A method for identifying surface defects of a stainless steel hot-rolled pickled plate derived from a continuous casting billet

Abstract

The present application belongs to the technical field of stainless steel metallographic microscopic detection and steel production quality traceability, a kind of stainless steel hot-rolled pickled plate surface defect discrimination method derived from continuous casting billet, the micro-morphology and element detection of stainless steel hot-rolled pickled plate surface defect are carried out by scanning electron microscope combined with X-ray energy spectrometer, a macro-morphology screening-oxide skin layering quantitative characterization-multiple point retest consistency triple linkage closed-loop determination system is constructed, the present application sets the oxide skin element, thickness quantitative determination interval, uses instrument objective detection value as determination basis, reduces the artificial determination deviation caused by qualitative observation, and the consistency of detection results of different operators is high.

Description

[0001] manual Technical Field

[0002] This invention belongs to the field of metallographic microscopic detection of stainless steel and traceability of steel production quality. Specifically, it relates to a method for identifying the process source of surface defects on hot-rolled pickled stainless steel sheets, which is applicable to the process source traceability detection of various surface defects on hot-rolled pickled stainless steel sheets of 304 austenitic stainless steel, 316 austenitic stainless steel, and 2101 duplex stainless steel. Background Technology

[0003] The production process of hot-rolled pickled stainless steel sheets is as follows: continuous casting of molten steel, high-temperature heating of the billet, rough rolling, finish rolling, high-pressure descaling, and pickling. During the prolonged high-temperature environment in the heating furnace, multiple layers of primary oxide scale form inside surface gaps and subcutaneous defects in the billet. The hot rolling process, with its shorter oxidation time, only produces a single thin oxide film. The structures of these two types of oxide scale are significantly different and can serve as a basis for defect tracing.

[0004] Existing technologies have disclosed the use of scanning electron microscopy and X-ray energy dispersive spectroscopy to observe the morphology of oxide scale defects in stainless steel. They have also disclosed techniques such as the three-layer basic structure of oxide scale on stainless steel billets, the classification of defects on stainless steel plates, and the addition of re-measurement when single-point observations show anomalies. However, existing technical solutions have the following technical problems: Only qualitative observation of oxide scale layering was performed, without setting a unified and quantifiable range for element and thickness determination, resulting in poor consistency of determination results among different operators; The macroscopic classification of defects, the microscopic quantitative detection, and the multi-point retesting are independent of each other, and no linkage judgment logic has been established, resulting in redundant detection process and large judgment error. Conventional water washing during sample preparation will generate a secondary oxide film on the sample surface, which will cover up the layering details of the original oxide scale. The retest points were randomly selected, which did not completely cover the entire defect area, and single-point observations were easily affected by local trace oxides. Different grades of stainless steel use different evaluation standards, making it difficult to compare test data across different grades.

[0005] Existing publicly available literature only records the above-mentioned single technical means, and does not disclose the technical solution that combines standardized sample preparation, graded quantitative characterization, oxide scale quantification threshold, defect linkage screening, and multi-point retesting across the entire range to form a complete closed-loop judgment system. Summary of the Invention

[0006] The purpose of this invention is to overcome at least one technical problem existing in the prior art and to provide a method for identifying surface defects of hot-rolled pickled stainless steel plates derived from continuously cast billets. By unifying and standardizing the detection process, setting a quantitative judgment range for oxide scale, and constructing a triple linkage judgment logic, the method achieves standardized defect source tracing and judgment, reproducible results, reduces human judgment bias, and improves the accuracy and efficiency of defect source tracing.

[0007] The technical solution adopted in this invention is: a method for identifying surface defects in hot-rolled pickled stainless steel sheets derived from continuously cast billets. This method employs scanning electron microscopy combined with X-ray energy dispersive spectroscopy to perform microscopic morphology and elemental analysis on the surface defects of hot-rolled pickled stainless steel sheets. A triple-linked closed-loop judgment system is constructed, consisting of macroscopic morphology screening, quantitative characterization of oxide scale layering, and multi-point retest consistency analysis. The method includes the following steps: Step 1: Preparation of standardized defect specimens: Cut specimens and clean them; after processing, prepare specimens with complete cross-sections of defects for observation. Step 2: Hierarchical and Layered Microscopic Quantitative Characterization: The scanning electron microscope (SEM) operating voltage was set to 20.00 kV, and backscattered electron imaging mode was adopted. Oxide residues in the defect interlayer were located at 100–200x magnification, and the thickness of each layer of the oxide scale was observed and measured at high magnification of 600–3000x. X-ray energy dispersive spectroscopy (EDS) was used to perform point-to-point elemental analysis on the inner, middle, and outer layers of the oxide scale, collecting the mass fractions of Cr, Fe, and O in each layer. The defect cross-section was observed using a scanning electron microscope (SEM) in backscattered electron (BSE) mode. Special attention was paid to observing whether there were oxide residues at the bottom or inside the defect, and analyzing the layered structure of these residual oxides. Simultaneously, point analysis or area scanning analysis was performed using energy dispersive spectroscopy (EDS) to determine the characteristic distribution of the main elements in each layer of oxides.

[0008] Step 3: Preliminary quantitative judgment of oxide scale layering: The primary oxide scale formed by stainless steel continuous casting billets after long-term high-temperature heating above 1000℃ consists of an inner layer of chromium-rich oxide, an intermediate layer of iron-chromium composite spinel, and an outer layer of continuous iron-rich oxide from the matrix outwards; the hot-rolled short-time oxidation products do not have a complete three-layer composite structure. Quantitative determination threshold: The main components of the inner layer of chromium-rich oxide are Cr2O3 and FeCr3O4, with a Cr element mass fraction of 28% to 40%; the main components of the outer layer of iron-rich oxide are Fe2O3 and Fe3O4, with a Fe element mass fraction of 65% to 85%, and the outer layer of iron-rich oxide is continuous and has a thickness of 3μm to 10μm. If the residual oxide in the defective interlayer has a complete three-layer structure and all indicators fall within the corresponding value range, the initial microscopic judgment is qualified; if any indicator is not met, the defect is directly judged as a hot-rolled secondary defect. The core criteria for determining that the defect originated from the continuously cast billet are: if the residual oxide observed at the defect site exhibits a distinct multi-layered structure, and there is a continuous, iron-rich oxide layer in the multi-layered structure, or there is an independent, complete iron oxide layer, then the defect can be determined to have originated from the continuously cast billet.

[0009] Scientific Principle: The oxide scale formed on stainless steel (especially high-chromium stainless steel) in the high-temperature zone (>1000℃) of continuously cast billets typically has the following structure from the outside to the inside: iron-rich oxide layer (Fe2O3 / Fe3O4) → composite oxide layer ((Fe,Cr)3O4) → chromium-rich oxide inner layer (Cr2O3). The outermost iron-rich oxide layer is the hallmark of the high-temperature oxide scale on continuously cast billets. When defects such as inclusions, bubbles, or cracks exist on the surface of the billet, this high-temperature oxide scale will be trapped inside or at the bottom of the defects. During subsequent hot rolling and pickling processes, the oxide scale on the surface of the plate is removed, but this "high-temperature scale" remaining inside the defects may be preserved, becoming a "fingerprint" for tracing its origin.

[0010] Step 4: Pre-screening of binary defects: Surface defects of hot-rolled pickled stainless steel plates are divided into two categories. The first category is defects that are easily inherited from the original defects of the continuous casting billet, and the second category is secondary defects generated separately by the hot rolling process. Only when the macroscopic morphology of the defect belongs to the first category of defects can it enter the multi-point retest process. If it belongs to the second category, it is directly determined that the defect originated from the hot rolling process. Step 5, Multi-point Retest Consistency Verification: When the initial microscopic judgment in Step 3 is qualified and the defect in Step 4 is determined to be of the first type, select no less than 3 defect interlayer points and repeat Step 2 and Step 3; only when all retest points meet the quantitative threshold in Step 3, and the triple judgment conditions are met simultaneously, is the defect ultimately determined to be inherited from the original defect of the continuously cast billet; if any point fails to meet the standard, the defect is determined to be a secondary defect of hot rolling.

[0011] Based on the macroscopic morphology of the defects (such as their distribution along the rolling direction and the shape of the opening) and the above microscopic analysis results, a final judgment is given.

[0012] In step one, the sample size is 20mm×15mm and is a stainless steel hot-rolled pickled plate sample containing complete defects; the cleaning is to remove oil stains, pickling residue and solid impurities from the sample surface by ultrasonic cleaning; the treatment is resin inlay, multi-stage sandpaper grinding and diamond polishing paste mirror polishing.

[0013] For the defect area to be identified in a hot-rolled and pickled stainless steel sheet, a metallographic sample is taken perpendicular to the sheet surface, ensuring that the sample contains a complete cross-section of the defect. The sample is then mounted, ground, and polished to prepare a test sample suitable for microscopic observation of the defect cross-section. Alternatively, a sample can be directly cut from the defect area of ​​the hot-rolled and pickled stainless steel sheet for direct surface observation and analysis.

[0014] In step two, the thickness of the oxide layer is measured using the built-in scale of a scanning electron microscope; at least three sets of elemental detection data are collected for each layer of oxide, and the arithmetic mean is taken as the elemental mass fraction of that layer.

[0015] The discrimination method is applicable to the surface defect detection of hot-rolled pickled plates of 304 austenitic stainless steel, 316 austenitic stainless steel and 2101 duplex stainless steel. The quantitative judgment threshold described in step three is used uniformly for the three types of steel.

[0016] In step one, the ultrasonic cleaning equipment has a power of 400W, the cleaning medium is anhydrous ethanol, and the cleaning time is no less than 5 minutes to avoid secondary oxide film interference from water washing.

[0017] In step five, three retesting points were selected: the head sandwich, the middle sandwich, and the tail sandwich, respectively, to completely cover the rolling extension range of the defect.

[0018] The beneficial effects of this invention are: this invention sets up quantitative judgment ranges for oxide scale elements and thickness, uses objective instrument detection values ​​as the judgment basis, reduces human judgment bias caused by qualitative observation, and ensures high consistency of detection results among different operators.

[0019] This invention constructs a three-pronged judgment logic of macroscopic screening, microscopic quantification, and multi-point retesting, with hierarchical constraints and cross-validation to reduce the probability of misjudgment in tracing caused by local trace oxides.

[0020] This invention standardizes the ultrasonic sample preparation of anhydrous ethanol, avoiding the formation of secondary oxide films during water washing that obscure the primary oxide layering and improving the clarity of microscopic observation.

[0021] This invention limits the retesting of defects at three points: the beginning, middle, and end, to fully characterize the true distribution of oxide scale in defect interlayers, eliminate the randomness of single-point observation, and adopt a set of quantitative standards to adapt to three types of mainstream stainless steel. Test data within the enterprise and from different branches can be compared horizontally to unify quality inspection standards.

[0022] This invention's macroscopic defect pre-screening can skip meaningless retesting processes, reduce the workload of high-precision microscopic detection, and improve laboratory testing efficiency. Attached Figure Description

[0023] Figure 1 This is a 200x SEM image of the peeling surface of 304 stainless steel. Figure 2 This is an 800x SEM image of the peeling surface of 304 stainless steel. Figure 3 This is a 1600x SEM image of the peeling surface of 304 stainless steel. Figure 4 This is a 3000x SEM image of the peeling surface of 304 stainless steel. Figure 5 This is a diagram showing the location of EDS measurement points for the Cr-rich oxide layer on 304 stainless steel with peeling surface. Figure 6 yes Figure 5 Spectral composition curves of the measurement points; Figure 7 This is a diagram showing the location of EDS measurement points for the Fe-rich oxide layer on the peeling surface of 304 stainless steel. Figure 8 yes Figure 7 Spectral composition curves of the measurement points; Figure 9 This is a 112x SEM image of a heavy-duty 2101 duplex stainless steel sheet. Figure 10 This is a 1600x SEM image of a heavy-duty 2101 duplex stainless steel sheet. Figure 11 This is a 600x SEM image of a heavy-duty 2101 duplex stainless steel sheet. Figure 12 This is a 3000x SEM image of a heavy-duty 2101 duplex stainless steel sheet. Figure 13 This is a map showing the EDS measurement points of the Cr-rich oxide layer on the heavy skin of 2101 duplex stainless steel. Figure 14 yes Figure 13 Spectral composition curves of the measurement points; Figure 15 This is a map showing the location of EDS measurement points for the Fe-rich oxide layer on the heavy skin of 2101 duplex stainless steel. Figure 16 yes Figure 15 Spectral composition curves of the measurement points. Detailed Implementation

[0024] Example 1: Tracing the source of linear "peeling" defects distributed along the rolling direction on the surface of hot-rolled pickled 304 austenitic stainless steel sheets from a steel plant.

[0025] A method for identifying surface defects in hot-rolled pickled stainless steel sheets derived from continuously cast billets employs scanning electron microscopy combined with X-ray energy dispersive spectroscopy to perform microscopic morphology and elemental analysis on the surface defects of hot-rolled pickled stainless steel sheets. A triple-linked closed-loop judgment system is constructed, consisting of macroscopic morphology screening, quantitative characterization of oxide scale layering, and multi-point retest consistency analysis. The method includes the following steps: Step 1: Preparation of standardized defect sample: Cut a 20mm×15mm stainless steel hot-rolled pickled plate sample containing complete defects, and use 400W anhydrous ethanol for 6 minutes to ultrasonically clean the sample surface to remove oil stains and avoid secondary oxide film interference caused by water washing, pickling residue and solid impurities; prepare a sample with complete defect cross section for observation by resin inlay, multi-level sandpaper grinding and diamond polishing paste mirror polishing.

[0026] Step 2: Hierarchical and Layered Microscopic Quantitative Characterization: Set the scanning electron microscope operating voltage to 20.00 kV and use BSE mode; (e.g.) Figure 1As shown, at 200x magnification, a layer of scattered oxide scale residue was observed between the bottom of the "peeling" defect and the substrate, such as... Figure 2-3 As shown, at magnifications of 800 and 1600, the oxide scale exhibits a layered structure. The thickness of each layer of the oxide scale was measured at a high magnification of 3000, and the arithmetic mean of three sets of EDS data was taken as the elemental mass fraction of that layer. X-ray energy dispersive spectroscopy was used to perform point-based elemental analysis on the inner, middle, and outer layers of the oxide scale, and the mass fractions of Cr, Fe, and O in each layer were collected. Figure 5 As shown, X-ray energy dispersive spectroscopy (EDS) analysis indicates that the inner layer of the layered oxide crust is rich in O and Cr, such as... Figure 7 As shown, the outer layer is rich in Fe and O.

[0027] Step 3: Preliminary quantitative judgment of oxide scale layering: The primary oxide scale formed by stainless steel continuous casting billets after long-term high-temperature heating above 1000℃ consists of an inner layer of chromium-rich oxide, an intermediate layer of iron-chromium composite spinel, and an outer layer of continuous iron-rich oxide from the matrix outwards; the hot-rolled short-time oxidation products do not have a complete three-layer composite structure. Quantitative determination threshold: The main components of the inner layer of chromium-rich oxide are Cr2O3 and FeCr3O4, with a Cr element mass fraction of 28% to 40%; the main components of the outer layer of iron-rich oxide are Fe2O3 and Fe3O4, with a Fe element mass fraction of 65% to 85%, and the outer layer of iron-rich oxide is continuous and has a thickness of 3μm to 10μm. If the residual oxide in the defective interlayer has a complete three-layer structure and all indicators fall within the corresponding value range, the initial microscopic assessment is qualified; if any indicator is not met, the defect is directly judged as a hot-rolled secondary defect.

[0028] like Figure 5-8 As shown, in this embodiment, the average Cr content of the inner layer is 32.5%, the average Fe content of the outer layer is 72.1%, the thickness of the iron-rich layer is 4.6 μm, the three-layer structure is complete, and the initial microscopic assessment is satisfactory.

[0029] Step 4: Pre-screening of binary defects: Surface defects of hot-rolled pickled stainless steel plates are divided into two categories. The first category is defects that are easily inherited from the original defects of the continuous casting billet, and the second category is secondary defects generated separately by the hot rolling process. Only when the macroscopic morphology of the defect belongs to the first category of defects can it enter the multi-point retesting process. If it belongs to the second category, it is directly determined that the defect originated from the hot rolling process.

[0030] This embodiment belongs to the first category of defects that are easily inherited from the original defects of the continuously cast billet, and enters the retesting process.

[0031] Step 5, Consistency Verification of Multi-Point Retesting: If the initial microscopic judgment in Step 3 is qualified and the defect in Step 4 is classified as a Class I defect, select 3 defect interlayer locations and repeat Steps 2 and 3. Only when all retesting locations meet the quantitative threshold in Step 3, and all three judgment conditions are met simultaneously, is the defect ultimately determined to be inherited from the original defect in the continuously cast billet. If any location fails to meet the standard, the defect is determined to be a secondary defect in hot rolling. The 3 retesting locations are selected as the defect head interlayer, middle interlayer, and tail interlayer, respectively, completely covering the defect rolling extension range.

[0032] In this embodiment, three points were retested: the detection results of the defect head, middle and tail interlayer all met the quantification range, and the three conditions were met simultaneously. Conclusion: The "peeling" defect was determined to originate from the continuously cast billet. The likely cause is the presence of micro-cracks or inclusions on the billet surface, allowing high-temperature oxide scale to penetrate and become exposed during hot rolling due to failure to weld it in. Optimization direction: Optimize the uniformity of secondary cooling in the crystallizer to reduce micro-cracks on the billet surface.

[0033] Example 2: Tracing the source of linear "heavy skin" defects on the surface of hot-rolled pickled 2101 duplex stainless steel sheets from a steel plant.

[0034] A method for identifying surface defects in hot-rolled pickled stainless steel sheets derived from continuously cast billets employs scanning electron microscopy combined with X-ray energy dispersive spectroscopy to perform microscopic morphology and elemental analysis on the surface defects of hot-rolled pickled stainless steel sheets. A triple-linked closed-loop judgment system is constructed, consisting of macroscopic morphology screening, quantitative characterization of oxide scale layering, and multi-point retest consistency analysis. The method includes the following steps: Step 1: Preparation of standardized defect sample: Cut a 20mm×15mm stainless steel hot-rolled pickled plate sample containing complete defects, and use 400W anhydrous ethanol for 5 minutes to ultrasonically clean the sample surface to remove oil stains and avoid secondary oxide film interference caused by water washing, pickling residue and solid impurities; prepare a sample with complete defect cross section for observation by resin inlay, multi-level sandpaper grinding and diamond polishing paste mirror polishing.

[0035] Step 2: Hierarchical and Layered Microscopic Quantitative Characterization: Set the scanning electron microscope operating voltage to 20.00 kV and use BSE mode; (e.g.) Figure 9 As shown, at 112x magnification, a layer of oxide scale residue was observed between the bottom of the "double-skin" defect and the substrate, as... Figure 10-12 As shown, at magnifications of 600 and 1600, the oxide scale exhibits a layered structure. The thickness of each layer of the oxide scale was measured at a high magnification of 3000, and the arithmetic mean of three sets of EDS data was taken as the elemental mass fraction of that layer. X-ray energy dispersive spectroscopy was used to perform point-based elemental analysis on the inner, middle, and outer layers of the oxide scale, and the mass fractions of Cr, Fe, and O in each layer were collected. Figure 13As shown, X-ray energy dispersive spectroscopy (EDS) analysis indicates that the inner layer of the layered oxide crust is rich in O and Cr, such as... Figure 15 As shown, the outer layer is rich in Fe and O.

[0036] Step 3: Preliminary quantitative judgment of oxide scale layering: The primary oxide scale formed by stainless steel continuous casting billets after long-term high-temperature heating above 1000℃ consists of an inner layer of chromium-rich oxide, an intermediate layer of iron-chromium composite spinel, and an outer layer of continuous iron-rich oxide from the matrix outwards; the hot-rolled short-time oxidation products do not have a complete three-layer composite structure. Quantitative determination threshold: The main components of the inner layer of chromium-rich oxide are Cr2O3 and FeCr3O4, with a Cr element mass fraction of 28% to 40%; the main components of the outer layer of iron-rich oxide are Fe2O3 and Fe3O4, with a Fe element mass fraction of 65% to 85%, and the outer layer of iron-rich oxide is continuous and has a thickness of 3μm to 10μm. If the residual oxide in the defective interlayer has a complete three-layer structure and all indicators fall within the corresponding value range, the initial microscopic assessment is qualified; if any indicator is not met, the defect is directly judged as a hot-rolled secondary defect.

[0037] like Figure 13-16 As shown in this embodiment, The inner layer has an average Cr content of 30.2%, the outer layer has an average Fe content of 70.8%, and the iron-rich layer has a thickness of 5.1 μm. The three-layer structure is intact and the initial microscopic assessment is satisfactory.

[0038] Step 4: Pre-screening of binary defects: Surface defects of hot-rolled pickled stainless steel plates are divided into two categories. The first category is defects that are easily inherited from the original defects of the continuous casting billet, and the second category is secondary defects generated separately by the hot rolling process. Only when the macroscopic morphology of the defect belongs to the first category of defects can it enter the multi-point retesting process. If it belongs to the second category, it is directly determined that the defect originated from the hot rolling process.

[0039] In this embodiment, the macroscopic feature is a heavy skin, which belongs to the first category of defects that are easily inherited from the original defects of the continuously cast billet, and thus enters the retesting process.

[0040] Step 5, Consistency Verification of Multi-Point Retesting: If the initial microscopic judgment in Step 3 is qualified and the defect in Step 4 is classified as a Class I defect, select 3 defect interlayer locations and repeat Steps 2 and 3. Only when all retesting locations meet the quantitative threshold in Step 3, and all three judgment conditions are met simultaneously, is the defect ultimately determined to be inherited from the original defect in the continuously cast billet. If any location fails to meet the standard, the defect is determined to be a secondary defect in hot rolling. The 3 retesting locations are selected as the defect head interlayer, middle interlayer, and tail interlayer, respectively, completely covering the defect rolling extension range.

[0041] In this embodiment, three-point retesting was conducted: the detection results of the defect head, middle and tail interlayer all met the quantification range, and the three conditions were met simultaneously. Source tracing conclusion: The "replenishment scale" defect was determined to originate from the continuously cast billet. The likely cause is the presence of micro-cracks on the billet surface, into which high-temperature oxide scale penetrated, failing to fuse during hot rolling and extending into a linear replenishment scale defect. Optimization directions: Stabilize the liquid level in the crystallizer and optimize the performance of the protective slag.

Claims

1. A method for identifying surface defects in hot-rolled pickled stainless steel sheets derived from continuously cast billets, comprising using scanning electron microscopy combined with X-ray energy dispersive spectroscopy to perform microscopic morphology and elemental analysis on surface defects of hot-rolled pickled stainless steel sheets, characterized in that, A closed-loop judgment system is constructed, which combines macroscopic morphology screening, layered quantitative characterization of oxide scale, and consistency evaluation at multiple points. The method includes the following steps: Step 1: Preparation of standardized defect specimens: Cut specimens and clean them; after processing, prepare specimens with complete cross-sections of defects for observation. Step 2: Hierarchical and Layered Microscopic Quantitative Characterization: The scanning electron microscope (SEM) operating voltage was set to 20.00 kV, and backscattered electron imaging mode was adopted. The oxide residue in the defect interlayer was located at 100-200x magnification, and the thickness of each layer of oxide scale was observed and measured at 600-3000x high magnification. X-ray energy dispersive spectroscopy was used to perform point-to-point elemental detection on the inner, middle, and outer layers of oxide scale, and the mass fractions of Cr, Fe, and O elements in each layer were collected. Step 3: Preliminary quantitative judgment of oxide scale layering: The primary oxide scale formed by stainless steel continuous casting billets after long-term high-temperature heating above 1000℃ consists of an inner layer of chromium-rich oxide, an intermediate layer of iron-chromium composite spinel, and an outer layer of continuous iron-rich oxide from the matrix outwards; the hot-rolled short-time oxidation products do not have a complete three-layer composite structure. Quantitative determination threshold: The main components of the inner layer of chromium-rich oxide are Cr2O3 and FeCr3O4, with a Cr element mass fraction of 28% to 40%; the main components of the outer layer of iron-rich oxide are Fe2O3 and Fe3O4, with a Fe element mass fraction of 65% to 85%, and the outer layer of iron-rich oxide is continuous and has a thickness of 3μm to 10μm. If the residual oxide in the defective interlayer has a complete three-layer structure and all indicators fall within the corresponding value range, the initial microscopic judgment is qualified; if any indicator is not met, the defect is directly judged as a hot-rolled secondary defect. Step 4: Pre-screening of binary defects: Surface defects of hot-rolled pickled stainless steel plates are divided into two categories. The first category is defects that are easily inherited from the original defects of the continuous casting billet, and the second category is secondary defects generated separately by the hot rolling process. Only when the macroscopic morphology of the defect belongs to the first category of defects can it enter the multi-point retest process. If it belongs to the second category, it is directly determined that the defect originated from the hot rolling process. Step 5, Multi-point Retest Consistency Verification: When the initial microscopic judgment in Step 3 is qualified and the defect in Step 4 is determined to be of the first type, select no less than 3 defect interlayer points and repeat Step 2 and Step 3; only when all retest points meet the quantitative threshold in Step 3, and the triple judgment conditions are met simultaneously, is the defect ultimately determined to be inherited from the original defect of the continuously cast billet; if any point fails to meet the standard, the defect is determined to be a secondary defect of hot rolling.

2. The method for identifying surface defects in hot-rolled pickled stainless steel plates derived from continuously cast billets according to claim 1, characterized in that: In step one, the sample size is 20mm×15mm and is a stainless steel hot-rolled pickled plate sample containing complete defects; the cleaning is to remove oil stains, pickling residue and solid impurities from the sample surface by ultrasonic cleaning; the treatment is resin inlay, multi-stage sandpaper grinding and diamond polishing paste mirror polishing.

3. The method for identifying surface defects in hot-rolled pickled stainless steel plates derived from continuously cast billets according to claim 1, characterized in that: In step two, the thickness of the oxide layer is measured using the built-in scale of a scanning electron microscope; at least three sets of elemental detection data are collected for each layer of oxide, and the arithmetic mean is taken as the elemental mass fraction of that layer.

4. The method for identifying surface defects in hot-rolled pickled stainless steel plates derived from continuously cast billets according to claim 1, characterized in that: The discrimination method is applicable to the surface defect detection of hot-rolled pickled plates of 304 austenitic stainless steel, 316 austenitic stainless steel and 2101 duplex stainless steel. The quantitative judgment threshold described in step three is used uniformly for the three types of steel.

5. The method for identifying surface defects in hot-rolled pickled stainless steel plates derived from continuously cast billets according to claim 1, characterized in that: In step one, the ultrasonic cleaning equipment has a power of 400W, the cleaning medium is anhydrous ethanol, and the cleaning time is no less than 5 minutes to avoid secondary oxide film interference from water washing.

6. The method for identifying surface defects in hot-rolled pickled stainless steel plates derived from continuously cast billets according to claim 1, characterized in that: In step five, three retesting points were selected: the head sandwich, the middle sandwich, and the tail sandwich, respectively, to completely cover the rolling extension range of the defect.

Images