field of stainless steel. Inclusions in stainless steel can damage the passive layer on the surface, forming pitting corrosion pits near the inclusions, leading to large-scale failure of the stainless steel. Rare earth elements have a significant modifying effect on inclusions in steel, making the areas near the inclusions less susceptible to dissolution in corrosive media, thereby improving the corrosion resistance of the stainless steel. Therefore, rare earth-modified inclusions play a crucial role in improving the performance of stainless steel.

[0003] The prior art provides a method for detecting and analyzing small-sized non-metallic inclusions in steel using a scanning electron microscope. By automatically detecting inclusions using a scanning electron microscope and an energy dispersive spectrometer, small-sized non-metallic inclusions can be counted. However, the prior art cannot detect the modified state of inclusions in rare earth-containing steel.

[0004] Therefore, there is an urgent need for a detection method to detect the modified state of inclusions in rare earth-containing stainless steel. Summary of the Invention

[0005] Based on this, it is necessary to provide a fully automatic method for detecting inclusions in rare earth-containing stainless steel to address the above technical problems, so as to detect the state of inclusions in stainless steel after modification by rare earth elements and describe the composite inclusions.

[0006] The present invention adopts the following technical solutions: The present invention provides a method for fully automatically detecting inclusions in rare earth-containing stainless steel, comprising: The oxide scale on the surface of the metal sample is removed by a pre-grinder, and the metal sample is mechanically ground and polished to obtain a rare earth-containing stainless steel metal sample; the metal sample is rare earth-containing stainless steel; According to the size range of different inclusions in rare earth-containing stainless steel metal samples, the inclusion detection method, detection area and measurement accuracy are determined, and the minimum detection size limit is determined; according to the equivalent circle diameter of the inclusion, the inclusion scanning analysis mode is determined; according to the detection area of the inclusion, the magnification range of the scanning electron microscope is determined; The OTS automatic inclusion analysis software is used to detect and count inclusions in rare earth-containing stainless steel metal samples based on the established detection method, detection area, measurement image accuracy, minimum detection size limit, scanning analysis mode, and scanning electron microscope magnification range. The type, size, distribution, and chemical composition of the inclusions are obtained and imported into the database model. The database model is then optimized using a machine learning algorithm to obtain a detailed report on the type, size distribution, and chemical composition of the inclusions.

[0007] Preferably, the metal sample is subjected to mechanical grinding and polishing, specifically comprising: The metal sample was ground using sandpaper of different grit sizes; the scratch direction of each grit size sandpaper was consistent; The ground metal sample surface is polished using diamond polishing paste; the polishing time ranges from 1 minute to 5 minutes.

[0008] Preferably, the inclusion detection method, detection area, and measurement accuracy are determined based on the size range of different inclusions in the rare earth-containing stainless steel metal sample, specifically including: When the size range of different inclusions in the rare earth-containing stainless steel metal sample is within the range of 0 μm to 1 μm, determining that the detection method for inclusions is small-area multi-region point detection, and determining that the detection area and measurement accuracy of a single region are the first preset detection area and the first preset measurement accuracy; When the size range of different inclusions in the rare earth-containing stainless steel metal sample is within the range of 1 μm to 10 μm, the inclusion detection method is determined to be small-area multi-region point detection, and the detection area and measurement accuracy of a single area are determined to be the second preset detection area and the second preset measurement accuracy; When the sizes of the different inclusions in the sample to be tested are statistically determined to be within the range of 10 μm to 100 μm, the inclusion detection method is determined to be a large-area single-region detection method, and the detection area and measurement accuracy of the single region are determined to be the third preset detection area and the third preset measurement accuracy; The first preset detection area is smaller than the second preset detection area, and the second preset detection area is smaller than the third preset detection area; the first preset measurement accuracy is greater than the second preset measurement accuracy, and the second preset measurement accuracy is greater than the third preset measurement accuracy.

[0009] Preferably, the minimum detection size lower limit is 5 times the pixel size.

[0010] Preferably, the scanning analysis mode of the inclusions is determined according to the equivalent circle diameter of the inclusions, specifically including: When the equivalent circle diameter is less than or equal to a preset analysis threshold, determining that the scanning analysis mode of different inclusions in the rare earth-containing stainless steel is a point analysis mode; When the equivalent circle diameter is greater than a preset analysis threshold, it is determined that the scanning analysis mode of different inclusions in the rare earth-containing stainless steel is a surface analysis mode.

[0011] Preferably, during the detection of inclusions in a rare earth-containing stainless steel metal sample, the scanning time for the inclusions is greater than 200 milliseconds and the expected counting rate is greater than 10,000.

[0012] Preferably, the magnification range of the scanning electron microscope is determined according to the detection area of the inclusions, specifically including: When the inspection area of inclusions is less than or equal to 2 square millimeters, the magnification range of the scanning electron microscope is determined to be 600 times to 1000 times; When the inspection area of the inclusion is larger than 2 square millimeters and smaller than 20 square centimeters, the magnification range of the scanning electron microscope is determined to be 200 times to 600 times.

[0013] At least one of the above technical solutions adopted by the present invention can achieve the following beneficial effects: The present invention statistically analyzes the size ranges of different inclusions in the mirror metal sample, and specifically determines the detection method and detection area of rare earth inclusions according to the actual size range of the inclusions, thereby avoiding unnecessary detection ranges and improving detection efficiency. The invention accurately identifies the size of inclusions, flexibly adjusts scanning parameters, and sets reasonable analysis thresholds according to the equivalent circle diameters of different inclusions in rare earth-containing stainless steel. Based on this, a scanning analysis mode for inclusions is determined, and dual qualitative confirmation is performed using a scanning electron microscope and OTS software. Automatic positioning detection is enabled to accurately determine the overall distribution state of rare earth inclusions in the stainless steel material. The present invention not only improves the precision and accuracy of detection, but also effectively avoids omissions or misjudgments caused by inappropriate detection conditions. It can effectively detect the state of inclusions in steel after modification by rare earth elements, and provides a detailed description of composite inclusions through a database model. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0015] Figure 1 A schematic flow chart of a method for fully automatic detection of inclusions in rare earth-containing stainless steel provided by the present invention; Figure 2 This is a graph showing the proportion of various types of inclusions automatically scanned by the OTS in Example 1 provided by the present invention; Figure 3 This is a graph showing the proportion of various types of inclusions automatically scanned by the OTS in Example 2 provided by the present invention; Figure 4This is a graph showing the proportion of various types of inclusions automatically scanned by the OTS in Example 3 provided by the present invention; Figure 5 This is a diagram showing the overall number and size distribution of all embodiments provided herein. DETAILED DESCRIPTION

[0016] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments of the present invention and corresponding drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in the specification, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0017] The following describes in detail the technical solutions provided by various embodiments of the present application in conjunction with the accompanying drawings.

[0018] Figure 1 The following is a flow chart of a method XX in the present invention, which specifically includes the following steps: S101: removing oxide scale on the surface of a metal sample by a pre-grinder, and mechanically grinding and polishing the metal sample to obtain a rare earth-containing stainless steel metal sample; the metal sample is rare earth-containing stainless steel.

[0019] Specifically, the metal sample was ground using sandpaper with a grit size of 120# to 800#, and the scratch direction of each grit size sandpaper was consistent, so that there was no scratch on the surface of the metal sample; the metal sample surface was polished using 2.5μm and 1μm diamond polishing paste, and the polishing time was 1 minute to 5 minutes.

[0020] S102: Based on the size range of different inclusions in the rare earth-containing stainless steel metal sample, determine the inclusion detection method, detection area and measurement diagram accuracy, and determine the minimum detection size limit; based on the equivalent circle diameter of the inclusion, determine the inclusion scanning analysis mode; based on the detection area of the inclusion, determine the magnification range of the scanning electron microscope.

[0021] Optionally, according to the size range of different inclusions in the rare earth-containing stainless steel metal sample, the inclusion detection method, detection area and measurement accuracy are determined, specifically including: When the size range of different inclusions in the rare earth-containing stainless steel metal sample is within the range of 0um to 1um, the inclusion detection method is determined to be small-area multi-region point detection, and the detection area and measurement accuracy of a single area are determined to be the first preset detection area and the first preset measurement accuracy; when the size range of different inclusions in the rare earth-containing stainless steel metal sample is within the range of 1um to 10um, the inclusion detection method is determined to be small-area multi-region point detection, and the detection area and measurement accuracy of a single area are determined to be the second preset detection area and the second preset measurement accuracy; when the size range of different inclusions in the statistically tested sample is within the range of 10um to 100um, the inclusion detection method is determined to be large-area single-region detection method, and the detection area and measurement accuracy of a single area are determined to be the third preset detection area and the third preset measurement accuracy; the first preset detection area is smaller than the second preset detection area, and the second preset detection area is smaller than the third preset detection area; the first preset measurement accuracy is greater than the second preset measurement accuracy, and the second preset measurement accuracy is greater than the third preset measurement accuracy.

[0022] Specifically, when the size range of different inclusions in the statistical metal sample is 10 microns to 500 nanometers, the detection method for inclusions is determined to be small-area multi-region point detection, the detection area of a single area is determined to be 2 square millimeters, and the measurement accuracy is 1536×1024; when the size range of different inclusions in the statistical metal sample is 100 microns to 10 microns, the detection method for inclusions is determined to be large-area single-region detection method, the detection area is determined to be greater than 16 square centimeters, and the measurement accuracy is 1024×768.

[0023] The minimum detection size is 5 times the pixel size.

[0024] Optionally, determining the scanning analysis mode of the inclusions according to the equivalent circle diameter of the inclusions specifically includes: when the equivalent circle diameter is less than or equal to a preset analysis threshold, determining that the scanning analysis mode of different inclusions in the rare earth-containing stainless steel is a point analysis mode; when the equivalent circle diameter is greater than the preset analysis threshold, determining that the scanning analysis mode of different inclusions in the rare earth-containing stainless steel is a surface analysis mode.

[0025] Specifically, the equivalent circle diameter of the inclusions in the rare earth-containing stainless steel metal sample is set as the analysis threshold, specifically 3 microns; when the equivalent circle diameter is less than or equal to 3 microns, the point analysis mode is used to scan and analyze different inclusions in the rare earth-containing stainless steel; when the equivalent circle diameter is greater than 3 microns, the surface analysis mode is used to scan and analyze different inclusions in the rare earth-containing stainless steel.

[0026] Among them, different inclusions in rare earth-containing stainless steel are scanned and analyzed, with an acquisition time greater than 200 milliseconds and an expected count rate greater than 10,000.

[0027] Optionally, the magnification range of the scanning electron microscope is determined according to the detection area of the inclusion, specifically including: when the detection area of the inclusion is less than or equal to 2 square millimeters, determining the magnification range of the scanning electron microscope to be 600 times to 1000 times; when the detection area of the inclusion is greater than 2 square millimeters and less than 20 square centimeters, determining the magnification range of the scanning electron microscope to be 200 times to 600 times.

[0028] S103: Using the OTS automatic inclusion analysis software, based on the established detection method, detection area, measurement image accuracy, minimum detection size limit, scanning analysis mode, and scanning electron microscope magnification range, the inclusions in the rare earth-containing stainless steel metal sample are detected and counted to obtain the type, size, distribution, and chemical composition of the inclusions.

[0029] The metal sample to be tested is tested by the above method and steps, including: Example 1 The weight percentage of the chemical composition of the experimental steel of the metal sample to be tested is shown in Table 1: Table 1 Chemical composition weight percentage of the experimental steel of the metal sample to be tested Clean the sample using a pre-grinder to remove any oxide scale and restore its metallic luster. Mechanically grind the sample using 120# to 800# sandpaper, ensuring that each grit size produces scratches in only one direction and that no visible scratches remain on the surface when using the finest sandpaper. Polish the sample using 2.5μm and then 1μm diamond polishing pastes, preferably for 1-5 minutes. This results in a mirror-like surface with no visible scratches.

[0030] Place the sample to be tested and the standardized sample into the sample chamber of the scanning electron microscope at the same time, and evacuate the sample chamber. Adjust the height between the sample surface and the lens, preferably 7-20 mm, more preferably 10-15 mm. Switch the SEM signal to backscattered signal, set the voltage to 20 kV, the aperture to 60, and select "high beam mode." Inspect the surface of the metallographic sample to be tested and adjust the image to be clear.

[0031] The automatic image analysis and measurement system is used to set the characteristic values of different steel inclusions: Field of view area setting: Set 1 area point, and the total market area of a single area point is 2 square millimeters.

[0032] Image parameter settings: Select "High" for scanning accuracy, select 1024*768 for measurement image accuracy, and the minimum detection size limit is required to be more than 5 times the pixel size.

[0033] X-ray parameter settings: Generally, the point analysis mode is used, the acquisition time is more than 200 milliseconds, the expected count rate is more than 10,000, the analysis threshold (based on the equivalent circle diameter) is set at 3 microns, and inclusions exceeding the threshold are collected using the "surface scan" mode.

[0034] Electron microscope working parameters: The electron microscope magnification is between 1000 times.

[0035] Brightness contrast setting: Adjust the contrast and brightness of the scanning electron microscope by using the steel sample and aluminum foil sample, and the background grayscale range is 120-170.

[0036] After adjusting the parameters, start the OTS automatic inclusion analysis software to detect and count the inclusions in the sample.

[0037] The obtained OTS data is imported into the optimized database, and the data is classified and statistically analyzed using a machine learning algorithm. The specific steps are as follows: Data preprocessing: clean the imported OTS data, remove noise data, and use the KNN algorithm to fill missing values; Feature extraction: extracting the type, size, chemical composition and other features of inclusions from the data; Model training: Convolutional neural network (CNN) is used to train the extracted features, and model parameters are optimized through cross-validation; Data analysis: Use the trained model to classify and count inclusions, and generate a detailed report on the type, size distribution, and chemical composition of inclusions.

[0038] The results of the inclusion quantity analysis of the stainless steel samples tested in this example are given in Figure 2 ,Depend on Figure 2 It can be seen that the rare earth inclusions in the experimental steel are mainly Al2O3 composite inclusions.

[0039] Example 2 The difference from Example 1 is that 0.0048% Ce is added to the test sample; the setting of the inclusion characteristic value is the same as the steps in Example 1.

[0040] The results of the inclusion quantity analysis of the stainless steel samples tested in this example are shown in Figure 3 ,Depend on Figure 3 It can be seen that the amount of CeAlO3 accounts for more than half of the total number of inclusions, and a large number of rare earth composite inclusions are found in the steel, and there are more types.

[0041] Example 3 The difference from Example 1 is that 0.0092% Ce is added to the test sample; the setting of the inclusion characteristic value is the same as that of Example 1. The results of the inclusion quantity analysis of the stainless steel samples tested in this example are given in Figure 4 .

[0042] See also Figure 5 , which shows the total number and size ratio of inclusions in Examples 1 to 3. It can be seen that the total number of inclusions is continuously decreasing, and the size of the inclusions is gradually increasing.

[0043] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present invention.

Claims

1. A method for fully automatic detection of inclusions in rare earth-containing stainless steel, characterized in that: include: Obtaining rare earth-containing stainless steel metal samples; According to the size range of different inclusions in rare earth-containing stainless steel metal samples, the inclusion detection method, detection area and measurement accuracy are determined, and the minimum detection size limit is determined; according to the equivalent circle diameter of the inclusion, the inclusion scanning analysis mode is determined; according to the detection area of the inclusion, the magnification range of the scanning electron microscope is determined; The OTS automatic inclusion analysis software is used to detect and count inclusions in rare earth-containing stainless steel metal samples based on the established detection method, detection area, measurement image accuracy, minimum detection size limit, scanning analysis mode, and scanning electron microscope magnification range to obtain the type, size, distribution, and chemical composition of the inclusions.

2. The method for fully automatic detection of inclusions in rare earth-containing stainless steel according to claim 1, characterized in that: The method of obtaining a rare earth-containing stainless steel metal sample specifically includes: Sandpaper of different grit sizes is used to grind rare earth-containing stainless steel. The scratch direction of each grit size sandpaper is consistent. The ground metal sample surface is polished with diamond polishing paste for a polishing time ranging from 1 minute to 5 minutes to obtain a rare earth-containing stainless steel metal sample.

3. The method for fully automatic detection of inclusions in rare earth-containing stainless steel according to claim 1, characterized in that: The method of determining the detection method, detection area, and measurement accuracy of inclusions according to the size range of different inclusions in the rare earth-containing stainless steel metal sample specifically includes: When the size range of different inclusions in the rare earth-containing stainless steel metal sample is within the range of 0 μm to 1 μm, determining that the detection method for inclusions is small-area multi-region point detection, and determining that the detection area and measurement accuracy of a single region are the first preset detection area and the first preset measurement accuracy; When the size range of different inclusions in the rare earth-containing stainless steel metal sample is within the range of 1 μm to 10 μm, the inclusion detection method is determined to be small-area multi-region point detection, and the detection area and measurement accuracy of a single area are determined to be the second preset detection area and the second preset measurement accuracy; When the sizes of the different inclusions in the sample to be tested are statistically determined to be within the range of 10 μm to 100 μm, the inclusion detection method is determined to be a large-area single-region detection method, and the detection area and measurement accuracy of the single region are determined to be the third preset detection area and the third preset measurement accuracy; The first preset detection area is smaller than the second preset detection area, and the second preset detection area is smaller than the third preset detection area; the first preset measurement accuracy is greater than the second preset measurement accuracy, and the second preset measurement accuracy is greater than the third preset measurement accuracy.

4. The method for fully automatic detection of inclusions in rare earth-containing stainless steel according to claim 1, characterized in that: The minimum detection size lower limit is 5 times the pixel size.

5. The method for fully automatic detection of inclusions in rare earth-containing stainless steel according to claim 1, characterized in that: Determining the scanning analysis mode of the inclusions according to the equivalent circle diameter of the inclusions specifically includes: When the equivalent circle diameter is less than or equal to a preset analysis threshold, determining that the scanning analysis mode of different inclusions in the rare earth-containing stainless steel is a point analysis mode; When the equivalent circle diameter is greater than a preset analysis threshold, it is determined that the scanning analysis mode of different inclusions in the rare earth-containing stainless steel is a surface analysis mode.

6. The method for fully automatic detection of inclusions in rare earth-containing stainless steel according to claim 1, characterized in that: In the process of detecting inclusions in a rare earth-containing stainless steel metal sample, the scanning time of the inclusions is greater than 200 milliseconds and the expected counting rate is greater than 10,000.

7. The method for fully automatic detection of inclusions in rare earth-containing stainless steel according to claim 1, characterized in that: Determining the magnification range of the scanning electron microscope based on the detection area of the inclusions specifically includes: When the inspection area of inclusions is less than or equal to 2 square millimeters, the magnification range of the scanning electron microscope is determined to be 600 times to 1000 times; When the inspection area of the inclusion is larger than 2 square millimeters and smaller than 20 square centimeters, the magnification range of the scanning electron microscope is determined to be 200 times to 600 times.