Online formulation identification and quality inspection method and system for longitudinally variable thickness steel coils

Abstract

This invention provides an online formula identification and quality inspection method and system for longitudinally variable thickness steel coils, belonging to the field of intelligent manufacturing and online inspection technology in the iron and steel metallurgy industry. The method includes: synchronously collecting continuous thickness data and corresponding position information of the steel coil along its length direction to generate a continuous thickness-length curve; identifying continuous segments with thickness fluctuations less than a preset threshold on the thickness-length curve as candidate stable segments, and selecting effective stable segments from the candidate stable segments; using pre-stored formula knowledge, matching the arrangement order of the effective stable segments along the length direction of the steel coil with the process sequence constraints to determine the set of continuous segments corresponding to the target formula, and completing the formula identification based on the set of continuous segments; performing full-length thickness tolerance detection on the steel coil according to the identified formula, and outputting the detection results and alarm information.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing and online inspection technology in the iron and steel metallurgy industry, specifically to a method and system for automatic recipe identification, matching and quality inspection of longitudinally variable thickness steel coils in continuous processing or post-processing based on process sequence constraints. Background Technology

[0002] In the manufacturing of high-end steel products, variable thickness steel coils (also known as unequal thickness steel plates or differential thickness plates) are widely used in automotive lightweighting, high-end home appliances, and special containers because they can achieve optimal material and thickness distribution on a single component. These steel coils are rolled along their length according to a pre-set thickness "formula" (i.e., thickness variation program) to form multiple continuous segments of different thicknesses.

[0003] Currently, the following practical problems exist in the subsequent processing (such as laser cutting, stamping, and slitting) and quality inspection of variable thickness steel coils:

[0004] 1. Loss of formula information and reliance on manual comparison: After the steel coils have passed through the production process, there are no intuitive thickness segment markings on the physical coils. Downstream operators must rely on paper or electronic process documents to manually compare and locate different thickness areas to switch processing parameters, which is inefficient and prone to errors, affecting production cycle and finished product qualification rate.

[0005] 2. Online identification is susceptible to strong noise interference: The production site is subject to complex interference factors such as equipment vibration, strip jitter, and temperature fluctuations, resulting in high noise and significant local fluctuations in the thickness data acquired online. Traditional judgment methods based on a single threshold or instantaneous data features are difficult to distinguish between actual thickness changes and measurement noise, easily leading to misidentification or missed identification, and failing to meet the requirements for identification stability and reliability under continuous production conditions.

[0006] 3. Traditional testing methods cannot achieve continuous formula verification: Existing thickness testing equipment (such as single-point thickness gauges) mostly perform sampling inspections and cannot continuously collect thickness data for the entire coil. Therefore, it is impossible to verify whether the actual rolled thickness distribution is consistent with the original design formula, posing potential quality risks. In a continuous data stream, accurately and robustly determining the end of one thickness segment and the beginning of another is a key challenge in achieving online automatic identification.

[0007] 4. Difficulty in seamless integration of automated production lines: In intelligent production lines, existing testing systems are mostly limited to quality judgment and lack the ability to convert the identified "recipe" information into a structured output that can drive downstream equipment (such as cutting machines and welding robots) to automatically adjust process parameters, thus hindering full-process automation.

[0008] While existing technologies include automatic thickness detection systems for uniform steel coils and image processing-based product surface recognition technologies, most are designed for uniform plates or only provide thickness deviation alarms. Neither effectively addresses the core industrial challenge of online, automatic, and highly reliable formula identification for longitudinally variable thickness steel coils in noisy, continuous production environments. Therefore, how to fully utilize the inherent process characteristics of steel processing in noisy, continuous production environments to achieve online, automatic, and highly reliable identification of longitudinally variable thickness steel coil formulas remains a critical technical problem that urgently needs to be solved. Summary of the Invention

[0009] In view of the technical defects and drawbacks existing in the prior art, the present invention provides an online formula identification and quality inspection method and system for longitudinally variable thickness steel coils to overcome the above problems or at least partially solve the above problems. The specific solution is as follows:

[0010] As a first aspect of the present invention, an online formula identification and quality inspection method for longitudinally variable thickness steel coils is provided, comprising the following steps:

[0011] S1. Data Acquisition and Preprocessing: Simultaneously acquire continuous thickness data and corresponding position information of the steel coil along its length direction, and perform noise reduction and smoothing processing on the thickness data to generate a continuous thickness-length curve.

[0012] S2. Extraction and screening of stable segments: On the thickness-length curve, continuous segments with thickness fluctuations less than a preset threshold are identified as candidate stable segments, and effective stable segments are screened from the candidate stable segments according to preset segment length rationality rules and thickness change physical continuity rules.

[0013] S3. Recipe identification based on process sequence constraints: Using pre-stored recipe knowledge, which includes at least the fixed order of each thickness segment in the target recipe as a process sequence constraint, the arrangement order of the effective stable segments in the length direction of the steel coil is matched with the process sequence constraint to determine the set of continuous segments corresponding to the target recipe, and the recipe identification is completed based on the set of continuous segments.

[0014] S4. Quality Inspection and Result Output: Based on the identified formula, perform full-length thickness tolerance inspection on the steel coil and output the inspection results and alarm information.

[0015] In some embodiments, in step S2, the preset segment length rationality rule is: to remove candidate stable segments whose length is less than a first preset ratio of the minimum formula segment length or whose length is greater than a second preset ratio of the maximum formula segment length; the thickness change physical continuity rule is: to remove candidate stable segments whose thickness change between adjacent sampling points within a segment exceeds the process allowable mutation threshold; wherein, the first preset ratio and the second preset ratio are determined based on the process parameters of the target formula, and the process allowable mutation threshold is determined based on the physical limits of the steel coil rolling process.

[0016] In some embodiments, step S2, after applying the paragraph length rationality rule and the thickness change physical continuity rule, further includes formula consistency prediction, specifically including: calculating the average thickness of the candidate stable segments after being screened by the first two rules, comparing the average thickness with the thickness of each standard formula segment in the formula knowledge base, and removing candidate stable segments whose average thickness does not fall within the allowable deviation range of any standard formula segment thickness to obtain the final effective stable segment; wherein, the formula knowledge base and the formula knowledge in S3 are the same data source, and the allowable deviation range of thickness is determined based on the process accuracy requirements of the target formula.

[0017] In some embodiments, step S3, the recipe identification based on process sequence constraints, specifically includes:

[0018] S31. Coarse positioning: Based on the arrangement order of the effective stable segments in the length direction of the steel coil, obtain the process sequence constraints of the target formula from the formula knowledge, find the continuous segment sequence that matches the process sequence constraints in the thickness segment type order, and preliminarily determine the possible position range of the target formula on the steel coil.

[0019] S32, Fine matching: For each possible position interval determined in S31, the continuous paragraph sequence corresponding to each effective stable segment is extracted, including the paragraph start point, paragraph end point, and at least one preset proportion position within the segment, forming a set of feature thickness points for the stable segment.

[0020] S33. Formula Confirmation: Based on the set of characteristic thickness points obtained in S32, a weighted comprehensive deviation calculation is performed between the characteristic thickness points of the target formula standard segment in the formula knowledge, and the formula with the smallest weighted comprehensive deviation is determined as the final identification result; wherein, the weighted comprehensive deviation is calculated by assigning weights to the deviation of each characteristic thickness point and then performing a weighted summation.

[0021] In some embodiments, in step S32, at least one preset proportional position within the segment is selected from at least one of the segment midpoint, quarter point, and three-quarter point; wherein, the selection of the preset proportional position is determined based on the thickness distribution characteristics of the steel coil rolling process, the thickness distribution characteristics including thickness uniformity requirements and detection accuracy requirements during the rolling process.

[0022] In some embodiments, in step S33, the weighted comprehensive deviation is calculated by assigning weights to the deviations of each feature thickness point and then performing a weighted summation; wherein, the weights corresponding to the start and end points of the paragraph are higher than the weights corresponding to other feature points within the paragraph, and the allocation of the weights is determined based on the importance of the feature thickness points in recipe recognition, and the importance is determined by the contribution of the feature thickness points to the positioning of the recipe segment boundary.

[0023] In some embodiments, step S4 includes: comparing the standard thickness and allowable deviation range corresponding to the identified formula with the actual thickness data of the steel coil point by point; when any of the following conditions occur, the thickness at that location is determined to be abnormal and marked:

[0024] (1) The actual thickness of multiple consecutive sampling points exceeds the allowable deviation range;

[0025] (2) The actual thickness of a single sampling point exceeds the allowable deviation range by more than a preset safety threshold;

[0026] The number of consecutive sampling points is determined based on process quality requirements and detection accuracy, and the preset safety threshold is determined based on the physical limits of the steel coil rolling process and the accuracy of the detection system.

[0027] In some embodiments, the method further includes result application, specifically including:

[0028] The identified formula and the precise start and end positions of each thickness segment are sent to the downstream processing equipment through a preset industrial communication interface, driving the downstream processing equipment to automatically call the matching processing parameters based on the received formula information and start and end position information.

[0029] In some embodiments, the method further includes manual-assisted identification, specifically including:

[0030] After step S2 or step S3 is executed, the current recognition result is displayed through the human-computer interaction interface, allowing users to manually adjust or specify the effective stable segment interval, and the formula recognition and thickness detection process is retried based on the manually adjusted interval. The system records the manual correction results and uses them to optimize the parameters of the automatic recognition model.

[0031] As a second aspect of the present invention, a longitudinally variable thickness steel coil online formula identification and detection system for implementing the method described in any of the above claims is provided, comprising:

[0032] The sensing and acquisition unit includes a thickness sensor and a displacement sensor, which are used to synchronously acquire continuous thickness data and position information of the steel coil;

[0033] The data processing unit is configured to perform the data acquisition and preprocessing steps and the stationary segment extraction and filtering steps;

[0034] The recipe identification unit is configured to perform the recipe identification step based on process sequence constraints;

[0035] The quality inspection unit is configured to perform the quality inspection and result output steps;

[0036] The human-machine interface and control unit is used to provide an operating interface, display results, receive instructions, and realize communication with downstream processing equipment.

[0037] The present invention has the following beneficial effects:

[0038] This invention first obtains high-quality continuous thickness-length curves through data acquisition and preprocessing, laying a reliable data foundation for subsequent analysis. Then, through the steps of extracting and screening stable segments, it effectively eliminates pseudo-stable signals caused by on-site noise, equipment vibration, etc., by applying rules such as the rationality of segment length and the physical continuity of thickness changes. On this basis, it utilizes the characteristic that each thickness segment in the variable thickness rolling process must follow a fixed sequence to match the arrangement order of effective stable segments with preset formula knowledge. Finally, based on the identified formula, it performs full-length thickness tolerance detection, realizing continuous and full-coverage quality inspection of the entire steel coil. Attached Figure Description

[0039] Figure 1 This is a flowchart illustrating an online formula identification and quality inspection method for longitudinally variable thickness steel coils provided in an embodiment of the present invention.

[0040] Figure 2 A schematic diagram of the thickness curve and effective stable section screening of longitudinally variable thickness steel coils provided in an embodiment of the present invention;

[0041] Figure 3 A schematic diagram of recipe segment matching based on process sequence constraints provided in an embodiment of the present invention;

[0042] Figure 4 This is a schematic diagram illustrating the extraction of effective stable segment feature thickness points provided in an embodiment of the present invention;

[0043] Figure 5This is a schematic diagram of full-length thickness tolerance detection based on formula identification provided in an embodiment of the present invention;

[0044] Figure 6 A structural block diagram of an online formula identification and detection system for longitudinally variable thickness steel coils provided in an embodiment of the present invention;

[0045] Figure 7 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0046] To enable those skilled in the art to better understand the technical solutions of the present invention, exemplary embodiments of the present invention are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0047] Where there is no conflict, the various embodiments of the present invention and the features thereof may be combined with each other.

[0048] As used herein, the term “and / or” includes any and all combinations of one or more related enumerated entries.

[0049] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that when the terms “comprising” and / or “made of” are used in this specification, the presence of the stated feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded. Terms such as “connected” or “linked” are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0050] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having the meaning consistent with their meaning in the context of the relevant art and the invention, and will not be interpreted as having an idealized or overly formal meaning unless expressly so defined herein.

[0051] In the technical solution of this invention, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information all comply with relevant laws and regulations and do not violate public order and good morals. The use of user data in this technical solution follows relevant national laws and regulations (e.g., the "Information Security Technology - Personal Information Security Specification"). For example: appropriate measures are taken for personal information access control; restrictions are imposed on the display of personal information; the purpose of using personal information does not exceed the scope of direct or reasonable association; and explicit identity targeting is eliminated when using personal information to avoid precisely locating a specific individual.

[0052] To address at least one of the technical problems existing in the aforementioned related technologies, the present invention provides an online formula identification and quality inspection method for longitudinally variable thickness steel coils. Figure 1 A flowchart illustrating an online formula identification and quality inspection method for longitudinally variable thickness steel coils provided in this embodiment of the invention includes the following steps:

[0053] S1. Data Acquisition and Preprocessing: Simultaneously acquire continuous thickness data and corresponding position information of the steel coil along its length direction, and perform noise reduction and smoothing processing on the thickness data to generate a continuous thickness-length curve.

[0054] S2. Extraction and screening of stable segments: On the thickness-length curve, continuous segments with thickness fluctuations less than a preset threshold are identified as candidate stable segments, and effective stable segments are screened from the candidate stable segments according to preset segment length rationality rules and thickness change physical continuity rules.

[0055] S3. Recipe identification based on process sequence constraints: Using pre-stored recipe knowledge, which includes at least the fixed order of each thickness segment in the target recipe as a process sequence constraint, the arrangement order of the effective stable segments in the length direction of the steel coil is matched with the process sequence constraint to determine the set of continuous segments corresponding to the target recipe, and the recipe identification is completed based on the set of continuous segments.

[0056] S4. Quality Inspection and Result Output: Based on the identified formula, perform full-length thickness tolerance inspection on the steel coil and output the inspection results and alarm information.

[0057] This invention first obtains high-quality continuous thickness-length curves through data acquisition and preprocessing, laying a reliable data foundation for subsequent analysis. Then, through the steps of extracting and screening stable segments, it effectively eliminates pseudo-stable signals caused by on-site noise, equipment vibration, etc., by applying rules such as the rationality of segment length and the physical continuity of thickness changes. On this basis, it utilizes the characteristic that each thickness segment in the variable thickness rolling process must follow a fixed sequence to match the arrangement order of effective stable segments with preset formula knowledge. Finally, based on the identified formula, it performs full-length thickness tolerance detection, realizing continuous and full-coverage quality inspection of the entire steel coil.

[0058] In some embodiments, in step S2, the preset segment length rationality rule is: to remove candidate stable segments whose length is less than a first preset ratio of the minimum formula segment length or whose length is greater than a second preset ratio of the maximum formula segment length; the thickness change physical continuity rule is: to remove candidate stable segments whose thickness change between adjacent sampling points within a segment exceeds the process allowable mutation threshold; wherein, the first preset ratio and the second preset ratio are determined based on the process parameters of the target formula, and the process allowable mutation threshold is determined based on the physical limits of the steel coil rolling process.

[0059] In this embodiment, by setting rules for the rationality of segment length and the physical continuity of thickness variation, pseudo-stable segments caused by measurement noise, local fluctuations, or process anomalies can be effectively eliminated, significantly improving the screening accuracy of effective stable segments. This provides a more reliable data foundation for subsequent formula identification based on process sequence constraints, thereby enhancing the anti-interference capability and identification accuracy of the entire identification system.

[0060] Taking the online formula identification of a certain longitudinally variable thickness steel coil as an example, let the target formula be a three-segment structure of "3mm-5mm-3mm", with standard lengths of 50m, 80m and 50m for each segment.

[0061] The specific implementation of step S2 includes:

[0062] After data preprocessing, multiple candidate stationary segments were identified on the thickness-length curve. Assumption:

[0063] Section A: Length 45m, average thickness 3.1mm.

[0064] Section B: Length 85m, average thickness 5.2mm.

[0065] Section C: Length 48m, average thickness 3.0mm.

[0066] Section D: Length 25m, average thickness 3.0mm (too short).

[0067] Segment E: 90m in length, 5.1mm in average thickness (but with abrupt changes inside).

[0068] Apply paragraph length appropriateness rules:

[0069] Minimum formulation section length: 50m, with the first preset ratio set at 0.6 (i.e., 30m).

[0070] Maximum formula section length: 80m, take the second preset ratio as 1.2 (i.e. 96m).

[0071] Removal rules: Segments with a length <30m or a length >96m are removed.

[0072] Result: Segment D (25m < 30m) was removed.

[0073] Apply the physical continuity rule of thickness variation:

[0074] The allowable abrupt change threshold for the process is set at 0.1 mm (based on the rolling process, the thickness variation between adjacent points should not exceed 0.1 mm).

[0075] Check the thickness variation of adjacent sampling points within each segment:

[0076] Segment E: At location 35m, the thickness of adjacent sampling points abruptly changes from 5.0mm to 5.3mm, a change of 0.3mm > 0.1mm.

[0077] Result: Segment E was removed.

[0078] The final effective stable segments: segments A, B, and C pass the screening and proceed to the subsequent formula identification step.

[0079] In some embodiments, step S2, after applying the paragraph length rationality rule and the thickness change physical continuity rule, further includes formula consistency prediction, specifically including: calculating the average thickness of the candidate stable segments after being screened by the first two rules, comparing the average thickness with the thickness of each standard formula segment in the formula knowledge base, and removing candidate stable segments whose average thickness does not fall within the allowable deviation range of any standard formula segment thickness to obtain the final effective stable segment; wherein, the formula knowledge base and the formula knowledge in S3 are the same data source, and the allowable deviation range of thickness is determined based on the process accuracy requirements of the target formula.

[0080] In this embodiment, by adding a formula consistency prediction step, based on the physical continuity screening of segment length and thickness changes, further screening is performed based on the matching degree between the thickness value itself and the target formula. This can effectively eliminate pseudo-stable segments whose thickness values ​​deviate from the standard formula due to measurement system errors, local material anomalies, or process fluctuations, forming a triple screening mechanism of "geometric features (length) + physical features (continuity) + numerical features (thickness value)". This significantly improves the screening accuracy and reliability of effective stable segments, providing a more accurate data foundation for subsequent formula identification based on process sequence constraints.

[0081] Following the aforementioned embodiment, after applying the paragraph length rationality rule and the thickness change physical continuity rule, three candidate stable segments are obtained: segment A (45m, 3.1mm), segment B (85m, 5.2mm), and segment C (48m, 3.0mm).

[0082] The specific implementation of the formula consistency prediction step includes:

[0083] Obtain the target formula parameters:

[0084] The target formulation is a three-segment structure of "3mm-5mm-3mm".

[0085] Standard formula segment thickness: first segment 3.0mm, second segment 5.0mm, third segment 3.0mm.

[0086] Thickness tolerance range: Based on process accuracy requirements, it is set to ±0.2mm (i.e. 2.8mm-3.2mm, 4.8mm-5.2mm).

[0087] Calculate the average thickness of each segment:

[0088] Section A: 45m, average thickness 3.1mm.

[0089] Section B: 85m, average thickness 5.2mm.

[0090] Section C: 48m, average thickness 3.0mm.

[0091] Thickness comparison and screening:

[0092] Segment A (3.1mm): falls within the range of 3.0mm±0.2mm (2.8mm-3.2mm), and is retained.

[0093] Segment B (5.2mm): falls within the range of 5.0mm±0.2mm (4.8mm-5.2mm), and is retained.

[0094] Segment C (3.0mm): If it falls within the range of 3.0mm±0.2mm, retain it.

[0095] Final effective stable segments: Segments A, B, and C all passed the formulation consistency prediction and were selected as the final effective stable segments to enter the subsequent formulation identification steps.

[0096] In some embodiments, the recipe identification based on process sequence constraints in step S3 specifically includes:

[0097] S31. Coarse positioning: Based on the arrangement order of the effective stable segments in the length direction of the steel coil, obtain the process sequence constraints of the target formula from the formula knowledge, find the continuous segment sequence that matches the process sequence constraints in the thickness segment type order, and preliminarily determine the possible position range of the target formula on the steel coil.

[0098] S32, Fine matching: For each possible position interval determined in S31, the continuous paragraph sequence corresponding to each effective stable segment is extracted, including the paragraph start point, paragraph end point, and at least one preset proportion position within the segment, forming a set of feature thickness points for the stable segment.

[0099] S33. Formula Confirmation: Based on the set of characteristic thickness points obtained in S32, a weighted comprehensive deviation calculation is performed between the characteristic thickness points of the target formula standard segment in the formula knowledge, and the formula with the smallest weighted comprehensive deviation is determined as the final identification result; wherein, the weighted comprehensive deviation is calculated by assigning weights to the deviation of each characteristic thickness point and then performing a weighted summation.

[0100] In this embodiment, the formula identification process is clearly divided into two stages: "coarse positioning (based on process sequence constraints)" and "fine matching (based on weighted deviation of feature thickness points)". This not only utilizes the rapid locking effect of process sequence constraints on the identification range, but also improves the identification accuracy through the weighted comprehensive deviation calculation of multiple feature points, forming a dual verification mechanism of "coarse positioning + fine matching". This effectively avoids misidentification caused by inaccurate matching of a single feature point, significantly improves the accuracy and robustness of formula identification, and reduces computational complexity, thereby improving online detection efficiency.

[0101] Following the aforementioned embodiment, after screening, the effective stable sections are obtained: section A (45m, 3.1mm), section B (85m, 5.2mm), and section C (48m, 3.0mm), arranged in the order of A→B→C along the length of the steel coil.

[0102] The specific implementation of step S3 includes:

[0103] S31 coarse positioning:

[0104] The process sequence constraint for the target formulation is: 3mm segment → 5mm segment → 3mm segment.

[0105] The effective smooth segment arrangement order is: segment A (3.1mm, possibly 3mm segment) → segment B (5.2mm, possibly 5mm segment) → segment C (3.0mm, possibly 3mm segment).

[0106] Matching judgment: The thickness segment types are in the same order (thin → thick → thin), and the possible location range is initially determined to be from the start of segment A to the end of segment C.

[0107] S32 fine matching:

[0108] For this possible location interval (segment A → segment B → segment C), extract the feature thickness points of each segment:

[0109] Segment A: Starting point thickness 3.1mm, ending point thickness 3.1mm, midpoint thickness 3.1mm.

[0110] Segment B: starting point thickness 5.2mm, ending point thickness 5.2mm, midpoint thickness 5.2mm.

[0111] Segment C: Starting point thickness 3.0mm, ending point thickness 3.0mm, midpoint thickness 3.0mm.

[0112] This constitutes a set of characteristic thickness points (3 points for each segment).

[0113] S33 formula confirmed:

[0114] Characteristic thickness points of the standard section of the target formulation (assuming that the standard section has a uniform thickness):

[0115] The first segment is 3mm: starting point 3.0mm, ending point 3.0mm, midpoint 3.0mm.

[0116] The second segment is 5mm: starting point 5.0mm, ending point 5.0mm, midpoint 5.0mm.

[0117] The third segment is 3mm: starting point 3.0mm, ending point 3.0mm, midpoint 3.0mm.

[0118] Calculate the weighted average deviation (assuming weights: starting point 0.4, ending point 0.4, midpoint 0.2):

[0119] Segment A deviation: (|3.1-3.0|×0.4 + |3.1-3.0|×0.4 + |3.1-3.0|×0.2) = 0.1.

[0120] Segment B deviation: (|5.2-5.0|×0.4 + |5.2-5.0|×0.4 + |5.2-5.0|×0.2) = 0.2.

[0121] Segment C deviation: (|3.0-3.0|×0.4 + |3.0-3.0|×0.4 + |3.0-3.0|×0.2) = 0.

[0122] Total deviation: 0.1 + 0.2 + 0 = 0.3.

[0123] If the deviation from other formulas (such as "3mm-3mm-5mm") is greater, then confirm that the current formula is "3mm-5mm-3mm".

[0124] In some embodiments, in step S32, at least one preset proportional position within the segment is selected from at least one of the segment midpoint, quarter point, and three-quarter point; wherein, the selection of the preset proportional position is determined based on the thickness distribution characteristics of the steel coil rolling process, the thickness distribution characteristics including thickness uniformity requirements and detection accuracy requirements during the rolling process.

[0125] In some embodiments, in step S33, the weighted comprehensive deviation is calculated by assigning weights to the deviations of each feature thickness point and then performing a weighted summation; wherein, the weights corresponding to the start and end points of the paragraph are higher than the weights corresponding to other feature points within the paragraph, and the allocation of the weights is determined based on the importance of the feature thickness points in recipe recognition, and the importance is determined by the contribution of the feature thickness points to the positioning of the recipe segment boundary.

[0126] In the above embodiments, by reasonably selecting the location of characteristic thickness points within a segment (such as the midpoint, quarter point, etc.) and determining the location selection criteria based on process characteristics, and by setting a weighting strategy with higher weights for the start and end points, the characteristics of steel coil thickness distribution and the boundary information of the formula segment can be reflected more accurately, thereby improving the representativeness and recognition accuracy of the characteristic thickness point set, thus enhancing the accuracy and robustness of formula recognition and reducing the risk of misidentification caused by improper selection of characteristic points or unreasonable weight allocation.

[0127] Following the aforementioned embodiments, fine matching is performed on segments A (45m, 3.1mm), B (85m, 5.2mm), and C (48m, 3.0mm).

[0128] The specific implementation of step S32 includes:

[0129] Location selection based on process characteristics: In the steel coil rolling process, the thickness distribution is usually relatively uniform, but there may be transition zones at the segment boundaries. Therefore, the midpoint, quarter point, and three-quarter point of the segment are selected as the location of the feature points.

[0130] Extracting feature thickness points:

[0131] Section A (45m): Thickness 3.1mm at the starting point (0m), 3.1mm at the ending point (45m), 3.1mm at the midpoint (22.5m), 3.1mm at the quarter point (11.25m), and 3.1mm at the three-quarter point (33.75m).

[0132] Section B (85m): starting point thickness 5.2mm, ending point thickness 5.2mm, midpoint thickness 5.2mm, quarter point thickness 5.2mm, three-quarter point thickness 5.2mm.

[0133] Section C (48m): starting point thickness 3.0mm, ending point thickness 3.0mm, midpoint thickness 3.0mm, quarter point thickness 3.0mm, three-quarter point thickness 3.0mm.

[0134] Construct a set of feature thickness points: You can choose to extract three points, namely the start point, the end point, and the midpoint, or multiple points such as the start point, the end point, the quarter point, the three-quarter point, and the midpoint.

[0135] Specific implementation of step S33:

[0136] Weights are determined based on importance: the start and end points of a paragraph are crucial for locating the boundary of the paragraph, so they are given higher weights (e.g., 0.4 for the start point and 0.4 for the end point), while other points within the paragraph (e.g., the midpoint) are given lower weights (e.g., 0.2).

[0137] Calculate the weighted aggregate deviation (taking segment A as an example):

[0138] Target formulation standard segment: 3mm segment, starting point 3.0mm, ending point 3.0mm, midpoint 3.0mm.

[0139] Actual measurement of segment A: starting point 3.1mm, ending point 3.1mm, midpoint 3.1mm.

[0140] Deviation calculation:

[0141] Starting point deviation: |3.1-3.0| = 0.1mm, weighted: 0.1×0.4 = 0.04.

[0142] Endpoint deviation: |3.1-3.0| = 0.1mm, weighted: 0.1×0.4 = 0.04.

[0143] Midpoint deviation: |3.1-3.0| = 0.1mm, weighted: 0.1×0.2 = 0.02.

[0144] Weighted average deviation: 0.04 + 0.04 + 0.02 = 0.1.

[0145] Formula confirmation: After calculating the deviations of each segment, sum them up, and the formula with the smallest total deviation is the identification result.

[0146] In some embodiments, step S4 includes: comparing the standard thickness and allowable deviation range corresponding to the identified formula with the actual thickness data of the steel coil point by point; when any of the following conditions occur, the thickness at that location is determined to be abnormal and marked:

[0147] (1) The actual thickness of multiple consecutive sampling points exceeds the allowable deviation range;

[0148] (2) The actual thickness of a single sampling point exceeds the allowable deviation range by more than a preset safety threshold;

[0149] The number of consecutive sampling points is determined based on process quality requirements and detection accuracy, and the preset safety threshold is determined based on the physical limits of the steel coil rolling process and the accuracy of the detection system.

[0150] In this embodiment, by setting two judgment conditions, namely "multiple consecutive sampling points exceeding the allowable deviation range" and "a single sampling point exceeding the safety threshold", the continuous thickness deviation and local abrupt defects of the steel coil can be detected simultaneously. This avoids false alarms caused by instantaneous measurement noise and can promptly capture serious local defects, forming a dual judgment mechanism that takes into account both detection sensitivity and anti-interference. This significantly improves the accuracy and reliability of thickness quality detection and ensures that unqualified steel coils can be effectively identified and marked.

[0151] Following the aforementioned embodiment, the identified formula is "3mm-5mm-3mm", with standard thicknesses of 3.0mm for the first segment, 5.0mm for the second segment, and 3.0mm for the third segment; the allowable deviation range is set to ±0.2mm (i.e., 2.8mm-3.2mm, 4.8mm-5.2mm).

[0152] The specific implementation of step S4 includes:

[0153] Parameter settings:

[0154] Number of consecutive sampling points: Based on process requirements, it is set to 5 consecutive sampling points (sampling interval 0.5m, i.e., a continuous length of 2.5m).

[0155] Preset safety threshold: Based on the limits of the rolling process, it is set to exceed the allowable deviation range by 0.5mm (i.e., the thickness of a single point exceeds the allowable range by more than 0.5mm).

[0156] Point-by-point comparison and detection (taking a certain segment as an example):

[0157] Assuming the actual thickness at a certain location on the steel coil is:

[0158] Sampling point 1: 3.3mm (0.1mm beyond the 3.2mm upper limit).

[0159] Sampling point 2: 3.4mm (0.2mm over).

[0160] Sampling point 3: 3.5mm (0.3mm beyond).

[0161] Sampling point 4: 3.6mm (0.4mm over).

[0162] Sampling point 5: 3.7mm (0.5mm over).

[0163] Sampling point 6: 3.8mm (0.6mm over).

[0164] Sampling point 7: 3.9mm (0.7mm over).

[0165] Sampling point 8: 4.0mm (exceeding 0.8mm).

[0166] Sampling point 9: 4.1mm (0.9mm over).

[0167] Sampling point 10: 4.2mm (exceeding 1.0mm).

[0168] Judgment process:

[0169] Starting from sampling point 1, five consecutive points (points 1-5) exceed the allowable deviation range, triggering condition (1) and determining that the thickness of the area is abnormal.

[0170] Alternatively, if sampling point 10 exceeds the allowable deviation range of 1.0 mm and exceeds the preset safety threshold of 0.5 mm, trigger condition (2) and determine that the thickness of the point is abnormal.

[0171] Tagging results:

[0172] The system marks abnormal locations (such as the area between points 1 and 10) and outputs alarm information.

[0173] In some embodiments, reference Figure 2The diagram shown illustrates the thickness curve and effective stable segment selection process of a longitudinally variable thickness steel coil according to an embodiment of the present invention. It shows the thickness curve along the length of the steel coil and the process of selecting effective stable segments. The horizontal axis represents the length of the steel coil (meters), and the vertical axis represents the thickness (millimeters). Multiple stable thickness segments are marked on the continuous thickness curve. S1, S2, S3, and S4 represent multiple effective stable segments extracted from the continuous thickness curve. These effective stable segments, together with their adjacent transition segments, constitute the effective thickness formula segment D1. N1 represents an abnormal stable segment, which constitutes the abnormal thickness formula segment D2. The abnormal segments include, but are not limited to, unstable thickness regions caused by factors such as measurement noise, equipment vibration, and steel material vibration. D1 and D2 in the diagram are only used to represent the schematic range of different continuous intervals and do not limit the specific length or proportion. Different line types or colors are used to distinguish effective stable segments, transition segments, and abnormal segments in the diagram, and the length and thickness units are clearly marked on the coordinate axes.

[0174] refer to Figure 3 The diagram illustrates a formula segment matching method based on process sequence constraints, as provided in an embodiment of the present invention. It demonstrates the principle of formula segment matching based on process sequence constraints. The diagram uses a top-bottom contrast layout. The upper part shows the process sequence of preset thickness-stable segments in the formula, where P1, P2, P3, and P4 represent the thickness type sequence of the standard formula segments (e.g., thin → thick → thin). The lower part shows multiple effective stable segments extracted from actual measured thickness data, where S1, S2, S3, and S4 represent the actual measured segments. Matching is performed using process sequence constraints (e.g., thickness segment type sequence). D1 represents the continuous segment interval corresponding to the formula process sequence in the actual measured data. Arrows indicate the process sequence direction, and matching lines connect the corresponding segments, visually demonstrating the segment consistency judgment process.

[0175] refer to Figure 4 The diagram illustrates the extraction method for effective stable segment feature thickness points according to an embodiment of the present invention. The horizontal axis represents the length of the stable segment, and the vertical axis represents the thickness. On the stable segment S1 (or S2, S3, S4), feature points at the segment start point, segment end point, and preset proportional positions within the segment (such as the segment midpoint, quarter point, and three-quarter point) are clearly marked. Different symbols (such as dots, triangles, and squares) distinguish different feature point types, and the corresponding thickness values ​​(such as 3.1mm, 3.0mm, etc.) are marked next to the feature points. Multiple feature points are extracted to form a feature thickness point set, which is used for subsequent weighted comprehensive deviation calculation.

[0176] refer to Figure 5The diagram illustrates the principle of full-length thickness tolerance detection based on formula identification, as provided in this embodiment of the invention. The horizontal axis represents the length of the steel coil, and the vertical axis represents the thickness. The continuous thickness curve is marked with a standard thickness line, upper tolerance limit, and lower tolerance limit (distinguished by different line types, such as solid lines, dashed lines, and dotted lines). When the actual thickness data exceeds the allowable deviation range, the abnormal area is marked with a shaded area or a different color. The diagram marks abnormal areas where multiple consecutive sampling points exceed the tolerance range, as well as abnormal points where a single sampling point exceeds the safety threshold. The coordinate axes clearly indicate the units of length and thickness, and key parameters such as "allowable deviation range" and "safety threshold" are marked in appropriate positions within the diagram.

[0177] In some embodiments, the method further includes result application, specifically including:

[0178] The identified formula and the precise start and end positions of each thickness segment are sent to the downstream processing equipment through a preset industrial communication interface, driving the downstream processing equipment to automatically call the matching processing parameters based on the received formula information and start and end position information.

[0179] The type and communication protocol of the industrial communication interface are determined based on the interface specifications of the downstream processing equipment. The processing parameters include at least one of the cutting parameters, straightening parameters, or winding parameters corresponding to the thickness range of the formula. The matching relationship of the processing parameters is pre-stored in the control system of the downstream processing equipment.

[0180] In this embodiment, by automatically sending the formula identification results and thickness segment location information to the downstream processing equipment and driving the equipment to automatically call the matching processing parameters, a fully automated closed loop from formula identification to processing execution is realized. This avoids the operational errors and inefficiencies caused by traditional manual parameter configuration, significantly improves the automation level and processing accuracy of the longitudinal variable thickness steel coil production line, while reducing manual intervention and improving production efficiency and product quality consistency.

[0181] In some embodiments, the method further includes manual-assisted identification, specifically including:

[0182] After step S2 or step S3 is executed, the current recognition result is displayed through the human-computer interaction interface, allowing users to manually adjust or specify the effective stable segment interval, and the formula recognition and thickness detection process is retried based on the manually adjusted interval. The system records the manual correction results and uses them to optimize the parameters of the automatic recognition model.

[0183] The human-computer interaction interface includes a graphical display and interactive operation controls. The optimization of the automatic recognition model includes adjusting feature extraction parameters or recognition algorithm parameters based on manual correction results. The optimization process is implemented through machine learning algorithms or parameter adaptive adjustment algorithms.

[0184] In this embodiment, by setting up a manual assisted identification step, a manual correction mechanism is provided on the basis of automatic identification. This can effectively handle complex working conditions or abnormal situations that are difficult for the automatic identification system to accurately identify, thereby improving the system's adaptability and robustness. At the same time, by recording the results of manual correction and using them to optimize the automatic identification model, the system achieves self-learning and continuous improvement, gradually reducing the need for manual intervention and improving the overall intelligence level and long-term operational stability of the identification system.

[0185] Based on the same inventive concept, embodiments of the present invention also provide an online formula identification and detection system for longitudinally variable thickness steel coils for implementing any of the methods described above, comprising:

[0186] The sensing and acquisition unit includes a thickness sensor and a displacement sensor, which are used to synchronously acquire continuous thickness data and position information of the steel coil;

[0187] The data processing unit is configured to perform the data acquisition and preprocessing steps and the stationary segment extraction and filtering steps;

[0188] The recipe identification unit is configured to perform the recipe identification step based on process sequence constraints;

[0189] The quality inspection unit is configured to perform the quality inspection and result output steps;

[0190] The human-machine interface and control unit is used to provide an operating interface, display results, receive instructions, and realize communication with downstream processing equipment.

[0191] In some embodiments, reference Figure 6The diagram shown is a structural block diagram of an online formula identification and detection system for longitudinally variable thickness steel coils provided by an embodiment of the present invention. It illustrates the structure and data flow of the online formula identification and detection system implementing the method of the present invention. The system adopts a standard block diagram format, with each module represented by a box and data flows connected by arrows. The system includes: a thickness sensor, a displacement sensor, a data acquisition module, a data processing module, a formula identification module, a thickness detection module, a human-machine interaction module, and a downstream equipment interface module. The data processing and formula identification module is used to preprocess the acquired thickness data, extract stable segments, and complete the consistency judgment and identification of formula segments based on process sequence constraints. The data flow arrows are labeled with the data type being transmitted (such as "thickness data", "position information", "identification result", etc.). The system boundary box is labeled "Online Detection System," distinguishing between hardware modules and software / control modules.

[0192] Based on the same inventive concept, embodiments of the present invention also provide an electronic device. Figure 7 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Figure 7 As shown, an embodiment of the present invention provides an electronic device including: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement an online formula identification and quality inspection method for longitudinally variable thickness steel coils as described in any of the above embodiments; the one or more I / O interfaces 103 are connected between the processor and the memory, configured to enable information interaction between the processor and the memory.

[0193] The processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read / write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus).

[0194] In some embodiments, the processor 101, memory 102, and I / O interface 103 are interconnected via bus 104, and thus connected to other components of the computing device.

[0195] In some embodiments, the one or more processors 101 include a field-programmable gate array.

[0196] This invention also provides a computer-readable medium. The computer-readable medium stores a computer program, which, when executed by a processor, implements an online formula identification and quality inspection method for longitudinally variable thickness steel coils as described in any of the above embodiments. The computer-readable storage medium can be volatile or non-volatile.

[0197] This invention also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code. When the computer-readable code is run in the processor of an electronic device, the processor in the electronic device executes any of the above-described methods for online formula identification and quality inspection of longitudinally variable thickness steel coils.

[0198] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0199] As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable program instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, it is known to those skilled in the art that communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0200] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0201] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0202] The computer program product described herein can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0203] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0204] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0205] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0206] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0207] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A method for online formulation identification and quality inspection of longitudinally variable thickness steel coils, characterized in that, Includes the following steps: S1. Data Acquisition and Preprocessing: Simultaneously acquire continuous thickness data and corresponding position information of the steel coil along its length direction, and perform noise reduction and smoothing processing on the thickness data to generate a continuous thickness-length curve. S2. Extraction and screening of stable segments: On the thickness-length curve, continuous segments with thickness fluctuations less than a preset threshold are identified as candidate stable segments, and effective stable segments are screened from the candidate stable segments according to preset segment length rationality rules and thickness change physical continuity rules. S3. Recipe identification based on process sequence constraints: Using pre-stored recipe knowledge, which includes at least the fixed order of each thickness segment in the target recipe as a process sequence constraint, the arrangement order of the effective stable segments in the length direction of the steel coil is matched with the process sequence constraint to determine the set of continuous segments corresponding to the target recipe, and the recipe identification is completed based on the set of continuous segments. S4. Quality Inspection and Result Output: Based on the identified formula, perform full-length thickness tolerance inspection on the steel coil and output the inspection results and alarm information.

2. The method according to claim 1, characterized in that, In step S2, the preset segment length rationality rule is: to remove candidate stable segments whose length is less than the minimum formula segment length by a first preset ratio or whose length is greater than the maximum formula segment length by a second preset ratio; the thickness change physical continuity rule is: to remove candidate stable segments whose thickness change between adjacent sampling points within a segment exceeds the process allowable abrupt change threshold; wherein, the first preset ratio and the second preset ratio are determined based on the process parameters of the target formula, and the process allowable abrupt change threshold is determined based on the physical limits of the steel coil rolling process.

3. The method according to claim 2, characterized in that, In step S2, after applying the paragraph length rationality rule and the thickness change physical continuity rule, the formula consistency prediction is also included. Specifically, this includes: calculating the average thickness of the candidate stable segments after being screened by the first two rules, comparing the average thickness with the thickness of each standard formula segment in the formula knowledge base, and removing candidate stable segments whose average thickness does not fall within the allowable deviation range of any standard formula segment thickness to obtain the final effective stable segments. The formula knowledge base and the formula knowledge in S3 are from the same data source, and the allowable deviation range of thickness is determined based on the process accuracy requirements of the target formula.

4. The method according to claim 1, characterized in that, In step S3, the recipe identification based on process sequence constraints specifically includes: S31. Coarse positioning: Based on the arrangement order of the effective stable segments in the length direction of the steel coil, obtain the process sequence constraints of the target formula from the formula knowledge, find the continuous segment sequence that matches the process sequence constraints in the thickness segment type order, and preliminarily determine the possible position range of the target formula on the steel coil. S32, Fine matching: For each possible position interval determined in S31, the continuous paragraph sequence corresponding to each effective stable segment is extracted, including the paragraph start point, paragraph end point, and at least one preset proportion position within the segment, forming a set of feature thickness points for the stable segment. S33. Formula Confirmation: Based on the set of characteristic thickness points obtained in S32, a weighted comprehensive deviation calculation is performed between the characteristic thickness points of the target formula standard segment in the formula knowledge, and the formula with the smallest weighted comprehensive deviation is determined as the final identification result; wherein, the weighted comprehensive deviation is calculated by assigning weights to the deviation of each characteristic thickness point and then performing a weighted summation.

5. The method according to claim 4, characterized in that, In step S32, at least one preset proportional position within the segment is selected from at least one of the segment midpoint, quarter point, and three-quarter point; wherein, the selection of the preset proportional position is determined based on the thickness distribution characteristics of the steel coil rolling process, the thickness distribution characteristics including the thickness uniformity requirements and detection accuracy requirements during the rolling process.

6. The method according to claim 4, characterized in that, In step S33, the weighted comprehensive deviation is calculated by assigning weights to the deviations of each feature thickness point and then performing a weighted summation. The weights corresponding to the start and end points of the paragraph are higher than the weights corresponding to other feature points within the paragraph. The weight allocation is determined based on the importance of the feature thickness points in recipe recognition, and the importance is determined by the contribution of the feature thickness points to the positioning of the recipe segment boundary.

7. The method according to claim 1, characterized in that, In step S4, the full-length thickness tolerance detection includes: comparing the standard thickness and allowable deviation range corresponding to the identified formula with the actual thickness data of the steel coil point by point; when any of the following conditions occur, the thickness at that location is determined to be abnormal and marked: (1) The actual thickness of multiple consecutive sampling points exceeds the allowable deviation range; (2) The actual thickness of a single sampling point exceeds the allowable deviation range by more than a preset safety threshold; The number of consecutive sampling points is determined based on process quality requirements and detection accuracy, and the preset safety threshold is determined based on the physical limits of the steel coil rolling process and the accuracy of the detection system.

8. The method according to claim 1, characterized in that, The method also includes result application, specifically including: The identified formula and the start and end position information of each thickness segment are sent to the downstream processing equipment through a preset industrial communication interface, which then drives the downstream processing equipment to automatically call the matching processing parameters based on the received formula information and start and end position information.

9. The method according to claim 1, characterized in that, The method also includes manual-assisted identification, specifically including: After step S2 or step S3 is executed, the current recognition result is displayed through the human-computer interaction interface, allowing users to manually adjust or specify the effective stable segment interval, and the formula recognition and thickness detection process is retried based on the manually adjusted interval. The system records the manual correction results and uses them to optimize the parameters of the automatic recognition model.

10. A longitudinally variable thickness steel coil online formula identification and detection system for implementing the method of any one of claims 1-9, characterized in that, include: The sensing and acquisition unit includes a thickness sensor and a displacement sensor, which are used to synchronously acquire continuous thickness data and position information of the steel coil; The data processing unit is configured to perform the data acquisition and preprocessing steps and the stationary segment extraction and filtering steps; The recipe identification unit is configured to perform the recipe identification step based on process sequence constraints; The quality inspection unit is configured to perform the quality inspection and result output steps; The human-machine interface and control unit is used to provide an operating interface, display results, receive instructions, and realize communication with downstream processing equipment.

Images