Steel coil cold rolling process control method, device, equipment, medium and product

Abstract

This application discloses a method, apparatus, equipment, medium, and product for controlling the cold rolling process of steel coils. The method involves acquiring a historical scratch dataset for the corresponding steel grade of the steel coil. This dataset includes multiple standard historical process parameter combinations and corresponding scratch risk prediction values. Based on these combinations and the predicted scratch risks, and with the constraints of meeting preset conditions for process parameter combinations and production parameters, and with the objective of minimizing a preset objective function, a target process parameter combination for the steel coil is calculated. The associated cold rolling unit is then controlled to perform cold rolling operations on the steel coil according to the target process parameter combination, ensuring that the scratch degree of the steel coil meets the preset target level. According to the embodiments of this application, the generation of coil scratch quality defects such as loose coiling, core pulling, and misalignment can be effectively suppressed, ultimately improving the quality of the steel coil product.

Description

Technical Field

[0001] This application belongs to the field of metal processing technology, and in particular relates to a method, apparatus, equipment, medium and product for controlling the cold rolling process of steel coils. Background Technology

[0002] The annealing mill is a core piece of equipment in cold rolling production, used to soften strip steel after multiple cold working processes. As an intermediate link connecting hot rolling, cold rolling, leveling, and other rolling processes, the annealing mill has long been responsible for the tempering of intermediate materials for high-strength steel, playing a key role in adjusting the microstructure and mechanical properties of the strip steel and contributing to the high-quality production of high-strength steel grades.

[0003] However, high-quality strip and sheet are not only about their shape and thickness requirements, but also about the surface quality of their cold-rolled coils. Due to the characteristics of the equipment and the process, the annealing unit does not include a coiling mechanism. The coil shape and coiling internal stress state of the annealed coil are inherited from the hot rolling and cold rolling processes, and are affected by the thermal expansion phenomenon during heat treatment, which is ultimately transferred to the subsequent cold rolling or leveling processes.

[0004] In this situation, if the interlayer stress field formed by coiling in the upstream cold rolling process, the residual thermal stress field formed in the heat treatment process, and the external force field applied during the hoisting process are unbalanced or uncoordinated, it is very easy to cause problems such as loose coiling, core pulling, and layering of the strip steel, which will eventually result in scratches and quality defects in the steel coil, seriously affecting product quality. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, medium, and product for controlling the cold rolling process of steel coils. It can effectively improve the problems of uneven distribution and incoordination of interlayer stress, residual thermal stress during annealing, and external force field during hoisting in the upstream cold rolling process. It can suppress the generation of quality defects such as loose coils, core pulling, and misaligned layers from the root, and ultimately improve the quality of steel coil products.

[0006] In a first aspect, embodiments of this application provide a method for controlling the cold rolling process of steel coils, the method comprising: Obtain the historical scratch dataset for the corresponding grade of steel coil. The historical scratch dataset includes multiple standard historical process parameter combinations and scratch risk prediction values ​​corresponding to each standard historical process parameter combination. Based on the multiple standard historical process parameter combinations and the predicted scratch risk values ​​corresponding to each standard historical process parameter combination, with the process parameter combination and production parameters meeting preset constraints as constraints, and with minimizing the preset objective function as the objective, the target process parameter combination of the steel coil is calculated. The target process parameter combination includes target cold continuous rolling coiling tension, target rotation speed, target heating rate, target isothermal temperature, target isothermal time, target cooling time, target internal temperature of the steel coil, target hoisting time, target maximum hoisting speed, target leveling uncoiling tension, and target leveling coiling tension. The cold rolling process unit of the steel coil is controlled to perform cold rolling operation on the steel coil according to the target process parameter combination, and the degree of scratches on the steel coil meets the preset target degree.

[0007] Secondly, embodiments of this application provide a steel coil cold rolling process control device, the device comprising: The first acquisition module is used to acquire the historical scratch dataset of the corresponding steel grade of the steel coil. The historical scratch dataset includes multiple standard historical process parameter combinations and scratch risk prediction values ​​corresponding to each standard historical process parameter combination. The calculation module is used to calculate the target process parameter combination of the steel coil based on the multiple standard historical process parameter combinations and the abrasion risk prediction value corresponding to each standard historical process parameter combination, with the process parameter combination and production parameters meeting the preset constraints as constraints, and with the goal of minimizing the preset objective function. The target process parameter combination includes target cold continuous rolling coiling tension, target rotation speed, target heating rate, target isothermal temperature, target isothermal time, target cooling time, target internal temperature of the steel coil, target hoisting time, target maximum hoisting speed, target leveling uncoiling tension, and target leveling coiling tension. The control module is used to control the cold rolling process associated unit of the steel coil to perform cold rolling operation on the steel coil according to the target process parameter combination, and the degree of scratches on the steel coil meets the preset target degree.

[0008] Thirdly, embodiments of this application provide an electronic device, the device including: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements the steel coil cold rolling process control method as described above.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the cold rolling process control method for steel coils as described above.

[0010] Fifthly, embodiments of this application provide a computer program product, wherein when the instructions in the computer program product are executed by the processor of an electronic device, the electronic device performs the steel coil cold rolling process control method as described in any of the above claims.

[0011] The steel coil cold rolling process control method, apparatus, equipment, medium, and product of this application embodiment can acquire historical scratch datasets for steel grades corresponding to the steel coils. The historical scratch datasets include multiple standard historical process parameter combinations and scratch risk prediction values ​​corresponding to each standard historical process parameter combination. Based on the multiple standard historical process parameter combinations and the scratch risk prediction values ​​corresponding to each standard historical process parameter combination, and with the process parameter combinations and production parameters meeting preset constraints as constraints, and minimizing a preset objective function as the objective, the target process parameter combination for the steel coil is calculated. The target process parameter combination includes target cold continuous rolling coiling tension, target rotation speed, target heating rate, target isothermal temperature, target isothermal time, target cooling time, target internal temperature of the steel coil, target hoisting time, target maximum hoisting speed, target leveling uncoiling tension, and target leveling coiling tension. The cold rolling process associated unit of the steel coil is controlled to perform cold rolling operations on the steel coil according to the target process parameter combination, and the degree of scratches on the steel coil meets the preset target degree. Thus, in this embodiment of the application, a dataset is constructed based on historical process parameters and corresponding predicted values ​​of scratch risk. The objective function related to scratch risk is minimized as the optimization objective. Combined with production constraints, the optimal combination of target process parameters for the entire cold rolling process is obtained and systematically optimized. This effectively improves the problems of uneven distribution and incoordination of upstream cold rolling interlayer stress, residual thermal stress from annealing, and external force field during hoisting. It fundamentally suppresses the generation of quality defects such as loose coils, core pulling, and misalignment of coils, and ultimately improves the quality of steel coil products. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a schematic flowchart of a cold rolling process control method for steel coils provided in an embodiment of this application; Figure 2 This is a schematic flowchart of another cold rolling process control method for steel coils provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of the cold rolling process control device for steel coils provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0014] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0015] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0016] The annealing mill is a core piece of equipment in cold rolling production, used to soften strip steel after multiple cold working processes. As an intermediate link connecting hot rolling, cold rolling, leveling, and other rolling processes, the annealing mill has long been responsible for the tempering of intermediate materials for high-strength steel, playing a key role in adjusting the microstructure and mechanical properties of the strip steel and contributing to the high-quality production of high-strength steel grades.

[0017] However, high-quality strip and sheet are not only about their shape and thickness requirements, but also about the surface quality of their cold-rolled coils. Due to the characteristics of the equipment and the process, the annealing unit does not include a coiling mechanism. The coil shape and coiling internal stress state of the annealed coil are inherited from the hot rolling and cold rolling processes, and are affected by the thermal expansion phenomenon during heat treatment, which is ultimately transferred to the subsequent cold rolling or leveling processes.

[0018] In this situation, if the interlayer stress field formed by coiling in the upstream cold rolling process, the residual thermal stress field formed in the heat treatment process, and the external force field applied during the hoisting process are unbalanced or uncoordinated, it is very easy to cause problems such as loose coiling, core pulling, and layering of the strip steel, which will eventually result in scratches and quality defects in the steel coil, seriously affecting product quality.

[0019] The acquisition, storage, use, and processing of data in this application comply with relevant national laws and regulations. It should be noted that certain software, components, models, and other existing industry solutions may be mentioned in the embodiments of this application. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0020] To address the problems of the prior art, embodiments of this application provide a method, apparatus, equipment, medium, and product for controlling the cold rolling process of steel coils. The method for controlling the cold rolling process of steel coils provided in this application embodiment will be described first below.

[0021] Figure 1 A schematic flowchart of a steel coil cold rolling process control method according to an embodiment of this application is shown. Figure 1 As shown, a method for controlling the cold rolling process of steel coils may include the following steps S101 to S103: S101. Obtain the historical scratch dataset for the corresponding grade of steel coil. The historical scratch dataset includes multiple standard historical process parameter combinations and the scratch risk prediction value corresponding to each standard historical process parameter combination. S102. Based on multiple standard historical process parameter combinations and the predicted scratch risk values ​​corresponding to each standard historical process parameter combination, with the process parameter combination and production parameters meeting the preset constraints as constraints, and with minimizing the preset objective function as the objective, the target process parameter combination of the steel coil is calculated. The target process parameter combination includes the target cold continuous rolling coiling tension, target rotation speed, target heating rate, target constant temperature, target constant temperature time, target cooling time, target internal temperature of the steel coil, target hoisting time, target maximum hoisting speed, target leveling uncoiling tension, and target leveling coiling tension. S103. The cold rolling process of the steel coil is controlled by the associated unit to perform cold rolling operation on the steel coil according to the target process parameter combination, and the degree of scratches on the steel coil meets the preset target degree.

[0022] The cold rolling process control method for steel coils in this application can acquire a historical scratch dataset for the corresponding steel grade of the steel coil. The historical scratch dataset includes multiple standard historical process parameter combinations and scratch risk prediction values ​​corresponding to each standard historical process parameter combination. Based on the multiple standard historical process parameter combinations and scratch risk prediction values ​​corresponding to each standard historical process parameter combination, and with the process parameter combinations and production parameters meeting preset constraints as constraints, and minimizing a preset objective function as the objective, a target process parameter combination for the steel coil is calculated. The target process parameter combination includes target cold continuous rolling coiling tension, target rotation speed, target heating rate, target isothermal temperature, target isothermal time, target cooling time, target internal temperature of the steel coil, target hoisting time, target maximum hoisting speed, target leveling uncoiling tension, and target leveling coiling tension. The cold rolling process associated unit of the steel coil is controlled to perform cold rolling operations on the steel coil according to the target process parameter combination, and the degree of scratches on the steel coil meets the preset target degree. Thus, in this embodiment of the application, a dataset is constructed based on historical process parameters and corresponding predicted values ​​of scratch risk. The objective function related to scratch risk is minimized as the optimization objective. Combined with production constraints, the optimal combination of target process parameters for the entire cold rolling process is obtained and systematically optimized. This effectively improves the problems of uneven distribution and incoordination of upstream cold rolling interlayer stress, residual thermal stress from annealing, and external force field during hoisting. It fundamentally suppresses the generation of quality defects such as loose coils, core pulling, and misalignment of coils, and ultimately improves the quality of steel coil products.

[0023] In S101, the aforementioned historical abrasion dataset may include multiple standard historical process parameter combinations and the abrasion risk prediction value corresponding to each standard historical process parameter combination.

[0024] In some embodiments of this application, obtaining a historical scratch dataset for a steel coil of a corresponding grade can, for example, involve obtaining multiple historical process parameter combinations for the steel coil of the corresponding grade and the scratch defect quality level corresponding to each historical process parameter combination; preprocessing the multiple historical process parameter combinations to obtain multiple standard historical process parameter combinations, the preprocessing including at least one of rejection processing, normalization processing, and deduplication processing; determining the scratch risk prediction value corresponding to each standard historical process parameter combination based on the multiple standard historical process parameter combinations and the scratch defect quality level corresponding to each standard historical process parameter combination, the scratch risk prediction value being used to indicate the degree of scratches on the steel coil under the standard historical process parameter combination; and constructing a historical scratch dataset based on the multiple standard historical process parameter combinations and the scratch risk prediction values ​​corresponding to each standard historical process parameter combination.

[0025] In S102, the aforementioned combination of target process parameters may include target cold rolling coiling tension, target rotational speed, target heating rate, target isothermal temperature, target isothermal time, target cooling time, target internal temperature of the steel coil, target hoisting time, target maximum hoisting speed, target leveling uncoiling tension, and target leveling coiling tension. Specifically, the cold rolling coiling tension set on the cold rolling Carousel coiler... c Rotation speed oh Annealing temperature-time curve parameters: including heating rate R h Constant temperature T h Constant temperature time t h Cooling time t c internal temperature of steel coil t in Lifting parameters: Lifting time t l Maximum hoisting speed V l Flattening tension parameters: Flattening unwinding tension un With flat winding tension re .

[0026] In some embodiments of this application, the above objective function can be as follows:

[0027] in, F(X) Let be the objective function. X For process parameter combinations, f scratch ( X () represents the predicted abrasion risk value corresponding to the combination of process parameters. T cycle ( X The total annealing cycle time is calculated based on the combination of process parameters. E ( X To estimate energy consumption, T base As the baseline annealing cycle, E base As a baseline energy consumption, a、b、c These are the weighting coefficients, and a+b+c=1.

[0028] In some embodiments of this application, the aforementioned constraints may include: process constraint sub-conditions and production constraint sub-conditions. The process constraint sub-conditions can be used to constrain the operating range of each process parameter in the process parameter combination, which may include cold continuous rolling coiling tension, rotational speed, heating rate, isothermal temperature, isothermal time, cooling time, internal temperature of the steel coil, hoisting time, maximum hoisting speed, leveling uncoiling tension, and leveling coiling tension. The production constraint sub-conditions can be used to constrain the operating range of production parameters, including the total annealing cycle time.

[0029] In some embodiments of this application, based on multiple standard historical process parameter combinations and the predicted scratch risk values ​​corresponding to each standard historical process parameter combination, and with the constraint that the process parameter combination and production parameters meet preset constraints, and with the objective of minimizing a preset objective function, the target process parameter combination for the steel coil is calculated. For example, this can be achieved by initially generating N randomly generated particles, each particle corresponding to a different process parameter combination, where N is a positive integer greater than or equal to 1, and the process parameter combination and production parameters corresponding to the N particles meet preset constraints. The process parameter combination of each particle is then substituted into the preset objective function to calculate the fitness of each particle. The fitness of the N particles is compared to determine the global optimal fitness, which is the minimum fitness. With the constraint that the process parameter combination and production parameters meet the constraints, the process parameter combination of the N particles is iteratively updated, and with the objective of minimizing the objective function, the fitness of each particle is iteratively calculated and the global optimal fitness is updated. When the number of iterations reaches a preset number, and / or the change in the global optimal fitness within several consecutive generations is less than a preset threshold, the process parameter combination corresponding to the global optimal fitness is determined as the target process parameter combination for the steel coil.

[0030] In S103, the aforementioned cold rolling process associated units, exemplarily, may include: a cold continuous rolling Carrousel coiler, a bell-type annealing unit, coil hoisting equipment, and a leveling unit. It should be noted that the cold rolling process associated units are not limited to the above examples and can be configured according to actual application scenarios; no specific limitations are made here.

[0031] The aforementioned target level could mean that the steel coil shows no scratches or only very slight scratches. Of course, in this embodiment, the target level can be set according to actual application needs, and is not specifically limited here.

[0032] In some embodiments of this application, the cold rolling process associated unit of the steel coil is controlled to perform cold rolling operation on the steel coil according to the target process parameter combination. For example, if the scratch defect quality level corresponding to the target process parameter combination does not meet the preset target level, the target process parameter combination may be adjusted until the scratch defect quality level corresponding to the target process parameter combination meets the target level; if the scratch defect quality level corresponding to the target process parameter combination meets the scratch risk prediction value range corresponding to the target level, the cold rolling process associated unit of the steel coil is controlled to perform cold rolling operation on the steel coil according to the target process parameter combination. Alternatively, it could involve controlling the cold rolling process of the steel coil to perform cold rolling operations on the steel coil according to the target process parameter combination; acquiring real-time monitoring data of the cold rolling process associated units; compensating and adjusting sensitive process parameters in the target process parameter combination when the deviation between the real-time monitoring data and the target process parameter combination exceeds a preset deviation threshold; and storing the steel coil grade, target process parameter combination, deviation threshold, sensitive process parameters, and compensation settings for each sensitive process parameter in the target process parameter combination.

[0033] In some embodiments, the above-described S101 may specifically include: Obtain multiple historical process parameter combinations for the corresponding steel grade of the steel coil and the scratch defect quality level corresponding to each historical process parameter combination; Multiple historical process parameter combinations are preprocessed to obtain multiple standard historical process parameter combinations. The preprocessing includes at least one of rejection processing, normalization processing and deduplication processing. Based on multiple standard historical process parameter combinations and the scratch defect quality level corresponding to each standard historical process parameter combination, the scratch risk prediction value corresponding to each standard historical process parameter combination is determined. The scratch risk prediction value is used to indicate the degree of scratches on the steel coil under the standard historical process parameter combination. A historical abrasion dataset is constructed based on multiple standard historical process parameter combinations and the corresponding abrasion risk prediction values ​​for each standard historical process parameter combination.

[0034] The above-mentioned combination of multiple historical process parameters may include: historical production data of cold rolling mill and historical production data of bell-type annealing mill.

[0035] Historical production data for cold rolling mills may include: historical data of the cold continuous rolling Carousel coiler and historical data of actual exit tension; among which, historical data of the cold continuous rolling Carousel coiler may include: cold continuous rolling coil tension and rotation speed.

[0036] Historical production data for the bell-type annealing unit may include: bell-type unit temperature data, hoisting data, and leveling tension data; among which, the bell-type unit temperature data may include: bell-type annealing unit temperature data, bell-type annealing unit time data, and internal temperature data of the steel coil; the bell-type annealing unit temperature data may include: temperature rise rate data during the heating stage, temperature data during the isothermal stage, and temperature fall rate data during the cooling stage; the bell-type annealing unit time data may include: heating time data, isothermal time data, and cooling time data; hoisting data may include: hoisting speed and time data; and leveling tension data may include leveling uncoiling tension and leveling coiling tension data.

[0037] The aforementioned scratch defect quality levels can be determined through visual inspection and simple measuring tools to ascertain the depth and area of ​​scratches, as well as the proportion of scratches across the width of the steel coil. For example, they can be categorized into four scratch defect quality levels: A, B, C, and D. A corresponds to no scratches or very minor scratches; B corresponds to moderate scratches; C corresponds to severe scratches; and D corresponds to extremely severe scratches. It should be noted that the scratch defect quality levels in this application are not limited to the four levels A, B, C, and D mentioned above, and can be set according to actual application requirements; no specific limitations are made here.

[0038] The preprocessing described above may include at least one of the following: removal processing, normalization processing, and deduplication processing. Removal processing may involve removing missing values ​​and / or outliers.

[0039] In some embodiments of this application, multiple historical process parameter combinations are preprocessed to obtain multiple standard historical process parameter combinations. For example, missing values ​​and outliers in multiple sample historical process parameter combinations may be identified, and samples containing missing values ​​and / or outliers may be deleted. Then, the data in the sample historical process parameter combinations are standardized to make each variable within the same dimension, which is convenient for optimization algorithm processing. Finally, the multiple processed samples are deduplicated to obtain multiple standard historical process parameter combinations.

[0040] The above-mentioned scratch risk prediction values ​​can be used to indicate the degree of scratches that may occur on steel coils under standard historical process parameter combinations.

[0041] In this embodiment, historical process parameter combinations and scratch defect quality levels for the corresponding steel grades are first collected. Data preprocessing, including rejection, normalization, and deduplication, effectively eliminates invalid, duplicate, and dimensionally inconsistent data interference, improving the reliability of the process data. Then, the discrete scratch defect quality levels are converted into quantified scratch risk prediction values, achieving a digital description of the severity of scratches, ultimately forming a standardized and valid historical scratch dataset. A high-quality dataset ensures accurate and reliable subsequent process parameter optimization results, facilitating precise control of the stress field in each process, reducing coil shape anomalies and surface scratches, and guaranteeing the quality of strip steel products.

[0042] In some embodiments, determining the predicted scratch risk value for each standard historical process parameter combination based on multiple standard historical process parameter combinations and the scratch defect quality level corresponding to each standard historical process parameter combination may specifically include: Based on multiple standard historical process parameter combinations and the scratch defect quality levels corresponding to each standard historical process parameter combination, a preset machine learning model is trained to obtain a scratch risk prediction model. Different scratch defect quality levels correspond to different scratch risk prediction value ranges. Input each combination of standard historical process parameters into the scratch risk prediction model, and output the scratch risk prediction value corresponding to each combination of standard historical process parameters.

[0043] The different scratch defect quality levels mentioned above correspond to different scratch risk prediction value ranges. For example, scratch defect quality level A corresponds to the scratch risk prediction value range of [0, 0.25], scratch defect quality level B corresponds to the scratch risk prediction value range of [0.25, 0.5], scratch defect quality level C corresponds to the scratch risk prediction value range of [0.5, 0.75], and scratch defect quality level D corresponds to the scratch risk prediction value range of [0.75, 1].

[0044] In some embodiments of this application, a pre-set machine learning model is trained based on multiple standard historical process parameter combinations and the scratch defect quality level corresponding to each standard historical process parameter combination to obtain a scratch risk prediction model. For example, the pre-set machine learning model can be trained by regression fitting based on multiple standard historical process parameter combinations and the scratch defect quality level corresponding to each standard historical process parameter combination to obtain a scratch risk prediction model.

[0045] In this embodiment, a dedicated scratch risk prediction model is trained using standardized process parameters and corresponding scratch defect quality levels, transforming the traditional discrete qualitative quality level evaluation into a continuous quantitative scratch risk prediction value. By using machine learning to uncover the implicit coupling relationship between multi-process parameters and scratch defects, the severity of steel coil scratches under different production conditions can be precisely characterized, solving the problems of strong subjectivity and inability to quantify risk evaluation in traditional manual grading. This provides a high-precision and continuous evaluation basis for subsequent process parameter optimization, effectively improving the matching accuracy of process parameters throughout the entire process, thereby stabilizing the coil shape and reducing interlayer scratch defects.

[0046] In some embodiments, the above-described S102 may specifically include: Initialize and randomly generate N particles. Different particles correspond to different combinations of process parameters. N is a positive integer greater than or equal to 1. The combination of process parameters and production parameters corresponding to the N particles satisfy the preset constraints. The process parameters of each particle are combined and input into the preset objective function to calculate the fitness of each particle. Compare the fitness of N particles to determine the global optimal fitness, which is the minimum fitness. With the constraints of process parameter combination and production parameter satisfaction, the process parameter combination of N particles is iteratively updated, and the fitness of each particle is iteratively calculated and the global optimal fitness is updated with the goal of minimizing the objective function. When the number of iterations reaches a preset number, and / or the change in the global optimal fitness is less than a preset threshold within several consecutive generations, the combination of process parameters corresponding to the global optimal fitness is determined as the target combination of process parameters for the steel coil.

[0047] In some embodiments of this application, N particles are randomly generated during initialization. For example, the particle swarm size may be set during initialization. N The position vector of each particle Represents a set of decision variables X The value of, dimension d The number of decision variables is equal to the number of particles; the position of each particle is randomly initialized. and speed Position initialization must meet the above constraints; set algorithm parameters: initial inertia weights. oh start and termination inertia weight oh end Individual learning factor c 1 Social learning factor c 2 Maximum number of iterationsK max Convergence threshold e .

[0048] In some embodiments of this application, the process parameters of each particle are combined and input into a preset objective function to calculate the fitness of each particle. For example, the fitness evaluation can be performed for each particle. i to position Decode the specific combination of process parameters and substitute them into the objective function. F ( X Calculate its fitness value F i .

[0049] In some embodiments of this application, the fitness of N particles is compared to determine the global optimal fitness. For example, this can be achieved by comparing the current fitness of each particle with its historical best fitness. pbest i ,renew pbest i and individual optimal position Compare the individual optimal fitness of all particles and update the global optimal fitness. gbest i and global optimal position .

[0050] In some embodiments of this application, constraints are imposed on the combination of process parameters and the satisfaction of production parameters. For example, this can be boundary and constraint processing: the updated particle position is checked for out-of-bounds errors; if it exceeds the decision variable boundary, it is pulled back to the boundary. For positions that violate hard constraints such as product performance or equipment safety, a very large penalty fitness value is assigned, causing them to be eliminated in subsequent iterations.

[0051] In some embodiments of this application, the process parameter combination of N particles is iteratively updated. For example, the particle velocity and position can be updated according to the following formula:

[0052] .

[0053] in, r 1 , r 2 The random number is within the interval [0,1], with inertial weight. oh ( t A linear decreasing strategy is adopted: .

[0054] In some embodiments of this application, when the number of iterations reaches a preset number, and / or the change in the global optimal fitness within multiple consecutive generations is less than a preset threshold, the combination of process parameters corresponding to the global optimal fitness is determined as the target combination of process parameters for the steel coil. For example, this could be when the maximum number of iterations... K max or global optimal fitness gbest i In continuous The change within a generation is less than a preset threshold. When the algorithm converges, the iteration terminates, and the globally optimal position at that point is output. e As the target combination of process parameters.

[0055] In this embodiment, a particle swarm optimization algorithm is used to optimize process parameters. Particles are initialized and iteratively updated within preset constraints, and the optimal combination of process parameters is selected based on minimizing the objective function. This algorithm boasts high search efficiency and strong global optimization capabilities, enabling rapid matching of coordinated parameters across all processes, including cold rolling, annealing, hoisting, and leveling, effectively balancing the stress fields of each process. The resulting target process parameter set is well-suited to actual production boundary conditions, exhibiting high parameter matching accuracy. This effectively suppresses defects such as loose coiling, core pulling, and misalignment, further reducing the risk of scratches on the steel coil.

[0056] As one implementation of this application, in order to further ensure the stability of the steel coil surface quality and improve the production qualification rate, before S103 above, the above method may further include: Input the target process parameter combination into the preset scratch risk prediction model, and output the scratch risk prediction value corresponding to the target process parameter combination; The scratch risk prediction value corresponding to the target process parameter combination is mapped to the scratch risk prediction value range corresponding to multiple scratch defect quality levels to determine the scratch defect quality level corresponding to the target process parameter combination. If the scratch defect quality level corresponding to the target process parameter combination does not meet the preset target level, adjust the target process parameter combination until the scratch defect quality level corresponding to the target process parameter combination meets the target level. Specifically, S103 mentioned above may include: When the scratch defect quality level corresponding to the target process parameter combination meets the scratch risk prediction value range corresponding to the target degree, the cold rolling process associated unit of the steel coil is controlled to perform cold rolling operation on the steel coil according to the target process parameter combination.

[0057] In some embodiments of this application, when the scratch defect quality level corresponding to the target process parameter combination does not meet the preset target level, the target process parameter combination is adjusted until the scratch defect quality level corresponding to the target process parameter combination meets the target level. For example, when the output scratch position and scratch degree do not reach the target level, that is, when the surface quality level does not reach A, the scratch position and scratch degree at this time and the optimized basic parameters (i.e., the target process parameter combination) are recorded, and the optimized basic parameters are iteratively predicted and adjusted by the particle swarm optimization algorithm.

[0058] In some embodiments of this application, when the scratch defect quality level corresponding to the target process parameter combination meets the scratch risk prediction value range corresponding to the target degree, the cold rolling process associated unit of the steel coil is controlled to perform cold rolling operation on the steel coil according to the target process parameter combination. For example, when the output scratch position and scratch degree reach the target, that is, when the surface quality level reaches A, the scratch position, scratch degree and optimized basic parameters are recorded at this time, and the optimized basic parameters are implemented in the actual production process.

[0059] In this embodiment, before formal production begins, a scratch risk prediction model is used to pre-verify the target process parameter combination. By matching the predicted risk value with the corresponding quality level, it is determined in advance whether the scratch defects meet the standards. If the preset requirements are not met, the parameters are readjusted until they meet the quality standards before production is organized. This method can proactively avoid quality risks caused by unreasonable process parameters, prevent the generation of batches of defective products, further ensure the stability of the steel coil surface quality, and improve the production qualification rate.

[0060] As another implementation of this application, in order to promptly offset the adverse effects of parameter fluctuations, continuously maintain the stress field balance of each process, and stabilize the scratch prevention effect, after S103 above, the above method may further include: Obtain real-time monitoring data of units associated with the cold rolling process; If the deviation between the real-time monitoring data and the target process parameter combination is greater than a preset deviation threshold, the sensitive process parameters in the target process parameter combination are compensated and adjusted. The sensitive process parameters are the process parameters in the target process parameter combination whose sensitivity coefficient to the objective function is greater than a preset sensitivity threshold. The steel grade, target process parameter combination, deviation threshold, sensitive process parameters, and compensation settings for each sensitive process parameter of the steel coil are stored in the preset production strategy library.

[0061] The aforementioned sensitive process parameters can be those process parameters in the target process parameter combination whose sensitivity coefficient to the objective function is greater than a preset sensitivity threshold. The sensitivity coefficient can be calculated using local or global sensitivity analysis methods for each decision variable.x j For the final objective function F Sensitivity coefficient S j The sensitivity threshold, for example, can be 0.1. Of course, in the embodiments of this application, the sensitivity threshold is not limited to this example, and can also be set according to the actual needs of the user, which is not specifically limited here.

[0062] In some embodiments of this application, when the deviation between the real-time monitoring data and the target process parameter combination is greater than a preset deviation threshold, the sensitive process parameters in the target process parameter combination are compensated and adjusted. For example, when there is a slight deviation between the real-time monitoring data (such as inlet plate shape, furnace temperature uniformity) and the basic parameters in the production strategy library, the highly sensitive variables (i.e. sensitive process parameters) can be compensated and adjusted in real time and with a small magnitude based on a pre-trained lightweight meta-model.

[0063] In some embodiments of this application, the steel grade, target process parameter combination, deviation threshold, sensitive process parameters, and compensation setting values ​​for each sensitive process parameter of the steel coil are stored in a preset production strategy library. In this way, based on the target process parameter combination of the optimization results and the sensitive process parameters of the sensitivity analysis, a graded and layered production strategy library of optimal process parameters is established for different steel grades (such as low carbon steel and high strength steel) and different product specifications (thickness and width). The compensation setting values ​​of the sensitive process parameters corresponding to each category of products and their allowable fluctuation range (i.e., deviation threshold) can be clearly listed.

[0064] In this embodiment, the on-site operating data and target process parameters are compared in real time during production. When the deviation exceeds a threshold, targeted compensation and correction are made for highly sensitive key process parameters to promptly offset the adverse effects of parameter fluctuations, continuously maintain the stress field balance of each process, and stabilize the scratch prevention effect. At the same time, information such as steel grade, process parameters, and control rules are stored in the production strategy library to realize the accumulation and reuse of experience data, providing a reference for subsequent production of the same grade of steel and improving the continuity and versatility of overall process control.

[0065] In some embodiments, the above objective function can be as follows:

[0066] in, F(X) Let be the objective function. X For process parameter combinations, f scratch ( X () represents the predicted abrasion risk value corresponding to the combination of process parameters. T cycle ( XThe total annealing cycle time is calculated based on the combination of process parameters. E ( X To estimate energy consumption, T base As the baseline annealing cycle, E base As a baseline energy consumption, a、b、c These are the weighting coefficients, and a+b+c=1.

[0067] The above f scratch ( X The value of ) represents the predicted abrasion risk corresponding to the combination of process parameters. Its value range is [0,1]. The larger the value, the higher the probability or severity of abrasion. This predicted value is obtained by regression analysis of the preprocessed feature data using the abrasion risk prediction model.

[0068] The above E ( X This is the estimated energy consumption calculated based on the temperature curve and time.

[0069] The above T base As the baseline annealing cycle, E base Both are used as the baseline energy consumption for normalization.

[0070] The above a、b、c These are the weighting coefficients, and a+b+c=1 It should be noted that the weighting coefficients can be adjusted according to the production strategy priorities (quality first, efficiency first, or energy saving first).

[0071] In this embodiment, the constructed objective function comprehensively considers three core indicators: scratch risk, annealing cycle, and production energy consumption, and achieves multi-objective balanced control through weighting coefficients. While prioritizing the reduction of scratch defects in steel coils, it also takes into account production efficiency and energy consumption control, avoiding the problems of prolonged production cycles and high energy consumption caused by solely pursuing surface quality, and achieving synergistic optimization of product quality, production efficiency, and production costs.

[0072] In some embodiments, the above constraints may specifically include: process constraint sub-conditions and production constraint sub-conditions. Process constraint sub-conditions are used to constrain the operating range of each process parameter in the process parameter combination. The process parameter combination includes cold continuous rolling coiling tension, rotation speed, heating rate, isothermal temperature, isothermal time, cooling time, internal temperature of the steel coil, hoisting time, maximum hoisting speed, leveling uncoiling tension, and leveling coiling tension. Production constraint subconditions are used to constrain the operating range of production parameters, which include the total duration of the annealing cycle.

[0073] The above combination of process parameters may include cold continuous rolling coiling tension, rotation speed, heating rate, constant temperature, constant temperature time, cooling time, internal temperature of the steel coil, hoisting time, maximum hoisting speed, leveling uncoiling tension, and leveling coiling tension.

[0074] In some embodiments of this application, process constraint sub-conditions are used to constrain the operating range of each process parameter in the process parameter combination. For example, it can constrain each decision variable (i.e., process parameter) to be within its allowed operating range. Therefore, process parameter boundary constraints are set as follows: Meanwhile, to ensure the strip is tight and flat, the uncoiling tension is... un and winding tension re Must meet: .

[0075] In some embodiments of this application, due to product performance constraints, the total annealing cycle time is... T cycle and the shortest time required to meet the product's mechanical performance requirements. T rmin The following conditions must be met: .

[0076] In this embodiment, the constraints are divided into two categories: process constraints and production constraints. On the one hand, reasonable operating ranges are defined for all process parameters throughout the entire process of cold rolling, bell-type annealing, hoisting, and leveling to ensure the safe and stable operation of each unit and the compliance of process execution. On the other hand, production constraints are set for the annealing cycle to meet the on-site capacity layout requirements. The dual constraints ensure that the parameter optimization results fully match the actual on-site working conditions, avoiding problems such as parameters exceeding limits or being unable to be implemented, and ensuring that the optimized process parameters can be directly applied to actual production.

[0077] To facilitate understanding of the cold rolling process control method for steel coils in the embodiments of this application, the actual application process of this cold rolling process control method for steel coils is described as follows: like Figure 2 As shown, this application provides another method for controlling the cold rolling process of steel coils, which may include the following steps S201 to S204: S201: Collect historical data related to scratches during cold rolling annealing in the production process of strip and sheet, preprocess the collected historical data, including data cleaning, outlier removal, feature extraction and normalization, and construct a historical scratch dataset for modeling and analysis. S202: Based on the actual production needs of the scratch problem, with minimizing the scratch risk as the core and taking into account the optimization objectives of efficiency and energy consumption, define the optimization objective function, clarify the optimization decision variables and their value range, and set relevant process constraints and production condition constraints. S203: Based on the historical scratch dataset, the particle swarm optimization algorithm is used to optimize the key parameters of the steel coil production process. By setting algorithm parameters such as population size, number of iterations, and learning factor, the algorithm gradually searches for the optimal solution of the objective function in multiple iterations until the algorithm converges to the preset target threshold, thereby obtaining the optimal combination of process parameters. S204: Further analyze and apply the optimized results. By analyzing the optimization results, assess the risk and severity of scratches under different production decisions, and formulate corresponding production adjustment strategies and preventive measures based on the analysis results, thereby effectively reducing the frequency and severity of scratches during annealing in actual production.

[0078] This application employs particle swarm optimization (PSO) to model and optimize the parameters of the cold-rolled annealing process, effectively uncovering the complex nonlinear relationship between production process parameters and scratch occurrence, thereby significantly reducing the scratch rate on the steel coil surface during annealing. Compared to traditional methods relying on manual experience or single-factor adjustments, this method uses data-driven global optimization to accurately identify critical process windows, improving the scientific rigor and reliability of process settings. By establishing a historical scratch dataset and combining iterative optimization, the system can dynamically adjust multi-dimensional decision variables such as temperature, tension, and speed in the annealing process, minimizing scratch occurrence while meeting production efficiency constraints. The optimization results can be directly used to guide actual production, helping to develop differentiated process strategies and preventative measures, reducing product downgrading and rework caused by scratches, thereby improving strip surface quality and product yield. Furthermore, this method helps extend the stable operating cycle of the annealing equipment and related roller systems, reducing unplanned downtime and maintenance frequency, and improving the overall operating rate and economic efficiency of the production line. By enabling proactive control of scratch problems, this method not only helps save energy and material consumption and reduce production costs, but also promotes the evolution of cold rolling production management towards intelligence and refinement, enhancing the competitiveness and sustainable development capabilities of enterprises in the field of high-quality strip production.

[0079] In S201, historical scratch data may include: historical production data of cold rolling mill and historical production data of bell-type annealing mill.

[0080] Historical production data for cold rolling mills may include: historical data of the cold continuous rolling Carousel coiler and historical data of actual exit tension; further, historical data of the cold continuous rolling Carousel coiler includes: coiling tension and rotation speed.

[0081] Historical production data for the bell-type annealing unit may include: bell-type unit temperature data, hoisting data, and leveling tension data; further, the bell-type unit temperature data may include: bell-type annealing unit temperature data, bell-type annealing unit time data, and internal temperature data of the steel coil; wherein, the bell-type annealing unit temperature data may include: temperature rise rate data during the heating stage, temperature data during the isothermal stage, and temperature fall rate data during the cooling stage; the bell-type annealing unit time data may include: heating time data, isothermal time data, and cooling time data; further, the hoisting data may include: hoisting speed and time data; further, the leveling tension data may include: uncoiling tension and coiling tension data.

[0082] Furthermore, preprocessing of the data in the historical dataset can include: Missing and outlier values ​​in the historical dataset are identified, and samples containing missing or outlier values ​​are removed. The data is standardized to ensure that all variables are within the same dimension, facilitating algorithm optimization. The preprocessed historical dataset is then deduplicated to obtain the deduplicated historical dataset.

[0083] Furthermore, deduplication of data in the preprocessed historical dataset can include: By comparing historical data of the same steel grade and analyzing the parameter differences, if the difference between steel coils is less than the preset difference threshold, the data corresponding to the steel coil is considered to be duplicated. In this case, only the historical data of one steel coil is retained, and the historical data of the other steel coils are deleted to complete the data deduplication.

[0084] In S202, the optimization objective is defined, and the decision variables and constraints for optimization are determined. The optimization objective is to minimize the inner-loop function relative to the objective function. F ( X ), find F ( X The minimum combination of decision variables, expressed by the objective function:

[0085] in, X This is the decision variable vector, i.e., the combination of the above process parameters; f scratch ( X ) is a predicted abrasion risk value trained on a historical abrasion dataset. Its value range is [0,1]. The larger the value, the higher the probability or severity of the abrasion. This predicted value is obtained by regressing the preprocessed feature data through a machine learning model. T cycle ( X() represents the total theoretical annealing cycle time calculated based on the decision variables; E ( X This is an estimated energy consumption calculated based on the temperature curve and time. T base and E base These are the baseline annealing cycle and baseline energy consumption, used for normalization. α , β , c These are the weighting coefficients, and a+b+c=1 It can be adjusted according to the production strategy priorities (quality first, efficiency first, or energy saving first).

[0086] Furthermore, the degree of scratches is evaluated by visual inspection and simple measuring tools to determine the depth and area of ​​the scratches, as well as the proportion of scratches on the width of the steel coil. The scratches are divided into four levels: A, B, C, and D, as shown in Table 1 below, and the classification data is entered into the database.

[0087] Table 1 Evaluation Table of Surface Quality Grade for Scratches

[0088] Furthermore, decision variables are determined. These decision variables are selected from adjustable key process parameters in the preprocessed and feature-engineered historical dataset. The minimum combination of decision variables, i.e., the aforementioned target combination of process parameters, can mainly include: Cold rolling coiling tension set on the cold rolling Carousel coiler c Rotation speed oh ; Annealing temperature-time curve parameters: including heating rate R h Constant temperature T h Constant temperature time t h Cooling time t c internal temperature of steel coil t in ; Lifting parameters: Lifting time t l Maximum hoisting speed V l ; Flattening tension parameters: Flattening unwinding tension un With flat winding tension re .

[0089] Furthermore, to ensure the feasibility and security of the optimization scheme, the following constraints are set: Each decision variable must be within its permissible operating range. Therefore, boundary constraints for the process parameters are set as follows:

[0090] Total annealing time due to product performance constraints T cycle and the shortest time required to meet the product's mechanical performance requirements. T rmin The following conditions must be met: .

[0091] To ensure the strip is tight and flat, the uncoiling tension is... un and winding tension re Must meet: .

[0092] In S203, the particle swarm optimization algorithm is trained using data from the basic parameter database, enabling it to optimize the steel coil production process. Through multiple iterations, the algorithm gradually converges to the target threshold.

[0093] Further, the algorithm is initialized, and the particle swarm size is set to... N The position vector of each particle Represents a set of decision variables X The value of, dimension d The number of decision variables is equal to the number of particles; the position of each particle is randomly initialized. and speed Position initialization must meet the above constraints; set algorithm parameters: initial inertia weights. oh start and termination inertia weight oh end Individual learning factor c 1 Social learning factor c 2 Maximum number of iterations K max Convergence threshold e .

[0094] Furthermore, in the iterative optimization process, in each iteration... t The process involves fitness evaluation, updating individual and global optima, updating particle velocity and position, and handling boundary and constraint issues. Fitness evaluation is performed for each particle... i to position Decode the specific combination of process parameters and substitute them into the objective function. F ( X Calculate its fitness value F i Update individual optimality and global optimality: Compare the current fitness of each particle with its historical individual optimal fitness. pbest i ,renew pbest i and individual optimal position Compare the individual optimal fitness of all particles and update the global optimal fitness. gbest i and global optimal position Update particle velocity and position: , ,in, r 1 , r 2 The random number is within the interval [0,1], with inertial weight. oh ( t A linear decreasing strategy is adopted: Boundary and constraint handling: The updated particle positions undergo boundary checks. If a position exceeds the decision variable boundary, it is pulled back to the boundary. For positions that violate hard constraints such as product performance or equipment safety, a very large penalty fitness value is assigned, causing them to be eliminated in subsequent iterations.

[0095] Furthermore, to prevent continuous iteration, when the maximum number of iterations is reached... K max or global optimal fitness gbest i In continuous The change within a generation is less than a preset threshold. When the algorithm converges, the iteration terminates. Output the global optimal position at this point. e This serves as the optimal process parameter setting scheme.

[0096] Furthermore, when the optimal process parameter setting scheme outputs scratch location and scratch severity to the target level, i.e., when the surface quality grade reaches A, the scratch location, scratch severity, and optimized basic parameters are recorded, and the optimized basic parameter setting scheme is implemented in actual production. When the output scratch location and scratch severity do not reach the target level, i.e., the surface quality grade does not reach A, the scratch location, scratch severity, and optimized basic parameters are recorded, and the difference model of the optimized basic parameters is re-inputted for iterative prediction and adjustment.

[0097] In S204, in-depth analysis is conducted based on the optimization results. Key factors are identified through comparative verification and sensitivity analysis. Based on this, a hierarchical production strategy library is constructed, and a dynamic application mechanism that can be fine-tuned online and updated in a closed loop is established to formulate and implement effective production strategies and measures, thereby significantly reducing the occurrence of scratches in annealing.

[0098] Furthermore, the optimization results are further analyzed and applied, including the verification and diagnostic analysis of the optimization results: comparative verification compares the optimized scheme with typical schemes in historical production, and calculates the theoretical risk reduction through the scratch risk prediction model.

[0099] Furthermore, key factor identification: using local or global sensitivity analysis methods, calculate the values ​​of each decision variable. x j For the final objective function F Sensitivity coefficient S j Identify highly sensitive variables S j >0.1, these are the key process points for controlling the risk of abrasion, and need to be closely monitored and stably controlled during production.

[0100] Furthermore, the construction of a tiered production strategy library involves establishing a tiered and layered optimal process parameter recommendation table (strategy library) for different steel grades (such as low-carbon steel and high-strength steel) and different product specifications (thickness and width) based on optimization results and sensitivity analysis conclusions. This table clearly lists the key parameter compensation settings and their allowable fluctuation ranges for each product category.

[0101] Furthermore, the system dynamically adjusts and optimizes the data through closed-loop processing, and develops a process setting auxiliary system to complete online fine-tuning. During actual production, the system automatically retrieves basic parameter settings (the combination of decision variables that minimizes F(X)) from the production strategy library based on the planned steel coil information. When there are slight deviations between real-time monitoring data (such as inlet plate shape and furnace temperature uniformity) and the basic parameters in the strategy library, the system can make immediate, small-amplitude compensation adjustments to highly sensitive variables based on a pre-trained lightweight meta-model.

[0102] Specifically, the particle swarm optimization algorithm model is trained using data from the basic parameter database (i.e., the production strategy library) so that it can minimize the scratches on the steel coil by adjusting the basic parameter settings; the optimal difference in the basic parameter settings is calculated to minimize the scratches, i.e., the scratches rating is A.

[0103] In some embodiments of this application, a periodic model update mechanism is established to complete the closed-loop update. Every time a certain amount of new production data (e.g., 500 volumes) is accumulated, the new data is added to the historical scratch dataset, the scratch risk prediction model is retrained, and the optimization algorithm is run again to update the production strategy library. This forms a continuous improvement closed loop of "data accumulation - model and strategy update - guiding production - generating new data," enabling scratch control capabilities to adaptively improve over time.

[0104] After implementing the above optimization strategies, an evaluation of economic benefits and quality is conducted. It is expected that the scratch defect rate in the bell-type annealing process can be significantly reduced, thereby improving the product surface quality pass rate and reducing losses due to scrap, downgrading, or rework caused by surface defects. At the same time, by coordinating and optimizing the annealing cycle and energy consumption, production efficiency and energy utilization can be improved while ensuring quality, achieving cost reduction and efficiency improvement.

[0105] Based on the steel coil cold rolling process control method provided in the above embodiments, this application also provides specific implementation methods of the steel coil cold rolling process control device. Please refer to the following embodiments.

[0106] like Figure 3 As shown, the cold rolling process control device 300 for steel coils provided in this application embodiment may include the following modules: a first acquisition module 301, a calculation module 302, and a control module 303.

[0107] The first acquisition module 301 is used to acquire the historical scratch dataset of the corresponding steel grade of the steel coil. The historical scratch dataset includes multiple standard historical process parameter combinations and the scratch risk prediction value corresponding to each standard historical process parameter combination. The calculation module 302 is used to calculate the target process parameter combination of the steel coil based on multiple standard historical process parameter combinations and the predicted scratch risk value corresponding to each standard historical process parameter combination, with the process parameter combination and production parameters meeting the preset constraints as constraints, and with minimizing the preset objective function as the objective, the target process parameter combination includes target cold continuous rolling coiling tension, target rotation speed, target heating rate, target isothermal temperature, target isothermal time, target cooling time, target internal temperature of the steel coil, target hoisting time, target maximum hoisting speed, target leveling uncoiling tension, and target leveling coiling tension; The control module 303 is used to control the cold rolling process associated unit of the steel coil to perform cold rolling operation on the steel coil according to the target process parameter combination, and the degree of scratches on the steel coil meets the preset target degree.

[0108] The cold rolling process control device for steel coils in this embodiment can acquire a historical scratch dataset for the corresponding steel grade of the steel coil. The historical scratch dataset includes multiple standard historical process parameter combinations and scratch risk prediction values ​​corresponding to each standard historical process parameter combination. Based on the multiple standard historical process parameter combinations and scratch risk prediction values ​​corresponding to each standard historical process parameter combination, and with the process parameter combinations and production parameters meeting preset constraints as constraints, and minimizing a preset objective function as the objective, the target process parameter combination for the steel coil is calculated. The target process parameter combination includes target cold continuous rolling coiling tension, target rotation speed, target heating rate, target isothermal temperature, target isothermal time, target cooling time, target internal temperature of the steel coil, target hoisting time, target maximum hoisting speed, target leveling uncoiling tension, and target leveling coiling tension. The device controls the associated cold rolling unit of the steel coil to perform cold rolling operations on the steel coil according to the target process parameter combination, and the degree of scratches on the steel coil meets the preset target degree. Thus, in this embodiment of the application, a dataset is constructed based on historical process parameters and corresponding predicted values ​​of scratch risk. The objective function related to scratch risk is minimized as the optimization objective. Combined with production constraints, the optimal combination of target process parameters for the entire cold rolling process is obtained and systematically optimized. This effectively improves the problems of uneven distribution and incoordination of upstream cold rolling interlayer stress, residual thermal stress from annealing, and external force field during hoisting. It fundamentally suppresses the generation of quality defects such as loose coils, core pulling, and misalignment of coils, and ultimately improves the quality of steel coil products.

[0109] In some embodiments, the first acquisition module 301 described above may specifically include: The acquisition unit is used to acquire multiple historical process parameter combinations for the corresponding steel grade of the steel coil and the scratch defect quality level corresponding to each historical process parameter combination. The preprocessing unit is used to preprocess multiple combinations of historical process parameters to obtain multiple standard combinations of historical process parameters. The preprocessing includes at least one of rejection processing, normalization processing and deduplication processing. The first determining unit is used to determine the scratch risk prediction value corresponding to each standard historical process parameter combination based on multiple standard historical process parameter combinations and the scratch defect quality level corresponding to each standard historical process parameter combination. The scratch risk prediction value is used to indicate the degree of scratches on the steel coil under the standard historical process parameter combination. The construction unit is used to construct a historical abrasion dataset based on multiple standard historical process parameter combinations and the abrasion risk prediction values ​​corresponding to each standard historical process parameter combination.

[0110] In some embodiments, the first determining unit may specifically include: The training subunit is used to train a preset machine learning model based on multiple standard historical process parameter combinations and the scratch defect quality level corresponding to each standard historical process parameter combination, so as to obtain a scratch risk prediction model. Different scratch defect quality levels correspond to different scratch risk prediction value ranges. The prediction sub-unit is used to input the various standard historical process parameter combinations into the scratch risk prediction model and output the scratch risk prediction value corresponding to each standard historical process parameter combination.

[0111] In some embodiments, the computing module 302 described above may specifically include: The initialization unit is used to initialize and randomly generate N particles. Different particles correspond to different combinations of process parameters. N is a positive integer greater than or equal to 1. The combination of process parameters and production parameters corresponding to the N particles meet the preset constraints. The calculation unit is used to input the process parameters of each particle into the preset objective function and calculate the fitness of each particle. The second determining unit is used to compare the fitness of N particles and determine the global optimal fitness, which is the minimum fitness. The update unit is used to iteratively update the process parameter combination of N particles with the constraints of process parameter combination and production parameter satisfaction, and to iteratively calculate the fitness of each particle and update the global optimal fitness with the goal of minimizing the objective function. The third determining unit is used to determine the combination of process parameters corresponding to the global optimal fitness as the target combination of process parameters for steel coils when the number of iterations reaches a preset number and / or the change in the global optimal fitness within multiple consecutive generations is less than a preset threshold.

[0112] As one implementation of this application, in order to further ensure the stability of the steel coil surface quality and improve the production qualification rate, the above-mentioned device 300 may further include: The prediction module is used to input the target process parameter combination into the preset scratch risk prediction model and output the scratch risk prediction value corresponding to the target process parameter combination. The determination module is used to map the scratch risk prediction value corresponding to the target process parameter combination to the scratch risk prediction value range corresponding to multiple scratch defect quality levels, and determine the scratch defect quality level corresponding to the target process parameter combination. The first adjustment module is used to adjust the target process parameter combination until the scratch defect quality level corresponding to the target process parameter combination meets the target level when the scratch defect quality level corresponding to the target process parameter combination does not meet the preset target level. Specifically, the control module 303 can be used to control the cold rolling process associated unit of the steel coil to perform cold rolling operation on the steel coil according to the target process parameter combination, provided that the scratch defect quality level corresponding to the target degree meets the scratch risk prediction value range corresponding to the target degree.

[0113] As another implementation of this application, in order to promptly offset the adverse effects of parameter fluctuations, continuously maintain the stress field balance of each process, and stabilize the scratch prevention effect, the above-mentioned device 300 may further include: The second acquisition module is used to acquire real-time monitoring data of the cold rolling process-related units; The second adjustment module is used to compensate and adjust the sensitive process parameters in the target process parameter combination when the deviation between the real-time monitoring data and the target process parameter combination is greater than a preset deviation threshold. The sensitive process parameters are the process parameters in the target process parameter combination whose sensitivity coefficient to the objective function is greater than a preset sensitivity threshold. The storage module is used to store the steel grade, target process parameter combination, deviation threshold, sensitive process parameters, and compensation settings of each sensitive process parameter of the steel coil into a preset production strategy library.

[0114] In some embodiments, the above objective function is as follows:

[0115] in, F(X) Let be the objective function. X For process parameter combinations, f scratch ( X () represents the predicted abrasion risk value corresponding to the combination of process parameters. T cycle ( X The total annealing cycle time is calculated based on the combination of process parameters. E ( X To estimate energy consumption, T base As the baseline annealing cycle, E base As a baseline energy consumption, a、b、c These are the weighting coefficients, and a+b+c=1.

[0116] In some embodiments, the above constraints include: process constraint sub-conditions and production constraint sub-conditions. Process constraint sub-conditions are used to constrain the operating range of each process parameter in the process parameter combination. The process parameter combination includes cold continuous rolling coiling tension, rotation speed, heating rate, isothermal temperature, isothermal time, cooling time, internal temperature of the steel coil, hoisting time, maximum hoisting speed, leveling uncoiling tension, and leveling coiling tension. Production constraint subconditions are used to constrain the operating range of production parameters, which include the total duration of the annealing cycle.

[0117] Figure 4 A schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application is shown.

[0118] An electronic device may include a processor 401 and a memory 402 storing computer program instructions.

[0119] Specifically, the processor 401 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0120] Memory 402 may include mass storage for data or instructions. For example, and not limitingly, memory 402 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 402 may include removable or non-removable (or fixed) media. Where appropriate, memory 402 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 402 is non-volatile solid-state memory.

[0121] In a particular embodiment, memory 402 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Thus, generally, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of this disclosure.

[0122] The processor 401 reads and executes computer program instructions stored in the memory 402 to implement any of the steel coil cold rolling process control methods in the above embodiments.

[0123] In one example, the electronic device may also include a communication interface 403 and a bus 410. For example, Figure 4 As shown, the processor 401, memory 402, and communication interface 403 are connected through bus 410 and complete communication with each other.

[0124] The communication interface 403 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0125] Bus 410 includes hardware, software, or both, that couples components of an electronic device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 410 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0126] This electronic device can execute the cold rolling process control method for steel coils in the embodiments of this application, thereby achieving a combination of Figure 1 and Figure 3 The method and apparatus for controlling the cold rolling process of steel coils are described.

[0127] Furthermore, in conjunction with the steel coil cold rolling process control method in the above embodiments, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the steel coil cold rolling process control methods in the above embodiments.

[0128] This application also provides a computer program product, including a computer program that, when executed by a processor, implements any of the steel coil cold rolling process control methods described in the above embodiments.

[0129] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0130] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0131] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0132] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer 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 these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0133] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for controlling the cold rolling process of steel coils, characterized in that, include: Obtain the historical scratch dataset for the corresponding grade of steel coil. The historical scratch dataset includes multiple standard historical process parameter combinations and scratch risk prediction values ​​corresponding to each standard historical process parameter combination. Based on the multiple standard historical process parameter combinations and the predicted scratch risk values ​​corresponding to each standard historical process parameter combination, with the process parameter combination and production parameters meeting preset constraints as constraints, and with minimizing the preset objective function as the objective, the target process parameter combination of the steel coil is calculated. The target process parameter combination includes target cold continuous rolling coiling tension, target rotation speed, target heating rate, target isothermal temperature, target isothermal time, target cooling time, target internal temperature of the steel coil, target hoisting time, target maximum hoisting speed, target leveling uncoiling tension, and target leveling coiling tension. The cold rolling process unit of the steel coil is controlled to perform cold rolling operation on the steel coil according to the target process parameter combination, and the degree of scratches on the steel coil meets the preset target degree.

2. The method according to claim 1, characterized in that, The process of obtaining the historical scratch dataset for the corresponding steel grade of the steel coil includes: Obtain multiple historical process parameter combinations for the corresponding steel grade of the steel coil and the scratch defect quality level corresponding to each historical process parameter combination; The multiple combinations of historical process parameters are preprocessed to obtain multiple combinations of standard historical process parameters. The preprocessing includes at least one of rejection processing, normalization processing and deduplication processing. Based on the multiple standard historical process parameter combinations and the scratch defect quality level corresponding to each standard historical process parameter combination, a scratch risk prediction value corresponding to each standard historical process parameter combination is determined. The scratch risk prediction value is used to indicate the degree of scratches on the steel coil under the standard historical process parameter combination. The historical abrasion dataset is constructed based on the multiple standard historical process parameter combinations and the abrasion risk prediction values ​​corresponding to each standard historical process parameter combination.

3. The method according to claim 2, characterized in that, The step of determining the predicted scratch risk value for each of the multiple standard historical process parameter combinations and the scratch defect quality level corresponding to each standard historical process parameter combination includes: Based on the multiple standard historical process parameter combinations and the scratch defect quality level corresponding to each standard historical process parameter combination, a preset machine learning model is trained to obtain a scratch risk prediction model. Different scratch defect quality levels correspond to different scratch risk prediction value ranges. The various combinations of standard historical process parameters are input into the scratch risk prediction model, and the scratch risk prediction value corresponding to each combination of standard historical process parameters is output.

4. The method according to claim 1, characterized in that, The step of calculating the target process parameter combination for the steel coil based on the multiple standard historical process parameter combinations and the predicted scratch risk values ​​corresponding to each standard historical process parameter combination, with the process parameter combination and production parameters meeting preset constraints as constraints, and minimizing a preset objective function as the objective, includes: N particles are randomly generated during initialization. Different particles correspond to different combinations of process parameters. N is a positive integer greater than or equal to 1. The combination of process parameters and production parameters corresponding to the N particles satisfy preset constraints. The process parameters of each particle are combined and input into a preset objective function to calculate the fitness of each particle. The fitness of the N particles is compared to determine the global optimal fitness, which is the minimum fitness. With the constraints that the process parameter combination and production parameters satisfy, the process parameter combination of the N particles is iteratively updated, and with the objective of minimizing the objective function, the fitness of each particle is iteratively calculated and the global optimal fitness is updated. When the number of iterations reaches a preset number, and / or the change in the global optimal fitness within multiple consecutive generations is less than a preset threshold, the combination of process parameters corresponding to the global optimal fitness is determined as the target combination of process parameters for the steel coil.

5. The method according to claim 1, characterized in that, Before the cold rolling unit associated with the steel coil performs cold rolling operation on the steel coil according to the target process parameter combination, the method further includes: The target process parameter combination is input into a preset scratch risk prediction model, and the scratch risk prediction value corresponding to the target process parameter combination is output. The scratch risk prediction value corresponding to the target process parameter combination is mapped to the scratch risk prediction value range corresponding to multiple scratch defect quality levels to determine the scratch defect quality level corresponding to the target process parameter combination. If the scratch defect quality level corresponding to the target process parameter combination does not meet the preset target level, the target process parameter combination is adjusted until the scratch defect quality level corresponding to the target process parameter combination meets the target level. The cold rolling unit controlling the steel coil performs cold rolling operations on the steel coil according to the target process parameter combination, including: When the scratch defect quality level corresponding to the target process parameter combination meets the scratch risk prediction value range corresponding to the target degree, the cold rolling process associated unit of the steel coil is controlled to perform cold rolling operation on the steel coil according to the target process parameter combination.

6. The method according to claim 1, characterized in that, After the cold rolling unit controlling the steel coil performs cold rolling operations on the steel coil according to the target process parameter combination, the method further includes: Obtain real-time monitoring data of the units associated with the cold rolling process; If the deviation between the real-time monitoring data and the target process parameter combination is greater than a preset deviation threshold, the sensitive process parameters in the target process parameter combination are compensated and adjusted. The sensitive process parameters are the process parameters in the target process parameter combination whose sensitivity coefficient to the objective function is greater than a preset sensitivity threshold. The steel grade, target process parameter combination, deviation threshold, sensitive process parameters, and compensation settings for each sensitive process parameter of the steel coil are stored in a preset production strategy library.

7. The method according to claim 1, characterized in that, The objective function is as follows: ， in, F(X) Let the objective function be... X For process parameter combinations, f scratch ( X ) represents the predicted abrasion risk value corresponding to the combination of process parameters. T cycle ( X The total annealing cycle time is calculated based on the combination of process parameters. E ( X To estimate energy consumption, T base As the baseline annealing cycle, E base As a baseline energy consumption, α, β, γ These are the weighting coefficients, and α+β+γ=1.

8. The method according to claim 1, characterized in that, The constraints include: process constraint sub-conditions and production constraint sub-conditions. The process constraint sub-conditions are used to constrain the operating range of each process parameter in the process parameter combination. The process parameter combination includes cold continuous rolling coiling tension, rotation speed, heating rate, isothermal temperature, isothermal time, cooling time, internal temperature of the steel coil, hoisting time, maximum hoisting speed, leveling uncoiling tension, and leveling coiling tension. The production constraint sub-condition is used to constrain the operating range of the production parameters, which include the total duration of the annealing cycle.

9. A steel coil cold rolling process control device, characterized in that, The device includes: The first acquisition module is used to acquire the historical scratch dataset of the corresponding steel grade of the steel coil. The historical scratch dataset includes multiple standard historical process parameter combinations and scratch risk prediction values ​​corresponding to each standard historical process parameter combination. The calculation module is used to calculate the target process parameter combination of the steel coil based on the multiple standard historical process parameter combinations and the abrasion risk prediction value corresponding to each standard historical process parameter combination, with the process parameter combination and production parameters meeting the preset constraints as constraints, and with the goal of minimizing the preset objective function. The target process parameter combination includes target cold continuous rolling coiling tension, target rotation speed, target heating rate, target isothermal temperature, target isothermal time, target cooling time, target internal temperature of the steel coil, target hoisting time, target maximum hoisting speed, target leveling uncoiling tension, and target leveling coiling tension. The control module is used to control the cold rolling process associated unit of the steel coil to perform cold rolling operation on the steel coil according to the target process parameter combination, and the degree of scratches on the steel coil meets the preset target degree.

10. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the steel coil cold rolling process control method as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the cold rolling process control method for steel coils as described in any one of claims 1-8.

12. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device causes the electronic device to perform the steel coil cold rolling process control method as described in any one of claims 1-8.

Images