A digital rating method for macroscopic subsurface crack of continuous casting billet

Abstract

The application discloses a continuous casting billet low-magnification subsurface crack digital rating method and relates to the technical field of continuous casting processes.The method comprises the following steps: acquiring the cross-sectional size parameters of a continuous casting billet, determining the subsurface crack evaluation area thickness and the risk critical distance based on the cross-sectional size parameters, acquiring the defect characterization data of the continuous casting billet, identifying the subsurface cracks in the subsurface crack evaluation area defined by the subsurface crack evaluation area thickness, and extracting the crack characteristic parameters, and performing combined grading based on the subsurface crack evaluation area thickness, the risk critical distance and the crack characteristic parameters to generate the subsurface crack rating result.The application takes the shortest distance d of the subsurface crack from the surface, the defect number n and the defect length l as the grading input, adopts the explicit threshold and the combination rule to output the 0-6 grade level G, and converts the subsurface crack evaluation from the experience judgment to the reproducible rule judgment.

Description

Technical Field

[0001] This invention relates to the field of continuous casting technology, and in particular to a digital rating method for low-magnification subcutaneous cracks in continuously cast billets. Background Technology

[0002] Subsurface cracks in continuously cast billets are a significant defect affecting billet quality and subsequent rolling yield. Accurate rating of subsurface cracks during low-magnification inspection of continuously cast billets is crucial for process optimization, quality assessment, and product release.

[0003] Currently, the assessment of subsurface cracks in continuously cast billets both domestically and internationally mainly relies on the national standard GB / T 226-2015. This standard provides a basic definition and qualitative description of cracks, but it has the following shortcomings in practical applications: Inconsistent judgment criteria: Different inspectors have subjective differences in their comprehensive judgment of crack depth, quantity, and length, and the inspection results of the same sample may be inconsistent in different shifts or by different personnel.

[0004] The grading granularity is too coarse: the existing standard grading is limited, making it difficult to distinguish subcutaneous cracks of different severity levels precisely, and failing to provide a clear basis for differentiated treatment such as subsequent resurfacing, excision, and downgrading.

[0005] Lack of digital representation: Judgment results are mostly qualitative descriptions or simple grades, making it difficult to form calculable and traceable structured data, which is not conducive to data integration and process attribution analysis of quality management systems (QMS / MES).

[0006] The definition of the subcutaneous region is ambiguous: For continuously cast billets of different specifications, there is no unified adaptive definition for the thickness range of the "subcutaneous" region, which leads to conceptual drift when rating across specifications.

[0007] Therefore, there is an urgent need for a solution that can achieve refined, standardized, and digital rating of subcutaneous cracks. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a digital rating method for low-magnification subcutaneous cracks in continuously cast billets.

[0009] To solve the above technical problems, the technical solution of the present invention is as follows: A digital rating method for low-magnification subcutaneous cracks in continuously cast billets includes: Obtain the cross-sectional dimension parameters of the continuously cast billet, and determine the thickness of the subcutaneous crack assessment zone and the critical risk distance based on the cross-sectional dimension parameters; Obtain defect characterization data of the continuously cast billet; Subcutaneous cracks are identified within the subcutaneous crack assessment area defined by the thickness of the subcutaneous crack assessment area, and crack feature parameters are extracted. The subcutaneous crack rating result is generated by combining the thickness of the subcutaneous crack assessment area, the critical risk distance, and the crack characteristic parameters.

[0010] As a preferred embodiment of the digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to the present invention, the determination of the thickness of the subcutaneous crack assessment area and the critical risk distance based on cross-sectional dimension parameters includes: Obtain the side length B of the cross-section of the continuously cast billet; Determine the thickness t of the subcutaneous crack assessment area, where t = min(0.25B, t max ), where t max This is the maximum thickness limit set for the subcutaneous region; Determine the critical risk distance d0, where d0 = min(β*t, d 0，max ), where β is the set proportionality coefficient, d 0，max This is the upper limit of the critical distance set for risk.

[0011] As a preferred embodiment of the digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to the present invention, the method for determining the thickness of the subcutaneous crack assessment area and the critical risk distance based on cross-sectional dimension parameters further includes: Determine the length grading thresholds for the maximum length of subcutaneous cracks, wherein the length grading thresholds include a first length grading threshold l1 and a second length grading threshold l2. in, , This is the first length scaling factor. This is the upper limit of the first length; , This is the second length scaling factor. This is the upper limit of the second length.

[0012] As a preferred embodiment of the digital rating method for low-magnification subcutaneous cracks in continuous casting billets described in this invention, the crack characteristic parameters include the number of defects, the distance of the subcutaneous crack closest to the surface, and the maximum length of the subcutaneous crack. The extracted crack feature parameters include: Each subcutaneous crack within the subcutaneous crack assessment area is identified, and the number of defects n is calculated. Obtain the shortest distance di from the i-th subcutaneous crack to the nearest surface, and determine the minimum value among all shortest distances di as the subcutaneous crack distance d closest to the surface; Obtain the maximum projected length li of the i-th subcutaneous crack along its principal axis, and determine the maximum value among all maximum projected lengths li as the maximum length l of the subcutaneous crack.

[0013] As a preferred embodiment of the digital rating method for subsurface cracks in continuous casting billets according to the present invention, the following steps are included: Combining judgment levels are performed based on the thickness of the subsurface crack evaluation area, the risk critical distance, and the crack characteristic parameters, and the generated subsurface crack rating results include: If the number of defects n = 0, it is determined as level 0; If the number of defects n ≥ 1 and the distance d from the subsurface crack closest to the surface is ≤ d0, it is determined as level 6; If the number of defects n ≥ 1 and d > t, it is determined that the crack is not within the scope of subsurface crack evaluation; If the number of defects n ≥ 1 and d0 < d ≤ t, combined judgment is performed based on the number of defects n, the maximum length l of the subsurface crack, and the length grading threshold, specifically including: If n < 5, when l < l1, it is determined as level 1; when l1 ≤ l ≤ l2, it is determined as level 2; when l > l2, it is determined as level 3; If 5 ≤ n ≤ 8, when l ≤ l1, it is determined as level 2; when l1 ≤ l ≤ l2, it is determined as level 3; when l > l2, it is determined as level 4; If n > 8, when l ≤ l1, it is determined as level 3; when l1 ≤ l ≤ l2, it is determined as level 4; when l > l2, it is determined as level 5.

[0014] As a preferred embodiment of the digital rating method for subsurface cracks in continuous casting billets according to the present invention, after the subsurface crack rating results are generated, the following steps are further included: Encapsulating the rating results, the crack characteristic parameters, and the identification information of the continuous casting billet into structured data; Digitally storing the structured data.

[0015] As a preferred embodiment of the digital rating method for subsurface cracks in continuous casting billets according to the present invention, the method for obtaining the shortest distance di from the i-th subsurface crack to the nearest surface includes: calculating the minimum Euclidean distance from the crack profile to the nearest surface boundary.

[0016] As a preferred embodiment of the digital rating method for subsurface cracks in continuous casting billets according to the present invention, the method for obtaining the maximum projected length li of the i-th subsurface crack along its main axis includes: calculating the maximum projected length of the crack skeleton in the main axis direction or calculating the length of the main axis of the minimum circumscribed rectangle of the crack.

[0017] As a preferred embodiment of the digital rating method for subsurface cracks in continuous casting billets according to the present invention, the defect characterization data includes at least one of low magnification images, ultrasonic detection data, or eddy current detection data.

[0018] As a preferred embodiment of the digital rating method for subsurface cracks in continuous casting billets according to the present invention, where: the value range of β is 0.10–0.20, and the first length ratio coefficient has a value range of 0.18~0.25, and the first length ratio coefficient has a value range of 0.35~0.50, the upper limit of the first length has a value range of 5~6mm, the upper limit of the second length has a value range of 10~12 mm, the upper limit of the maximum thickness tmax of the subsurface area is set to have a value range of 20~30 mm, and the upper limit d of the set risk critical distance 0，max has a value range of 8~10 mm.

[0019] The beneficial effects of the present invention are as follows: (I) The present invention uses the shortest distance d from the subsurface crack to the surface, the number of defects n, and the length of the defect l as the input for grading, and outputs the grade G of 0-6 levels using clear thresholds and combination rules, converting the assessment of subsurface cracks from empirical judgment to reproducible rule-based judgment. Different inspectors will obtain exactly the same rating results for the same sample based on the same set of rules, thus significantly reducing the grading differences and improving the consistency and traceability of the results.

[0020] (2) For subsurface cracks where d0 < d ≤ t, the present invention uses the combination rules of n and l to form a gradient grading of levels 1-5. This refined grading can provide a clear grade basis for subsequent treatment measures such as grinding, cutting, repair, and downgrading, which is beneficial to reducing unnecessary cutting / downgrading caused by excessive conservatism and at the same time reducing the risk of defect spillage caused by being too lenient.

[0021] (3) The present invention sets a red line rule that directly determines the grade as 6 when d ≤ d0 and n ≥ 1. This rule enables subsurface cracks close to the surface to be clearly classified as the highest risk level once they appear. Regardless of their length and quantity, they will not be underestimated. This design significantly improves the certainty of the quality release judgment and effectively prevents the spillage of high-risk defects.

[0022] (4) The present invention introduces the thickness t of the subsurface crack assessment area. For a 150 mm * 150 mm continuous casting billet, t = 22.5 mm; for continuous casting billets of 200 mm and above, t is controlled by the upper limit of 25 mm. This "ratio + upper limit capping" adaptive definition method not only ensures that the subsurface area changes reasonably with the specifications of the continuous casting billet, but also avoids the subsurface area of large-section continuous casting billets being too deep and deviating from the engineering meaning. At the same time, cracks with d > t are clearly classified as other internal defects, ensuring the consistency of the rating caliber between different specifications of continuous casting billets from the source and facilitating stable promotion and application within the range of 150-250 mm continuous casting billets. Description of the Drawings

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic flowchart of the digital rating method for low-magnification subcutaneous cracks in continuously cast billets provided by the present invention. Detailed Implementation

[0025] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0026] See Figure 1 This application provides a digital rating method for low-magnification subcutaneous cracks in continuously cast billets, which specifically includes the following steps: Step S1: Obtain the cross-sectional dimension parameters of the continuously cast billet, and determine the thickness of the subcutaneous crack assessment zone and the critical risk distance based on the cross-sectional dimension parameters.

[0027] Specifically, first, the side length B (unit: mm) of the continuous casting billet section is obtained. Then, based on the section size parameters, the thickness of the subcutaneous crack assessment zone and the critical risk distance are determined. For common continuous casting billets, B ranges from 150 to 250 mm.

[0028] Regarding the thickness t of the subcutaneous crack assessment region, the subcutaneous crack assessment region refers to a ring-shaped region with a thickness of t extending inward from the four surfaces of the continuously cast billet. This application adopts an adaptive method of "proportionality + upper limit capping" to determine t: t = min(0.25B, t) max ), where t max The maximum thickness of the subcutaneous region is set to be 20-30 mm, preferably 25 mm. For example, when B=150 mm, 0.25B=37.5 mm, so t=25 mm; when B=200 mm, 0.25B=50 mm, so t=25 mm is still used. This avoids the subcutaneous region of large-section continuously cast billets being too deep, which would deviate from the engineering requirements.

[0029] The critical risk distance d0 is used to define high-risk areas near the surface, also using a "proportional + upper limit cap" approach: d0 = min(β*t, d 0，max ), where β is a set proportionality coefficient, ranging from 0.10 to 0.20, preferably 0.2, d 0，maxis the upper limit of the set risk critical distance, taking 8 - 10 mm, preferably 8 mm. Under the preferred parameters, d0 is usually 5 - 8 mm.

[0030] Additionally, in order to classify the length of the subcutaneous crack, the present application sets the first length classification threshold l1 and the second length classification threshold l2 (l1 < l2), and also adopts the method of ratio + upper limit capping: , , where is the first length ratio coefficient, is the first length upper limit; is the second length ratio coefficient and is the second length upper limit.

[0031] Among them, takes 0.18 - 0.25, preferably 0.20; takes 5 - 6 mm, preferably 5 mm; takes 0.35 - 0.50, preferably 0.40; takes 10 - 12 mm, preferably 10 mm.

[0032] Once the above parameters are determined according to the continuous casting billet specifications, they can be used as fixed thresholds for subsequent grading.

[0033] Step S2: Obtain the defect characterization data of the continuous casting billet.

[0034] Specifically, the defect characterization data refers to the data sources that can reflect the characteristics such as the position, quantity, and length of the subcutaneous crack, including but not limited to: Macroscopic image: The image obtained by photographing or scanning after etching or corroding the macroscopic specimen of the continuous casting billet; Ultrasonic detection data: Converting the ultrasonic flaw detection result into equivalent defect position and size data; Eddy current detection data: Converting the eddy current detection result into equivalent defect position and size data.

[0035] If the obtained is a macroscopic image, it is necessary to complete the image scale calibration through a scale or calibration block to convert the pixel size to the millimeter scale. If the obtained is an ultrasonic or eddy current detection signal, it is necessary to convert the signal scale to the millimeter scale.

[0036] Step S3: Identify the subcutaneous crack within the subcutaneous crack assessment area defined by the thickness of the subcutaneous crack assessment area, and extract the crack characteristic parameters.

[0037] Specifically, within the subcutaneous crack assessment area defined by the thickness t, each subcutaneous crack is identified and parameter extraction is performed. Among them, the crack characteristic parameters include the number of defects, the distance of the subcutaneous crack closest to the surface, and the maximum length of the subcutaneous crack. The specific extraction of crack characteristic parameters includes: (1) Number of defects n: Count the total number of subcutaneous cracks within the assessment area.

[0038] (2) Distance d of the subcutaneous crack closest to the surface: Calculate the minimum Euclidean distance from the contour of each crack to the nearest surface boundary. For automated implementation, an image processing algorithm can be used to extract the crack contour, and then calculate the shortest distance from each pixel point on the contour to the surface boundary, and take the minimum value as di. Take the minimum value among all di: .

[0039] (3) Maximum length l of the subcutaneous crack: First, calculate the maximum projected length li of the i-th crack. Along the main axis direction of the crack, calculate the maximum projected length. Specifically, one of the following two preferred methods can be used: Skeleton method: Extract the skeleton line of the crack, determine the main axis direction, project all points on the skeleton onto the main axis direction, and take the maximum distance between the projected points; Minimum bounding rectangle method: Calculate the minimum bounding rectangle that can completely enclose the crack, and take the length of the longer side of the rectangle.

[0040] Then, take the maximum value among all li: [[ID=L19]].

[0041] Step S4: Based on the thickness of the subcutaneous crack assessment area, the risk critical distance, and the crack characteristic parameters, perform combined grading to generate the subcutaneous crack rating result.

[0042] Specifically, according to d, n, l obtained in step S3 and t, d0, l1, l2 determined in step S1, output the rating level G (0 - 6 levels) according to the following rules: If the number of defects n = 0, it is determined as level 0; <L If the number of defects n ≥ 1 and the distance d of the subcutaneous crack closest to the surface ≤ d0, it is determined as level 6; If the number of defects n ≥ 1 and d > t, it is determined that the crack does not fall within the scope of subcutaneous crack assessment; If the number of defects n ≥ 1 and d0 < d ≤ t, perform combined determination based on the number of defects n, the maximum length l of the subcutaneous crack, and the length grading threshold, specifically including: If n < 5, when l < l1, it is determined as level 1, when l1 ≤ l ≤ l2, it is determined as level 2, and when l > l2, it is determined as level 3; If 5 ≤ n ≤ 8, then when l ≤ l1, it is judged as level 2; when l1 ≤ l ≤ l2, it is judged as level 3; and when l > l2, it is judged as level 4. If n>8, then when l ≤ l1, it is judged as level 3; when l1≤ l ≤ l2, it is judged as level 4; and when l>l2, it is judged as level 5.

[0043] Step S5: Output and digital storage of results.

[0044] Specifically, after the grading is completed, the rating results, crack characteristic parameters, and identification information of the continuously cast billet are encapsulated into structured data. The structured data format can be CSV, JSON, relational database fields, or interface data structures of an industrial quality management system (QMS / MES). By writing the data into the quality management system through the interface, a closed-loop management of "inspection-grading-traceability" can be achieved.

[0045] The above technical solution will be further explained below through specific embodiments.

[0046] Example 1: This example uses the low-magnification inspection of a 150 mm × 150 mm continuous casting billet from a steel plant as an example to illustrate the method of the present invention in detail.

[0047] Step 1: Determine the thickness t of the subcutaneous crack assessment area and the critical risk distance d0.

[0048] The side length of the continuously cast billet is B=150 mm. Take t_max=25 mm, then: t = min(0.25×150, 25) = min(37.5, 25) = 25 mm.

[0049] If we take β=0.2 and d0,max=8 mm, then: d0 = min(0.2×25, 8) = min(5, 8) = 5 mm.

[0050] Pick =0.20, =5 mm, then: l1 = min(0.20×25, 5) = min(5, 5) = 5 mm.

[0051] Pick =0.40, =10 mm, then: l2 = min(0.40×25, 10) = min(10, 10) = 10 mm.

[0052] Step 2: Obtain defect characterization data.

[0053] The macrostructure specimens of the continuous casting billets are subjected to acid etching treatment to obtain macrostructure images. The calibration from pixel size to millimeter scale is completed through the image scale.

[0054] Step 3: Identify the subcutaneous cracks and extract the characteristic parameters.

[0055] Within the subcutaneous crack evaluation area defined by the thickness t = 25 mm, 4 subcutaneous cracks are identified. The measured results are as follows: The shortest distances di of each crack from the surface are respectively: 3.2 mm, 4.5 mm, 6.1 mm, 12.3 mm; The maximum projected lengths li of each crack are respectively: 2.1 mm, 3.5 mm, 4.7 mm, 1.8 mm.

[0056] Calculated: d = min(di) = 3.2 mm; n = 4; l = max(li) = 4.7 mm.

[0057] Step 4: Combined grading: Since n = 4 ≥ 1 and d = 3.2 mm ≤ d0 = 5 mm, according to the grading rule "if n ≥ 1 and d ≤ d0, it is graded as grade 6", so the grading result of the subcutaneous cracks of this specimen is grade 6.

[0058] Step 5: Result output and storage The grading result G = 6, along with d = 3.2 mm, n = 4, l = 4.7 mm, t = 25 mm, together with the identification information such as the furnace number, steel grade, sampling location, etc., are combined into structured data in JSON format and written into the quality management system.

[0059] Example 2: In this example, taking a continuous casting billet of 200 mm × 200 mm with another furnace number as an example, the grading process of subcutaneous cracks in the interval of d0 < d ≤ t is shown.

[0060] Step 1: Determine the parameters B = 200 mm, t = min(0.25 × 200, 25) = min(50, 25) = 25 mm; d0 = min(0.2 × 25, 8) = 5 mm; l1 = min(0.20 × 25, 5) = 5 mm; l2 = min(0.40 × 25, 10) = 10 mm.

[0061] Steps 2 - 3: Extract the characteristic parameters 6 subcutaneous cracks are identified, and the measured results are as follows: The values of di are respectively: 7.2 mm, 8.5 mm, 9.0 mm, 11.3 mm, 10.5 mm, 12.0 mm; The values of li are respectively: 3.5 mm, 6.2 mm, 4.8 mm, 5.5 mm, 7.0 mm, 8.5 mm.

[0062] Calculated: d = 7.2 mm, n = 6, l = 8.5 mm.

[0063] Step 4: Combined grading Since n = 6 ≥ 1, and d0 = 5 < d = 7.2 ≤ t = 25, enter the combined grading of n and l. n = 6 belongs to the interval "5 ≤ n ≤ 8", and l = 8.5 mm satisfies l1 = 5 ≤ l = 8.5 ≤ l2 = 10. It is determined to be Grade 3 according to the rules.

[0064] Step 5: Output G = 3 and store it in a structured manner.

[0065] Thus, the technical solution of this application uses the shortest distance d from the subcutaneous crack to the surface, the number of defects n, and the length of the defect l as the input for grading, and uses clear thresholds and combination rules to output the grade G from 0 to 6, converting the evaluation of subcutaneous cracks from empirical judgment to reproducible rule-based judgment.

[0066] In addition to the above embodiments, the present invention may have other embodiments; all technical solutions formed by equivalent replacement or equivalent transformation fall within the protection scope required by the present invention.

Claims

1. A digital rating method for low-magnification subcutaneous cracks in continuously cast billets, characterized in that: include: Obtain the cross-sectional dimension parameters of the continuously cast billet, and determine the thickness of the subcutaneous crack assessment zone and the critical risk distance based on the cross-sectional dimension parameters; Obtain defect characterization data of the continuously cast billet; Subcutaneous cracks are identified within the subcutaneous crack assessment area defined by the thickness of the subcutaneous crack assessment area, and crack feature parameters are extracted. The subcutaneous crack rating result is generated by combining the thickness of the subcutaneous crack assessment area, the critical risk distance, and the crack characteristic parameters.

2. The digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to claim 1, characterized in that: The determination of the subcutaneous crack assessment zone thickness and risk threshold distance based on cross-sectional dimension parameters includes: Obtain the side length B of the cross-section of the continuously cast billet; Determine the thickness t of the subcutaneous crack assessment area, where t = min(0.25B, t max In the formula, t max This is the maximum thickness limit set for the subcutaneous region; Determine the critical risk distance d0, where d0 = min(β*t, d 0，max ), where β is the set proportionality coefficient, d 0，max This is the upper limit of the critical distance set for risk.

3. The digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to claim 2, characterized in that: The method of determining the thickness of the subcutaneous crack assessment zone and the critical risk distance based on cross-sectional dimension parameters also includes: Determine the length grading thresholds for the maximum length of subcutaneous cracks, wherein the length grading thresholds include a first length grading threshold l1 and a second length grading threshold l2. in, , This is the first length scaling factor. This is the upper limit of the first length; , This is the second length scaling factor. This is the upper limit of the second length.

4. The digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to claim 1, characterized in that: The crack characteristic parameters include the number of defects, the distance of the subcutaneous crack closest to the surface, and the maximum length of the subcutaneous crack; The extracted crack feature parameters include: Each subcutaneous crack within the subcutaneous crack assessment area is identified, and the number of defects n is calculated. Obtain the shortest distance di from the i-th subcutaneous crack to the nearest surface, and determine the minimum value among all shortest distances di as the subcutaneous crack distance d closest to the surface; Obtain the maximum projected length li of the i-th subcutaneous crack along its principal axis, and determine the maximum value among all maximum projected lengths li as the maximum length l of the subcutaneous crack.

5. The digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to claim 4, characterized in that: The subcutaneous crack rating result is generated by combining the thickness of the subcutaneous crack assessment area, the critical risk distance, and the crack characteristic parameters for classification. If the number of defects n = 0, then it is determined to be level 0; If the number of defects n≥ 1, and the distance d of the subcutaneous crack closest to the surface ≤ d0, then it is judged as level 6; If the number of defects n ≥ 1 and d > t, then the crack is determined not to be within the scope of subcutaneous crack assessment. If the number of defects n ≥ 1, and d0 < d ≤ t, then a combined determination is made based on the number of defects n, the maximum length l of the subcutaneous crack, and the length grading threshold, specifically including: If n < 5, then when l < l1, it is judged as level 1; when l1 ≤ l ≤ l2, it is judged as level 2; and when l > l2, it is judged as level 3. If 5 ≤ n ≤ 8, then when l ≤ l1, it is judged as level 2; when l1 ≤ l ≤ l2, it is judged as level 3; and when l > l2, it is judged as level 4. If n > 8, then when l ≤ l1, it is judged as level 3; when l1 ≤ l ≤ l2, it is judged as level 4; and when l > l2, it is judged as level 5.

6. The digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to claim 1, characterized in that: Following the generation of the subcutaneous crack rating result, the following is also included: The rating results, the crack characteristic parameters, and the identification information of the continuously cast billet are encapsulated into structured data; The structured data is stored digitally.

7. The digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to claim 4, characterized in that: Obtaining the shortest distance di from the i-th subcutaneous crack to the nearest surface includes: calculating the minimum Euclidean distance from the crack profile to the nearest surface boundary.

8. The digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to claim 4, characterized in that: The step of obtaining the maximum projected length li of the i-th subcutaneous crack along its principal axis includes: calculating the maximum projected length of the crack skeleton along the principal axis or calculating the principal axis length of the minimum bounding rectangle of the crack.

9. The digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to claim 1, characterized in that: The defect characterization data includes at least one of low-magnification images, ultrasonic testing data, or eddy current testing data.

10. The digital rating method for low-magnification subcutaneous cracks in continuously cast billets according to claim 1, characterized in that: The value of β ranges from 0.10 to 0.20, and the first length scaling factor... The value range is 0.18~0.25, and the first length scaling factor The value range is 0.35~0.50, and the upper limit of the first length is... The value range is 5~6 mm, and the upper limit of the second length is... The value range is 10~12 mm, the maximum subcutaneous thickness limit tmax is set to a value range of 20~30 mm, and the maximum risk critical distance limit d is set. 0，max The value range is 8~10 mm.

Images